HORIZONS - DSTA

DSTAHORIZONS

2015

71 Science Park DriveSingapore 118253

www.dsta.gov.sg

DS

TA

HO

RIZ

ON

S 2015

DSTA HorizonsIssue 10ISSN 2339-529X (print) ISSN 2339-5303 (online)©2015 Defence Science and Technology Agency

No part of this publication may be reproduced, stored or transmitted in any form or by any means without the prior written permission of the Defence Science and Technology Agency, Singapore.

The opinions and view expressed in this publication are those of the authors and do not necessarily reflect the views of the Defence Science and Technology Agency, Singapore.

All information correct at time of publication.

DSTA HORIZONS EDITORIAL TEAM

EditorTan Yang How

MembersCher Sok Kheng PatriciaChiam DasenChiam Pack LuanChua Siew Ting PearlyFan Yue SangHeng Eu Chang LeonardHeng Tze Hua AndyHo Kwee Peng JuliLee Kian KongLee Kwee EngLeow Aik Siang

Readers can access current and past issues of DSTA Horizons atwww.dsta.gov.sg/dstahorizons

We welcome your feedback. Please send all correspondence to:

DSTA Horizons Editorial TeamDSTA Academy 1 Depot RoadSingapore 109679

Email: [email protected]

Loo Jang WeiNg Kok KengTan Beng HockTan Chee Hwai DennisTan Kok PengTeo Chong LaiTeo Seow KhyeTeo Siow HiangWang Yew KwangWong Lock LiangZee Sow Wai

Technical EditorProfessor Bernard TanDepartment of Physics, Faculty of ScienceNational University of Singapore

DSTA HORIZONS | 2015

CONTENTS2 Editorial

4 TransformingRangePracticeswiththeMulti-MissionRangeComplex LIMPeter,YEOQiuLingTammi,LIMMengKeeJohnson,LAUChinSengEric

14 InnovativeApproachesfortheAdvancedCombatManSystem LIMWeiQiang,PEHHanYongLester

22 TechnologicalAdvancementsandInnovationsinCombat EngineeringEquipment PHUAZhengqiDaryl,TANChun,WONGYeeYinKimberly

30 DeliveringNewMineCountermeasureCapabilitiestotheRSN GOHYongHan,LAMSuYingAudrey

38 eWorkplace:EvolvingDSTA’sKnowledgeManagementJourney KOOYihLiangKevin,LIMLayHarEvon,HOWeiLingAngela,SOHYunLinJason

46 Model-DrivenArchitectureApproachforEnterpriseSystems LAIKokKee,NGWendy,LOWKweeBoon

54 DataAnalyticsforOptimisingCyberandDataCentreOperations CHANGXuquanStanley,SIMSzeLiang,WONGMingQian

60 ChallengesandDesignConsiderationsforRadarOperation inLocalLittoral LOManLing,LOKEMunKwong

70 KaBandSatelliteCommunicationsDesignAnalysisandOptimisation LEONGSeeChuan,SUNRu-Tian,YIPPengHon

80 PerformanceChallengesforHighResolutionImagingSensors forSurveillanceinTropicalEnvironment LEECheowGim,EEKokTiong,HENGYinghuiElizabeth

90 SafetyManagementofNationalDayParadeFireworksDisplay SIMGimYoung,LEEChungKiat,OEISuCheok,ME5ONGWoeiLeng

100 ProtectionandResiliencyforSingapore’sCriticalInfrastructures ONGKweeSiangSteve,CHONGOiYinKaren,SEEThongHwee

2 DSTA HORIZONS | 2015

EDITORIAL

ThisyearmarksSingapore’sfiftiethyearof independence

and coincidentally, DSTA’s fifteenth anniversary and the

tenthissueofDSTAHorizons.Hence,itisonlyfittingthat

the 12 articles chosen for this issue reflect the diverse

competencies, innovations and expertise of DSTA which

havecontributedtowardsbuildingamodernarmedforces

thatishighlycapableandresourceefficient.

‘TransformingRangePracticeswith theMulti-MissionRange Complex’ traces the innovative development of

the Multi-Mission Range Complex (MMRC), an indoor

live-firing training facility that represents the next step

forwardinmarksmanshiptraining.Bearinguniquefeatures

such as the single-rail targetry system, the MMRC is a

prime example of how creative solutions can overcome

resource and space constraints. Innovative ideas are

also crucial as rapid advancements in technology have

changedthebattlefieldlandscapeandhowtheSingapore

Armed Forces (SAF) operates in the field. ‘InnovativeApproaches for the Advanced Combat Man System’chronicles the enhancements of the Advanced Combat

ManSystem(ACMS)intoitsnewerandlightweightvariant–

theACMSiLITE.Thelessonslearntduringitsdevelopment

serve to highlight possible concepts and technologies

to enhance the SAF’s combat capabilities. Focusing on

key areas of combat engineering tasks on the battlefield

today,‘TechnologicalAdvancementsandInnovationsinCombatEngineeringEquipment’ explores theevolution

ofcombatengineeringequipmentandhowithasshaped

the operational capabilities of the SAF. The article also

looksatfuturetechnologicaltrendsthatwillpossiblyshape

thedevelopmentoffuturecombatengineeringequipment.

DSTA utilises its pool of knowledge to deliver new and

exciting capabilities to the SAF. ‘Delivering New MineCountermeasureCapabilitiestotheRSN’offersinsightsinto the modernisation programme of the Navy’s Mine

CountermeasureVesselswhichimprovestheirminehunting

capabilities. It discusses the technical challenges of the

programmeand theprojectmanagement team’ssystems

engineeringbasedapproach that resulted ina significant

improvementinmissioneffectiveness.

Deriving technical lessons and insights from practical

experiencesisalsoavaluablecomponentofDSTA’swork.

Capturing the key features and design considerations

behind DSTA’s next-generation digital workplace is

‘eWorkplace:EvolvingDSTA’sKnowledgeManagementJourney’.ItdetailsthetransformationofDSTA’seWorkplace

Intranet platform that greatly enhances collaboration,

learningandproductivitywithin theorganisation. ‘Model-drivenArchitectureApproach for EnterpriseSystems’illustrates the features of a Model-driven Architecture

approachandhowitcanimprovetheefficiencyofenterprise

application development. It also shares the architecting

TanYangHowPresidentDSTAAcademy

3DSTA HORIZONS | 2015

effortsandbenefitsofadoptingsuchanapproachaspart

of the IT Application Lifecycle Management framework.

‘DataAnalytics forOptimisingCyberandDataCentreOperations’examineshowDSTAisutilisingdataanalytics

toovercometheincreasinglychallengingtaskofmanaging

cyberdefenceanddatacentreoperationsthroughanomaly

detection, discovery of hidden patterns and insights

and the optimisation of resources. It also explores other

challengesthathavetobeaddressedinordertomaximise

thepotentialapplicationofdataanalytics.

New ideas and perspectives are often triggered in the

face of technical challenges. ‘Challenges and DesignConsiderations for Radar Operation in Local Littoral’describes the challenges posed by Singapore’s unique

littoralenvironmenttoradardesign.Itsauthorsalsoshare

some of their best practices in the operationalisation of

radarsanddiscusspotentialdevelopmentsinthedomain.

‘Ka Band Satellite Communications Design AnalysisandOptimisation’examinesthefeasibilityandapplication

ofaKabandnetworkinsatellitecommunicationsbytaking

a systems approach and carrying out detailed trade-off

analysis of key operational parameters. ‘PerformanceChallenges for High Resolution Imaging Sensors forSurveillance in Tropical Environment’ delves into thescience behind environmental factors such as weather

andhazethatcanadverselyaffectasensor’sperformance

and looks at how the right kind of electro-optics can be

exploitedtoenhancesurveillanceperformance.

Drawing from its experiences in the areas of safety and

security, DSTA has been contributing its expertise in

fireworkssafetymanagementfortheNationalDayParade

(NDP). ‘Safety Management of National Day ParadeFireworks Display’ outlines how safety is addressedthrough the fireworks life cycle and also shares the

innovativesolutionsusedinthereal-timemanagementof

fireworks to deliver a safe fireworks display for theNDP.

Finally, ‘Protection and Resiliency for Singapore’sCritical Infrastructures’ leverages DSTA’s know-how indesigningcriticalinfrastructuresfortheMinistryofDefence

and the SAF to illustrate how protection and resiliency

can be balanced to improve the survivability of critical

infrastructuresinSingapore.

Wehopethatthearticleswillbeaninsightfulandenriching

read for our readers. We are also appreciative of the

authorsandreviewersforalltheireffortsandcommitment.

ThistenthissueofDSTAHorizonsrepresentsasignificant

milestoneinoureffortstoenrichthelearningandsharing

culture within the defence technology community. It is

henceourwishthatDSTAHorizonswillcontinuetoplaythis

importantroleformanymoreissuestocome.Thankyou.


TRANSFoRMINgRANgEPRACTICESWITHTHEMuLTI-MISSIoNRANgECoMPLExLIMPeter,YEOQiuLingTammi,LIMMengKeeJohnson,LAUChinSengEric

INTRODUCTION

Supported by efficient services to enable soldiers to fully

focusonshooting,theMulti-MissionRangeComplex(MMRC)

was conceptualised as a one-stop marksmanship training

hub for soldiers to hone and sustain varied marksmanship

competencies.

The three-storey MMRC features seven live-firing indoor

rangesencompassingadvancedsimulation,acousticsensing

and range technologies to provide realistic scenario-based

live-firingtraining.Iteliminatesthetraditionalprocessofrange

administration by outsourcing routine, non-core pre-range

and post-range administration, logistical and maintenance

functions.

TheMMRCisanaccumulationoftherequirementsofmultiple

shootingrangesintoonefacility.Ithasenhancedthewaythe

SingaporeArmytrainsbyprovidingitwiththeflexibilitytotrain

safelyunderdifferentrealisticscenariosandenvironments. It

hasalsoincreasedtheproductivityandefficiencyoftheArmy

by allowing it to conduct 50% more training opportunities

withinthesametimeframe.

OVERVIEW OF LIVE-FIRING SYSTEMS other than conventional targetry systems (e.g. the Portable

ElectronicTargetrySystemandStationaryElectronicTargetry

System)whicharedeployedinotheroutdoorranges,thelive-

firing systems in the MMRC consist of two new modules:

theVideoTargetrySystem (VTS)and theSingle-RailMoving

ElectronicTargetrySystem(METS).

VideoTargetrySystem

The VTS is a leading-edge computer-based marksmanship,

tacticalandjudgementallive-firingtargetrysystemcomprising

threemainsimulationsubsystems:aVTSboxtargetmeasuring

2m by 2.7m, a RangeControl Computer and a Firing Point

Computer.

TheVTSallowssoldierstotrainbeyondthebasicmarksmanship

settingsprovidedbytraditionalbaffledranges.usingcomputer

generated imageryorcustomisedvideos, itcangeneratean

ABSTRACT

TheMulti-MissionRangeComplex (MMRC) is a three-storey live-firing training hub jointly developedbyDSTAand theHeadquarters9thDivision/Infantryof theSingaporeArmy.TheMMRChasbeencited inmanydomainsasanexcellentexampleofhowSingaporeovercamethechallengesofresourceandspaceconstraintswithinnovativesolutions1.

Thisarticleoutlinestheevolutionofmarksmanshiptrainingfromtraditionalandmanuallyoperatedtargetsinanopenfieldtoadvancedandsafelive-firingtrainingsolutionsinanindoorenvironment.ItalsohighlightsthebenefitsoftheMMRC.

Keywords:indoorlive-firing,judgementalshooting,multi-tiershooting,reconfigurableurbanoperationrange,landsaving


assortmentofconditions realistically (seeFigure1).With the

VTS,soldiersareabletoconductlongdistancemarksmanship

trainingofupto1,000mina50mrangesetup(seeFigure2).

Figure1.Live-firingattheVTSscreen

Figure4.ThebackofaVTSboxtarget

Figure2.VTSscreenssetupinthe50mrange

Figure5.Anacousticsensor

The VTS’ shot detection system utilises precision acoustic

technology toprojectbullet trajectoriesbasedon the actual

ballistictableswithinmillimetreaccuracy(seeFigure3).

Figure3.Anexampleofaballistictable

0 1000 2000 3000 4000 5000 6000 7000

500

1000

1500

2000

2500

3000

3500

4000

4500

5000

Range (m)

Height (m

)

70o

60o

50o

40o

30o

20o

10o

The VTS box target resembles a box with a rubber screen

installed at the front and at the back which allows bullets

to pass through (see Figure 4). The construction of the box

contains the shockwaves (generated by the bullet passing

through) within the box to minimise noise. Twelve acoustic

sensorsareinstalledwithintheboxtodetectthesesupersonic

shockwaves(seeFigure5).


Theactuallocationofthebulletimpactisthendeterminedby

monitoringtheshockwavesviatheacousticsensorsoptimally

locatedattheedgesofthescreen(seeFigure6).Thelocation

ofthebulletwouldthenbecalculatedbyextrapolation.

However, subsonic ammunition such as the 9mm round is

designedtooperateatspeedslessthanthespeedofsound

andwill not create supersonic shockwave as it travels. The

detection is instead computed when the projectile hits the

rubberscreenlikeadrum.

Basedonmathematicalmodelsand theactualbullet impact

Figure6.Coordinatesofshot(x,y)determinedbyextrapolationofshockwaves/sounddetectedbythesensors

RESTRICTED

y

Acoustic sensors located within the box

x (0, 0)

𝑥𝑥𝑎𝑎, 𝑦𝑦𝑎𝑎 , 𝑡𝑡𝑎𝑎

Figure7.Simulatingtrajectoryofabulletbeyondthephysicaldistance

𝑑𝑑 = 𝑣𝑣2 sin 2𝜃𝜃 𝑔𝑔 , 𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑔𝑔 𝑜𝑜𝑓𝑓 𝑎𝑎 𝑓𝑓𝑓𝑓𝑎𝑎𝑓𝑓 𝑔𝑔𝑓𝑓𝑜𝑜𝑔𝑔𝑓𝑓𝑑𝑑

Simulated Trajectory Actual Trajectory

Range (ft)

Heig

ht (f

t)

pointonthescreen,thetrajectoryofthebulletisthencalculated

andsimulated fordistancesbeyondthephysicaldistanceof

thescreen(seeFigure7).


MovingElectronicTargetrySystem

Themulti-tierrangeattheMMRCfeaturesasingle-railMETS

co-inventedbyDSTA(seeFigure8).Thenewsingle-railMETS

eliminatesanypotentialline-of-sightissuesfortraineeslocated

onlowerlevelsfiringathigher-leveltargets.unlikeconventional

movingtargetrysystemsasshowninFigure9,thenewdesign

requires a shorter installation depth (0.96m instead of 2.4m)

andutilisesonlyonemotor todrive thesingle railof targets.

Hence,forarangewith10firinglanes,thisnewdesignsaves

anaverageofabout70m2perrange.

MAINTAINING HIGH SAFETY STANDARDS

AsthisistheSingaporeArmedForces’(SAF)firstindoormulti-

storey live-firing facilityof thisscale,maintaininghighsafety

standardsforfirers,facilitypersonnelandthefacilityitselfisof

thetoppriority.AllrangesintheMMRCarefittedwithrobust

ballisticprotectionsystemsthatweresubjectedtostrictand

rigorousvalidationpriortoinstallation,tomitigatethehazards

associatedwithfiringinanindoorrange.

OverallRoundContainment

TheMMRCisdesignedforfullroundcontainmentandhence

there is noneed to cater for aweapondanger areaoutside

the range. In order to ensure total round containment, steel

escalatorbullettrapsandgranularrubberbulletcatchersare

installedinthe50mand100mrangestocontaintheammunition

allowed intherangesfromdesignatedfiringpoints.Portable

hard traps that are constructed from armoured steel are

overlaidwithshreddedrubberpanelmountedonspacerbars

andareusedasbullettrapsintheurbanoperationsRange.

Inaddition,allrangeshavewalls(inclusiveofcolumns),floors

andceilingsthataremadeofreinforcedconcretedesignedin

compliance with requirements stipulated in the JSP403 uK

range safety handbookand verified tobeable to effectively

containtheroundsfiredintheMMRC.

Hazardous Ricochet and Backsplash Hazards

Ballisticprotectionsystemsareinstalledtopreventhazardous

ricochets and backsplash associated with firing in indoor

ranges. They also serve as additional protection to prevent

roundescapementandprotecttheranges’concretewallsand

ceilingsfromoccasionalshots.

Shreddedrubberpanel(SRP)isusedextensivelyintheranges

to prevent hazardous ricochet and backsplash hazards to

the firers.Majority of theSRParemountedon spacerbars,

whichinturnareweldedtoarmouredsteelplate.Thissystem

isusedfortherangewalls,ceilingbaffles,targetmechanism

protection system and floor baffles to prevent ricochet and

backsplash hazards to the firers (see Figure 10). Bullets

thatpass through theSRPand strike the steelwouldeither

fragmentordeform,andthedensityoftheSRPpreventsany

fragmentsordeformedprojectilesfrompassingbackthrough

intotherangearea.

Restricted

Bottom Tier

Top Tier

UNEQUAL LINE OF SIGHT

Restricted

Bottom Tier

Top Tier


Figure8.Single-railMETS Figure9.ConventionalMETS

Restricted


Restricted


TRANSFoRMINgRANgEPRACTICESWITHTHEMuLTI-MISSIoNRANgECoMPLEx


Redirectiveguardsarepositionedimmediatelyinfrontofand

abovethethroatofthebulletcatchers(seeFigure11).Inmost

cases,theredirectiveguardswillbewithintheweapondanger

area and therefore require thicker armoured steel plate. The

guards are hung at an angle to ensure that projectiles are

directedintothethroatofthebulletcatcherandwillcontinue

theirtravelintothedecelerationchamberofthebulletcatcher.

Inaddition,theangleatwhichtheplatesarehungwouldalso

ensure that the pitting on the steel plates would not cause

hazardousricochetandfragments,andinjurethefirers.

Figure10.Ceilingbafflesin50mRange1and2

Figure11.Redirectiveguardsusedatthebullettraps

Protection of Personnel in Control Room and Learning Gallery

Ballistic glass is installed in the control room and learning

gallerytoallowtrainersandsoldierstoobservethelive-firing

activitiessafely(seeFigure12).Theballisticglassisratedto

beabletowithstand7.62mmx51mmNATorounds,andits

performancewasverifiedduringcomponenttesting.


EnsuringSafetyduringFireandMovement

Toensurethesafetyofsoldiersduringlive-firetacticaltraining

involving the firing of small arms, a safety angle is applied

betweenthesideofthefiringpointandthenearestfireronthe

adjacent lane.This safetyanglecomprisesa ricochet safety

angleplustheappropriateconeoffireangle.

Forfireandmovementtraining2,itwasdeterminedthatawider

firinglanewasrequiredforeachfirer.Insteadofhaving10firing

lanesforstaticfiring,therangehadtobereconfiguredtohave

just seven firing lanes. To comply with the requirement, the

teamsuccessfully redesignedthe range toallowthe10-lane

range tobeeasily reconfigured intoaseven-lane range (see

Figure13).Thiswasachievedbymountingthestaticelectronic

targetsonmovablerailswhichallowthetargetstobemoved

manuallyandlockedintoposition.

Figure12.Ballisticglassinstalledatthecontrolroomandlearninggallery

Figure13.Comparisonbetweena10-laneandseven-lanerange



EnvironmentalSafety

The ranges in the MMRC are ventilated mechanically. The

designoftherangeventilationsystemiscrucialinminimising

the accumulation of contaminants in the range as well as

in preventing contaminants from flowing into the adjacent

spaces. This reduces the risk of contaminants inhalation by

soldiersandinstructors.

The range ventilation system is designed to create a low-

velocity,uni-directionalairflowtowardsthetargetryareawhere

anexhaustsystemfiltersthecontaminantsbeforedischarging

theairintotheexternalenvironment(seeFigure14).Forsuch

acomplextask,numericalmodellingusingComputationFluid

Dynamics (CFD)wasused indesigning the rangeventilation

system for the first time inSingapore (seeFigure15).Areas

of stagnation and high air velocity were identified and the

designwasrefinedoverseveraliterationstoimproveairflow.

Topreventtheoutflowofcontaminantstotheadjacentspaces,

theairpressureintherangeismaintainedataslightly lower

pressure than the surrounding spaces. Differential pressure

sensors are installed to measure the pressure differential

between the range and its adjacent spaces. The Building

AutomationSystemmonitors these sensors in real timeand

regulates the speed of the exhaust system to maintain a

differentialpressureof5Pato15Pabetweentherangeandthe

adjacentarea.

Figure14.Schematicofrangeventilationsystem

Figure15.CFDmodellingofairflowvelocityintherange


BENEFITS OF THE MMRC

The operationalisation of the MMRC has brought about many

benefits to the SAF and commercial entities.

Optimising Allocation of Resources

To ensure training hours are optimally utilised, a soldier-centric

workflow was created to enhance the positive experience

of soldiers from registration to exit. All routine, non-core

administrative tasks were streamlined and outsourced to

a commercial entity, allowing the SAF to focus on core

competency development. Conversely, the commercial entity

is able to leverage its creativity and experience to manage

its operation and staffing efficiencies to meet contractual

performance requirements. Unlike traditional outdoor ranges,

training is not subjected to external weather and lighting

conditions in the MMRC. In particular, soldiers do not need

to cease training due to a downpour or wait long hours after

their daytime training for nightfall before proceeding with their

nighttime training. Hence, the waiting time and time needed

for administration and logistics matters have been efficiently

converted into training time for soldiers.

Aligned Vision in National Development

Some of the existing outdoor live-firing ranges in Singapore

are sited on valuable state land that have to be returned to

the Singapore Land Authority (SLA) progressively for land

redevelopment. The DSTA Integrated Project Management

Team performed detailed studies and planned strategically for

the land required to build the MMRC and reduced the total

footprint for constructing seven typical outdoor ranges by at

least 3.7 times. This approach also enabled the Ministry of

Defence (MINDEF) and the SAF to achieve total land savings of

22 hectares, equivalent to 30 football fields. With the delivery

of the MMRC, the land occupied by existing outdoor firing

ranges can be returned to SLA progressively. The MMRC also

minimises the need to develop new conventional ranges.

CONCLUSION

The MMRC has become a critical training facility of the Army

and provides a pleasant experience for our current generation

of national servicemen. It has also paved the way for future

training infrastructure and systems development. It is at

the forefront of training development and has crossed new

boundaries, creating a paradigm shift in the way the SAF

conducts its live-firing training. The capabilities delivered by

the MMRC have provided the SAF with a safe yet challenging

environment to train and sustain instinctive and judgemental

shooting competencies, and instils a positive experience in

every soldier that trains in the facility.

ACKNOWLEDGEMENTS

The authors would like to thank the late Deputy Director

(Operations and Support - Army) Mr Ng Tiong Lee for his

guidance and for forming this team. The authors would also

like to acknowledge the guidance and ideas from the senior

management of DSTA, MINDEF and the SAF, as well as

colleagues who have contributed to the successful delivery of

the MMRC.

ENDNOTES

1 The MMRC was awarded the IES Prestigious Engineering

Achievement Award (2014), Defence Technology Prize Team

(Engineering) Award, MINDEF Innovation Project and Savings

And Value Enhancement (SAVE) Awards in 2013.

2 Fire and movement training is a shooting drill that requires

soldiers to take cover, take aim, fire and move forward to

another location to repeat the drill.

TRANSFORMINg RANgE PRACTICES WITH THE MULTI-MISSION RANgE COMPLEx


BIOGRAPHY

LIM Peter is a Programme Manager

(NetworkedSystems)wholedinthedelivery

of the Multi-Mission Range Complex

(MMRC).PetergraduatedwithaBachelorof

Technology(ElectronicsEngineering)degree

and a Master of Engineering (Electrical

and Computing) degree from the National

university of Singapore (NuS) in 2001

and2005 respectively.He furtherobtainedaMasterofScience

(Software Engineering) degree from the Naval Postgraduate

School,uSA,in2011.

YEOQiuLingTammisupportedthedelivery

of targetry and simulation systems in the

MMRC when she was a Project Manager

(Networked Systems). She is currently

pursuing a Master of Science (operations

Research)degreeattheNavalPostgraduate

School, uSA. Tammi graduated with

a Bachelor of Engineering (Electrical

Engineering)degreeandaMasterofScience(IndustrialSystems

Engineering)degreefromNuSin2006and2012respectively.

LIM Meng Kee Johnson is a Manager

(Buildingand Infrastructure). Hehasmore

than 15 years of experience in designing

and commissioning air-conditioning,

ventilation and fire protection systems for

military facilities. Johnson graduated with

a Bachelor of Engineering (Mechanical

Engineering)degree fromNanyangTechnologicaluniversityand

aMasterofScience(IndustrialandSystemsEngineering)degree

fromNuS in1996and2008 respectively.He furtherobtaineda

MasterofScience (FireProtectionEngineering)degree from the

universityofMaryland,CollegePark,uSA,in2010.

LAU Chin Seng Eric was a System

Manager (Systems Management). As the

range safety engineer for the MMRC, he

provided technical assessmentandadvice

inensuringthatthedesignoftherangesin

MMRCcompliedwithexistingrangesafety

guidelines.Erichas representedSingapore

in the annual International Range Safety

Advisorygroupmeetingwhichcomprisesexpertsandpractitioners

inrange.HegraduatedwithaBachelorofEngineering(Chemical

Engineering)degreefromNuSin2006.




INNoVATIVEAPPRoACHESFoRTHEADVANCEDCoMBATMANSySTEM

INTRODUCTION

LIMWeiQiang,PEHHanYongLester

ABSTRACT

TheSingaporeArmywasequippedwiththeAdvancedCombatManSystem(ACMS)in2010toenhancethesurvivabilityandcombatcapabilityofsoldiersinurbanoperations.WhiletheACMSachieveditsobjectives,theProjectManagementTeam(PMT) identifiedareasfor improvementwhichresultedina lightersystemandanenhanceduser interfaceutilisingmulti-touchtechnologysimilartocommercialelectronicdevicessuchassmartphones.

ThisarticlecapturesthejourneytakentodevelopanddesignanewerandlightweightvariantoftheACMS–theACMSiLITE, using a soldier-centric approach. The article also shares lessons learnt from this process, and introduces someconceptsand futuristic technologies that thePMT isexploring toenhance thefightingcapabilitiesof thesoldier in theSingaporeArmedForces.

Keywords:urbanoperations,ACMS,CoTS,smartphone,soldier-centric

With the prevalence of global urbanisation, the Singapore

Armywill inevitablyneedtoengageinurbanoperations.The

urban environment presents a whole newmulti-dimensional

battlefield – one which not only requires changes to

conventionalcombatoperationsandtactics,butalsoaneed

toovercomethestrategicadvantageaffordedtoanadversary

thatisconcealedandentrenchedinanurbanenvironment.

The Advanced Combat Man System (ACMS) is an urban

fightingsystemfortheThirdgenerationArmy.Itisdesignedto

addressthechallengesfacedinurbanoperationsbyenhancing

commandandcontrol(C2),situationalawareness,survivability

andlethalityofthesoldier.Inadditiontobeingequippedwith

theACMS, soldiers are equippedwith remote sensors such

as surveillance robotsandkeyhole sensors.With theACMS

andthesesensors,soldiersbecomepartofanetworkedforce.

Soldiers’ situational awareness is thus enhanced, allowing

themtoengagetheirtargetseffectively.Soldiersarealsoable

tonavigateaccurately through theurbanbattlefield toavoid

knowndangerareas.Thekeycomponentsof theACMSare

illustratedinFigure1.


Figure1.overviewoftheACMS

Figure2.Thenetworkedsoldierandhiscomplementarycapabilities

the lethality and situational awareness of their units (see

Figure 2). The soldiers themselves become sensors on the

ground,providingreal-timeinformationtothecommandersfor

improvedbattlefieldcoordination.

TheACMSalso allows selected appointment holders to tap

the wider resources of the battalion, such as the TERREx

InfantryCarrierVehicle(ICV)forfiresupportandsustenance,

andevenutilisehighercommandresourcestofurtherenhance

1. Personal Radio −  Enables soldiers to share

voice and data information with his peers and commanders.

−  Contains in-built Global Positioning System (GPS) for positional tracking.

2. Head Mounted Display (HMD) -  Allows soldiers to view

location of friendly and hostile forces on digital maps.

3. Portable Computer -  Processes and

analyses data from GPS, sensors and other ACMS sets.

4. Communication Keypad -  Enables soldiers to

access hot keys such as “Call-For-Medic” and “On-Contact”. Request will be sent to peers and commanders.

5. Weapon Interactor −  Allows the soldier to

perform round corner firing without exposing himself.

1

5

2

4

3












1

5

2

4

3












1

5

2

4

3












1

5

2

4

3


A FORCE MULIPLIER

Priortothefieldingof theACMS,battlefield informationwas

disseminated verbally. This limited the speed and accuracy

of engaging targets,with soldiers requiring a longer time to

interpret commands, identify targets and engage them. The

ACMS,withitsgraphicalpresentationofbattlefieldinformation,

has enabled soldiers to share up-to-date information in a

preciseandswiftmanner.

TheACMShasalsoconnecteddismounted soldiers to their

TERREx ICV. This has allowed the sharing of a common

operational picture and provision of support fire from the

weaponsystemoftheTERRExICV.

REINVENTING THE ACMS

WhiletheACMSmetitsintendedobjectivesintermsofweight

andcapabilities,andbenchmarkedwellagainstotherleading

soldier modernisation programmes globally, the Project

Management Team (PMT) was cognisant of the need to

continuallyreinventthesystemtoensurerelevancy.oneareaof

interestwasinthefieldofpersonalinfocommtechnology.The

introductionofsmartphonesredefinedthetraditionalconcept

of a phone. Features like emails, messaging, photography,

video,maps,gamesandotherapplicationswerepackedinto

an elegant, lightweight device. A multi-touch user interface

replacedphysicalbuttonsandkeyboards.

Not surprisingly, smartphones were well received by

consumers.Itdidnottakelongfortheconsumermarkettobe

saturatedwithsmartphonesandsimilarlifestyletechnologies.

Thesmartphonesoonbecameapervasivelifestyledevice.

As the smartphones packed tremendous computing power

in a small form factor, the PMT envisaged that the use of

smartphonesintheACMScouldachieveweightreductionand

enhanceusability.

THE BIRTH OF A LIGHTWEIGHT ACMS VARIANT

Inlightoftherapiddevelopmentoflifestyledevices,thePMT

determinedtheneedtodevelopanewversionoftheACMS

that would be more useful and better received by soldiers.

Theopportunityarosewhenfeedback,gatheredthroughtrials

and exercises, indicated that itwas desirable to reduce the

weightoftheACMS.Itwasalsonotedthatthelowerechelons

(section-levelandbelow)didnotrequirethefullsuiteofACMS

capabilitiessincetheywouldbeoccupiedwiththeimmediate

fire fight. Thus, the idea to create a new lightweight variant

oftheACMSwhichwouldonlyhavetheessentialcapabilities

required at the tactical levels, was conceived. This was the

iLITE.

It was envisaged that the iLITE would leverage both the

existing technologies employed for the in-service ACMS as

well as commercial-off-the-shelf (CoTS) technologies. The

iLITEwouldbelighter,simplerandmoreintuitivetousewhile

meeting the required operating duration and conditions. It

wasclearthatmaximisingbatterylifewouldbeakeydesign

consideration as this would reduce the number of batteries

required and consequently the overall system weight. To

make the iLITE simple and intuitive to use, familiar CoTS

smartphoneswouldbeadoptedasthemediumthroughwhich

thesoldier interactswith thesystem.Thesebecamethekey

designconsiderationsfortheiLITE.

DEVELOPMENT OF THE ILITE

QuesttoAdoptCOTSSmartphone

At the outset of the iLITE’s design, the PMT had initiated

the use of a CoTS smartphone as the input, display and

processor of iLITE. However, knowledge of hardening third

partysmartphoneswasstilllacking.Therewerealsoconcerns

about the ability of CoTS devices to comply with military

ruggedisation standards suitable for soldier use. Hence,

thePMTwas facedwith theundesirable optionof adopting

bespoke smartphones with features modelled after CoTS

smartphones.

These concerns did not deter the PMT from their vision of

an iLITE design based on CoTS smartphones. Moreover,

thePMTassessed that thecommercialsector,with itshuge

R&Dfunding,wouldcontinuetospearheadinnovationsinthe

fieldofpersonal infocommtechnology. Inaddition, thehuge

commercialmarketprovidedbetterleverageforaccesstothe

latesttechnologiessuchashigh-speedprocessorsandlonger

batterylife.ThisconvincedthePMTthatabespokesmartphone

wasunlikelytomatchuptotheCoTSsmartphonesthatthe

soldierswereaccustomedto.Itwasalsolikelythatabespoke

smartphonewouldbemadeobsoletebeforelong.

ThePMTovercamethechallengesinvolvedinhardeningCoTS

smartphonesbytappingthewiderexpertisewithinDSTA.At

thesametime,avarietyofCoTSsolutionsforwaterproofing


and shock protection for the more popular smartphones

was entering the market, potentially addressing the Army’s

ruggedisationneeds.

SelectingSmartphoneforiLITE

ThesearchforasuitableCoTSsmartphoneforiLITEbeganwith

theselectionofasuitablemobileoperatingSystem(oS).The

AndroidoSwaseventuallyselectedoverotheroSsduetoits

abilitytobecustomisedtomeetthePMT’srequirements,and

ithavingthelargestmarketshareatthetimeofdevelopment

whichwouldreducetheriskofhardwareobsolescence.

A study was then conducted on a range of leading CoTS

Androidsmartphonesintheconsumermarket.Itcametothe

PMT’sattentionduring themarket researchthateachphone

manufacturerhadsomethingdifferent toofferandtherewas

nosuchthingasabestsmartphone.

After much deliberation, the PMT decided on a set of key

considerations that would guide the selection of a suitable

smartphone. First, it should have the largest market share

amongtheAndroidsmartphonemakerswhichwouldincrease

theavailabilityofCoTSruggedisedcasings.Second,itshould

possess superior technical performance to provide greater

utility in the iLITE.Third, it shouldhavea screenwhichwas

large enough for viewing while still allowing operation with

one hand. Last of all, it should comewith a power-efficient

screentechnologythatwouldprolongtheoperatingduration.

A smartphone was finally selected based on the above

guidelines.

StrategyforWeightReduction

TheimplementationofaCoTSsmartphonewasacrucialpart

of the PMT’s strategy to reduce system weight. Instead of

incorporatingtwodistinctsubsystems(theportablecomputer

andthecommunicationkeypad)withinthein-serviceACMS,the

iLITEsmartphoneservedastheintegratedmobileprocessor,

inputanddisplaysubsystems.Thisreducedthesystemweight

by more than 50%. The lower power requirements of the

processor1used in theAndroidoSalsomade itpossible to

reducethenumberofbatteriesrequiredfortheiLITE.

There are two key differences between the current ACMS

andtheiLITE.First,theiLITEusesaCoTSsmartphoneasan

integratedmobileprocessor anddisplay subsystem, instead

of a separate soldier computer and head mounted display.

Second, the iLITE requires only one battery for the entire

missionasopposedtothreeintheACMS.Thekeydifferences

areshowninFigure3.

Figure3.KeydifferencesbetweenACMSandiLITE

Current ACMS – Full Suite of Software Features

Communication Keypad

Battery (x3)

HMD

Portable Computer

Navigation System

Smartphone

Battery (x1)

Comms System

Navigation System

iLITE – Essential Software Features

for Tactical Levels



Although the Army was prepared to trade ruggedness2 of

the smartphones for weight reduction, the PMT remained

steadfast in pursuing CoTS solutions to meet the Army’s

ruggedisationneeds.Atthattime,thecommercialmarketwas

rollingoutavarietyofproductsandsolutionsforwaterproofing

andshockprotectionofpopularCoTSphones.Theseinclude

waterproof phones, ruggedisedprotective casings and even

water-repellentnano-coatings.ThePMT’seffortpaidoffwhen

usertrialsvalidatedthattheCoTSprotectivecasesprovided

adequateruggedisationforthesmartphones.

StrategytoImproveSystemErgonomics

To improve soldier receptivity towards the iLITE, human

factors engineering expertise was engaged to enhance its

ergonomics.Thisincludedwearabilitystudiestodeterminethe

optimalplacementoftheiLITEcomponentsonthesoldier,as

wellasthedesignoftheiLITEC2graphicaluserInterfacefor

greatereaseofuse.Numerous trialswereconductedduring

the development of iLITE as part of an iterative process to

gatheruserfeedbackonthesystemdesign.

A phased equipping approach was also implemented for

the iLITE,where lessons learnt and feedback received from

preceding deliverieswould be incorporated into subsequent

batchdeliveries.

OPPORTUNITIES BROUGHT FORTH BY ILITE

The advent of the smartphone also provided functionalities

beyondtheiLITErequirements.Forinstance,thesmartphone

could be used by soldiers to access e-learning materials

suchasthoseprovidedontheLEARNetplatformattheirown

time.otherusefulutilityapplications thatwerecommercially

available in the Android’s application store (such as the

compassandrangefindingtools)couldalsobemadeavailable

tothesoldier.

onamorecomplexscale,theiLITEcouldbeintegratedwith

other systems such as the Tactical Engagement System3

(TES) to achieve weight and cost savings. Through suitable

applications, the iLITE’s smartphone could also be used

to control unmanned platforms to complement a soldier’s

mission.

LESSONS LEARNT

While CoTS products can be cost-effective solutions, the

fastpaceoftechnologicaladvancementsinCoTSinfocomm

technologies can render these solutions obsolete in two to

three years. A comprehensive obsolescence management

planwould need to be adopted to facilitate the insertion of

newtechnologiesandtacklepotentialhardwareobsolescence

issues:

a) Modular Architecture - System components should be

keptmodular,whereverpossible, to ensure that anupgrade

ormodificationinonesubsystemwouldnothaveasignificant

impactontherestofthesubsystems.

b) Phased Equipping - A phased equipping approach

shouldbeadoptedtoenableincorporationofrefinementsdue

totechnologicalimprovementsanduserfeedback.

c) Software Portability - The C2 software should be

designedforeaseofmodificationandexpansioninanticipation

of future upgrades, such as hardware changes or insertion

of new technologies. For the Android oS in particular, C2

functionalitiesshouldbeimplementedattheapplicationlevel,

whereapplicable, toallow thedevelopedC2software tobe

portedtothelatestfirmwareversionwithminimaleffort.

d) Regular Reviews - The PMT should keep abreast of

technological improvements and development in market

trendsinordertoassesstheevolutionofCoTSsmartdevices

andoS trendsat regularpre-planned reviews. If there is an

anticipatedshift in theoSmarket shareorobsolescenceof

existing hardware that could impact the project severely,

the obsolescence management plan should be updated

accordingly.

FUTURE TECHNOLOGIES

TheintroductionofiLITEhasputtheArmyattheforefrontof

soldier digitisation. It is imperative to stay updated on new

technologiesthatarebeingdevelopedconstantlythroughout

theworld.Thesenewtechnologiesneed tobeharnessedat

appropriate junctures to enhance mission effectiveness for

soldiers. Several emerging technologies that could possibly

seeapplicationintheACMSareasfollows:

a) WearableTechnologies -Thecommercialmarkethasa

varietyofwearabletechnologies,rangingfromsmartwatches

toeyewearsuchasthegoogleglass.Althoughthesewearable


technologieshaveyettogaintraction,itremainsaninteresting

andexcitingareatomonitor.Figure4highlightssomepotential

applicationsofwearabletechnologies.

b) SimultaneousLocalisationandMapping(SLAM)-SLAM

allowssoldierstoconstructamapofanunfamiliarenvironment

as they are navigating through it. This technologywould be

extremelyusefulwhensoldiersnavigatethroughbuildingsand

internal spaces. This could be achieved by networking the

iLITE to a SLAM-enabled unmanned ground vehicle (ugV).

Thistechnology,whilestillinitsinfancy,wouldfurtherenhance

thecapabilitiesofthenetworkedsoldier.

c) VoiceandGestureRecognitionApplications-Voiceand

gesturerecognitionapplicationswouldenhancetheabilityof

soldierstocontrolsensorsassetssuchasugVs.Inaddition,

voicerecognitionwouldenablesoldierstoperformthedesired

functiononthesmartphonewithouthavingtogothroughthe

processmanually,thusgivingrisetoquickerresponses.

d) WirelessCharging-Wirelesschargingcouldbeexplored

toprovideadditionalchargingoptionsfortheArmy.Potential

applicationsincludewirelesscharginginvehiclesthatenables

the batteries on soldiers to be charged while in vehicles.

Vehiclescouldalsoserveashot-spotsforwirelesscharging,

allowingsoldierstochargetheirdevicesincloseproximityto

thevehicle.

e) Energy Harvesting - As soldiers carry more electronic

gear,thegrowingenergydemandsposeanenormousburden

onthelogisticssupportsystem.Soldierswillalsohavetocope

withtheweightofextrabatteries.Energyharvestingtechniques

could be explored to improve the sustenance of soldiers in

the battlefield. one promising technique is biomechanical

harvesting, inwhichelectricity isgeneratedviabodymotion

suchaswalking.

CONCLUSION

The PMT’s decision to design the iLITE based on a CoTS

smartphonewas forward looking. The iLITE hasmet its key

objectivesofweightreduction, improvedsystemergonomics

andintuitiveness.ThisenhancedACMSvariantimprovesthe

Army’sfightingcapabilitiesgreatlybypackingmorepunchata

lowerweightrequirement.

The successful implementation of CoTS technologies in

iLITEhasproventheviabilityofleveragingCoTSsolutionsfor

operational equipment. Trade-offs were managed to deliver

cost-effective CoTS solutions that not only met the Army’s

requirements, but were also highly adaptive to the dynamic

paceoftechnologicaladvancement.

ACKNOWLEDGEMENTS

TheauthorswouldliketothankMrNgyuenCheongandMs

Ang Chai yit for their guidance and valuable inputs in the

preparationofthisarticle.

Figure4.Potentialapplicationsforwearabletechnologies

1. Eyewear −  Augmented reality

technology to enhance soldier’s situational awareness

−  Potentially guide soldiers to objectives/targets

3. Smart Watches −  Could be paired with

smartphone to alert the user to incoming messages

2. Smart Clothing −  Comes with sensors to

keep track of soldier’s health indicators e.g. heart rate, and alert medics when soldier is injured. In-built flexible batteries would power these sensors

−  Able to switch camouflaging pattern according to environment

−  Sensors to detect the presence of chemical agents in the environment



ENDNOTES

1 ARMprocessorsareafamilyof32-bitReducedInstruction

SetComputer(RISC)microprocessorsdevelopedbyAdvanced

RISCMachines.

2 SmartphonesareCoTSproductswithdesignsthatcannot

bemodifiedeasily.

3 TES is a laser-based system currently used in combat

trainingexercisestosimulatetheeffectsofweapons.

BIOGRAPHY

LIM Wei Qiang is an Engineer (Land

Systems) involved in identifying and

leveraging technologies to enhance power

managementandloadreductionforsoldiers.

Hewaspartoftheteamthatdevelopedthe

Advanced Combat Man System (ACMS)

and the iLITE. Wei Qiang graduated with

a Bachelor of Engineering (Mechanical

Engineering) degree from Nanyang Technological university in

2010withspecialisationinEnergyandtheEnvironment.

PEHHanYongLesterisaProjectManager

(Land Systems) currently overseeing

the development of the ACMS iLITE. In

particular, he is involved in leveraging

technologies of interest to further enhance

the capabilities of the ACMS iLITE. under

the DSTA undergraduate Scholarship,

Lester obtained a Bachelor of Science in

Engineering (Electrical Engineering) degree from the university

ofMichigan,uSA,aswellasaMasterofScience(Management

ScienceandEngineering)degreefromStanforduniversity,uSA,

in2007and2008respectively.




TECHNoLogICALADVANCEMENTSANDINNoVATIoNSINCoMBATENgINEERINgEQuIPMENT

INTRODUCTION

Combat engineers play an essential supporting role in any

force as they enhance forcemobility for friendly troops and

hinderthemobilityofadversaries.Someactivitiesundertaken

bycombatengineersincludebridging,obstacleclearance,soft

groundmobilityenhancementaswellasmineandimprovised

explosivedevice(IED)neutralisation.

Traditionally,combatengineeringtaskshavebeenmanpower

intensive, time-consuming, logistically demanding and

dangerous.Now,technologicaladvancementsandinnovations

havecontributedtocombatengineeringinthreemainareas:(a)

automation toenable leanermanningofcombatengineering

equipment;(b)reductionoftimespentoncombatengineering

tasks; and (c) improvement of man-machine interfaces and

ergonomicstomakesystemssaferandeasiertouse.

This article presents and explores the role of technological

advances and innovations in the evolution of combat

engineeringequipment,withafocusonthreeareasofcombat

engineering tasks: fixed bridging, wet bridging, and mine

clearing.

PHUAZhengqiDaryl,TANChun,WONGYeeYinKimberly

ABSTRACT

Combatengineeringisessentialinenablingforcestoovercomediverseobstacles.Traditionally,combatengineeringtaskshavebeenmanpowerintensive,time-consuming,logisticallydemandinganddangerous.Now,moderncombatengineeringequipment requires less manpower and logistics to operate, features people-centric designs and allows tasks to becompletedinasaferandfasterway.

Thisarticleillustratestheroleoftechnologicaladvancementsandinnovationsincombatengineeringequipment,andshareshowcombatengineeringequipmentisexpectedtoevolveovertime.

Keywords:combatengineering,militarybridges,mechanised,mineclearing

FIXED BRIDGING

Military fixed bridges are required to be strong enough to

transportheavymilitaryvehicles,lightenoughtobetransported

easily and simple enough to be constructed quickly. Bridge

engineers use the concepts of bendingmoments and shear

forces to design efficient bridges to achieve the optimal

balancebetweenspan,strengthandweight.

ClassificationofBridges

Military bridges are classified by their Military Load Class

(MLC)inaccordancewithNATo’sStandardisationAgreement

20211. The MLC is a single number which represents the

strengthofthebridgeandclassofvehicleitcantransport,and

isproportionatetothemaximumbendingmomentandshear

forcethebridgecanwithstandwithoutundergoingirreversible

yield.

VehiclesaresimilarlyassignedanMLCratingdependentonthe

maximumbendingmomentandshearforceitexertsonabridge.

This isdependenton thevehicle’sweight, length,numberof


axlesandwheelloading.Thedeterminationofavehicle’sMLC

rating involves tedious calculations of the maximum shear

forceandbendingmomentover44predefinedbridgespans

foratotalof88testcases.DSTAhasdevelopedanin-house

applicationtoautomatethesecalculations,allowingengineers

toestimateavehicle’sMLCquickly.

InnovationsonModernFixedBridges

Modernfixedbridgeshavecomea longwaysince theearly

days, when crudely bundled brushwood called fascines

and one-piece bridges were used to cross tank ditches.

Technologicaladvancementsinweldingandthedevelopment

ofstrongermaterialshaveenabledmodernbridgestobelonger,

strongeranddeployedquickly.Someoftheseinnovationsare

asfollows.

Launching of Long Bridges

Carryingbridgessignificantlylongerthanthelaunchingvehicle

hampers the mobility of the vehicle and makes launching

unwieldy. To enable the launching vehicle to carry longer

bridges,thesebridgeswereinnovativelyfoldedinhalf,andthen

launchedbyunfoldingthemwithascissor-launchmechanism

(seeFigure1).However,scissor-launchedbridgeshavealarge

visualsignatureduringlaunching–posingchallengesforthem

Figure1.Scissor-launchmechanismonaM60A1AVLB(©Quihuis/File:M60A1ArmoredVehicleLandingBridge.jpg/http://www.news.navy.mil/view_single.asp?id=5015/PublicDomain)(left)andhorizontal-launchmechanismona

Leopard2ArmouredVehicleLaunchedBridge(ong,2013)(right)ReprintedwithpermissionfromDefenceMediaCentre(DMC)

tobelaunchedinareaswithoverheadobstaclesandmakingit

easierforenemiestospotthem.

Further improvements have been made to the designs of

assaultbridgesandahorizontal-launchmechanismhasbeen

developedinstead.TheLeopard2ArmouredVehicleLaunched

Bridge (AVLB), introduced into theSingaporeArmedForces’

(SAF)servicein2010,usesthehorizontal-launchmechanism

(seeFigure1).Anotheradvantageofthisdesignistheabsence

of hydraulic components in the bridge that improves its

reliabilityandservicelife.

Automation of Bridge Systems

Earlyfixedbridgeswereconstructedmanuallywhichcalledfor

significantmanpower and time.Thenewgenerationbridges

are launched by bridge layer vehicles with fully automatic

launching and retrieval modes operated by a crew of two,

whichcanlauncha26mbridgeinlessthaneightminutes.In

theLeopard2AVLB,thisismadepossiblethroughtheuseof

anelectronicallycontrolledhydraulicsystemandanarrayof

sensorsandactuatorsonboardthebridgelayervehicle.With

thesesensorsandcamerasmonitoringthelaunch,theoperator

isabletolaunchthebridgewithouttheneedtocomeoutofthe

vehicleandopenthehatch,making itsafer for theoperator.


Figure2.M3g’sSchottelPumpJetup-close(C.Teuert,2014)(left)SchottelPumpJet360°movement(Schottelru,2011)(right)

forma100mbridgein25minutes-ascomparedtotheHeavy

AssaultBridgewhichrequired30trucks,60menand4hours

toconstruct.This isasignificantdecrease in themanpower,

logistics and construction time required. The technological

innovationsareasfollows.

Propulsion Systems

The earlier amphibious vehicles used a propeller for water

propulsion. The latest M3g utilises pump jets instead.

Traditionalpumpjetdesignsutiliseareversingbucketattached

totheendofthenozzletoachievereversethrustandbraking

ofthevehicleinwater,andplatestomanipulatethedirectionof

thejetforsteering.TheM3g’spumpjetisaninnovativedesign

where the nozzle can be rotated 360° to providemaximum

manoeuvrabilityofthevehicleinwater(seeFigure2).

Control System

The M3g has one pump jet at the fore and aft of the

vehicle, allowing the system to achieve a very high level of

manoeuvrability which a traditional propeller is unable to

achieve.Inordertocontrolthefullrangeofcomplexmotions

of the dual pump jets, an innovative control system was

implementedontheM3g(seeFigure3).Thisintuitivecontrol

allows the coxswain to retain full control over the complex

motionoftheM3gwithminimaltaskloading.

WET BRIDGING

Wet bridging (also known as float bridging) is a solution to

overcome long wet gaps which fixed bridges are unable to

overcome. This is done through the construction of float

bridges or forming of ferries. Wet bridges build upon the

principles of fixedbridging, but add a degree of complexity

withflotationconsiderations.

OriginsofWetBridgesandFerries

Wetbridginginvolvesthelinkingupofpontoonsorboatsuntil

thewetgapisbridgedoraraftofarequiredsizeisformed.The

traditionaltypesofwetbridgesolutionswerelogisticintensive,

and required multiple trucks to transport the pontoons,

tugboats, decks, ramps and associated accessories. For

example, a100massaultboatbridge requireda teamof60

men,20logistictrucksandover90minutesforconstruction.

InnovationsonModernFloatBridges

Compared to assault boat bridges and early amphibious

vehicles,modernfloatbridgesaremoreeffectiveandeasierto

operate.Forexample,theM3g,whichwasintroducedintothe

SAF’sservicein2008,requiresonlysixvehiclesand24mento


Furthermore, multiple vehicles connected together to form

a ferry canbenetworked for operationbya single console,

allowingasingleoperatorcontrolovertheferry.Thisimproves

thecontroloftheentireferryandenablesleanmanningofthe

system.

Reducing the Logistics Requirement of Float Bridges

PriortotheSAFacquiringtheM3g,thedesignoftheoriginal

equipment manufacturer (oEM) included three ramps per

vehicle. While this was adequate for the construction of a

Figure3.DemonstrationofthemarinecontrolsoftheM3g.Thelevelcontrolsthepowerandprimarydirection,andthe“horns”controltherelativerotationoftherig.

floatbridge, itwas insufficient for theconstructionofatwo-

vehicleferryinopen-coupleconfiguration2.Alogisticsvehicle

to transport the three additional ramps is required for this

configuration.

The SAF’s M3gs were modified to accommodate an extra

rampeach(seeFigure4).Modularpanelscanalsobeplaced

laterallybetweentworampstoformthethirdramp(seeFigure

5).Thisinnovationeliminatedtheneedforthelogisticsvehicle,

reducing the logistics footprint and achieving manpower

savings.

Figure4.oEM’sM3bridgewithonerampontheeachsideversusSingapore’sM3gwithtworampsoneachside.ThethirdramponoEM’sM3bridgeisstowedinthecentreofthevehicleandisnotvisible.



MINE CLEARING

Mine clearing is an important component in a combat

engineer’s tool list. The ability to clear a minefield quickly and

safely offers significant strategic and tactical advantage to the

force.

Mine Clearing Methods

Manual Methods – Mine Prodders and Metal Detectors

The most rudimentary method of clearing mines involves

personnel using mine prodders or metal detectors. These

methods relied on human operators to detect buried mines

before they could be neutralised, typically using a bomb-

disposal operator. Due to the close proximity of operators to

potential mines, these manual methods are time-consuming

and dangerous.

Figure 5. The modular panels of the M3G as a third ramp. The modular ramps are stored as seen on the right.

Mine Rollers and Mine Ploughs

Mechanised methods of mine clearing include using mine

rollers to trigger pressure sensitive mines, or mine ploughs to

push aside surface and shallow mines. However, mine rollers

are ineffective against mines which are not pressure activated,

and mine ploughs are only effective against shallow buried

mines on relatively soft ground.

Mine Flails

The method which provides the highest assurance of clearing

mines is the use of mine flails (see Figure 6). This method

employs a flail system to impact the ground physically to

neutralise mines. Modern mine flail vehicles have incorporated

the latest technology and innovations to allow soldiers to

clear a minefield faster, with less manpower and a high safety

margin. The SAF has two types of mine flailing systems – the

Mine Clearing Vehicle MCV910 and the Trailblazer Counter

Figure 6. Mine flail system on Singapore’s Trailblazer (Ong, 2013)Reprinted with permission from Defence Media Centre (DMC)


Mine Vehicle. The former was procured off the shelf with

customisationwhilethelatterwasdevelopedjointlybyDSTA

andSingaporeTechnologiesEngineering.

Mechanised methods enable the clearing of mines with

remarkably little effort compared to manual methods. The

latest innovations in mine clearing vehicles have made the

dangeroustaskofmineclearingsafer,fasterandpossiblewith

lessmanpower.

InnovationsinModernMineFlails

ThemineflailthattheSAFusescurrentlyincludesnumerous

innovationswhichmakeitsafetooperateandmoreefficientas

comparedtooff-the-shelfmineflails.Somemajorinnovations

areasfollows.

Hydro-Mechanical Continuously Variable Transmission

Whilemineclearingoperationsareconductedat lowspeeds

oflessthan1km/h,normaldrivingspeedsareupto70km/h.

Conventional vehicle transmissions are normally unable to

performwell at both low (less than 1km/h) and high (up to

70km/h)speedranges.

Instead of using two separate transmission systems3which

wouldrequireadditionalweightandspaceontheplatform,the

Trailblazerwasequippedwithahydro-mechanicalcontinuously

variable transmission which consists of a hydraulic pump

for the low speed range and a mechanically geared torque

convertor forhighspeed rangewithinasinglehousing.This

compactandlighttransmissionsystemenabledtheTrailblazer

tobecapableofboth extremely lowandhigh speedswhile

contributingtoweightandspacesavings.

Lane Marking System

ThelanemarkingsystemspeciallydesignedfortheTrailblazer

possesses pneumatically fired rods. Compared with other

systemsthatemploypyrotechnicfiring,thepneumaticsystem

enables safer andmore accurate lanemarking. Tominimise

the workload on the operators, the lane marking system is

deployed hydraulically and can be retracted with ease for

stowagefromthecrewcabin.

Fully Automated Control

ThefullyautomatedsoftwarecontroloftheTrailblazerreduces

the operator workload and sustains the operator for longer

missions. The software control achieves auto-contouring

effect for complete mine clearing. This creates operational

flexibility for theoperator toclearundulatingcontourswhen

necessary.Coupledwiththepneumaticlanemarkingsystem,

theTrailblazer is able to clear andmark the lane for follow-

onforcesautomaticallywithsignificantly lesseffortandtime

comparedtopastsystems.usingdigitalcontrolalsoenables

mobility to be controlled with a joystick, presenting better

ergonomicsfortheoperatoroverlongmissions.

FUTURE TRENDS

Combatengineerequipmenthasbenefittedfromtechnological

advancements and innovations over the years. This section

exploresthepotentialevolutionofcombatengineerequipment.

AdvancedMaterials

The advancement of materials technology will play a

significantroleinthefutureofcombatengineeringequipment.

one promising area is the use of composites in float and

fixed bridges which will bring about lighter, stronger and

longerbridges.However,theuseofcompositesisnotwithout

disadvantages.otherthancost,theuseofcompositesposes

severaltechnicalchallenges.

The maintenance of composite material structures is more

challenging than conventional material structures used in

bridges like steel or aluminium alloys. While conventional

materialscanberepairedbywelding,therepairisconsiderably

more complex if a delamination occurs between the fibre

and matrix in a fibre composite structure. Furthermore, the

delaminationmay not be visible to the operatorwhowould

not be able to sense the need formaintenance even when

potentially severe damage has occurred. Hence, more

experienceandexpertisearerequiredtomaintaincomposite

structures. Composites are also very sensitive to flaws

sustained during the manufacturing process as compared

to metals. Any deviation from a tightly controlled process

may lead to a compromise in thematerial properties of the

composite. These factors contribute to the high cost of

incorporatingcompositesincombatengineeringsystems.

However,asmoreadvancedcompositesaredeveloped, the

useof composites couldbemore cost effective. This could

resultinatrendtowardstheuseofmorecompositesincombat

engineeringsystems.



REFERENCES

C.Teuert(personalcommunication,September1,2014)

ong, H. T. (2013, June 3). Engineers roll out. Retrieved

from http://www.mindef.gov.sg/imindef/resourcelibrary/

cyberpioneer/topics/articles/features/2013/jun13_cs.html

Quihuis,K.,Jr.(2003,February6).File:M60A1ArmoredVehicle

Landing Bridge.jpg. In Wikimedia Commons. Retrieved

october 7, 2014, from http://commons.wikimedia.org/wiki/

File:M60A1_Armored_Vehicle_Landing_Bridge.jpg

Schottelru. (2011, November 16). Schottel SPJ [Video

file]. Retrieved from http://www.youtube.com/watch?v=c_

F4uZ5BKHy

ENDNOTES

1 The Standardisation Agreement 2021 MILITARy LoAD

CLASSIFICATIoN oF BRIDgES, FERRIES, RAFTS AND

VEHICLES is a standard for NATo forces that provides

methodsforcomputationofMLCforbridges,ferries,raftsand

vehicles(bothtrackedandwheeledvehicles).

2 Thisconfiguration is ideal as it notonlyprovidesa larger

deck space for ferrying vehicles, but also increases the

metacentricheightforadditionalstability.

3 The transmission systems comprise a mechanical

transmissionforhighspeedtravellingandahydrostaticdrive

for low speedmine clearing. This is because a mechanical

transmissionisunabletoprovidehightorquesandisinefficient

at lowspeeds,whereasahydrostaticdrive suffers from low

efficiencyandexcessivelossesathighspeeds.

AutonomousEquipment

Althoughunmannedtechnologyisrelativelymature,ithasnot

beenwidely implemented incombatengineeringequipment.

onemainfactor iscost.Asunmannedtechnologiesbecome

morecosteffective,therecouldbemoredrive-by-wirecombat

engineering systems to enable the development of remote-

controlledandautonomoussystems.

Fully autonomous equipment will require sufficient artificial

intelligence(AI).AsAItechnologyimprovesinthefuture,it is

likely that systems for dull, dirty and dangerous tasks such

asmineclearingwillmakethefirstpushtofullyautonomous

systems.

ImprovisedExplosiveDeviceNeutralisation

Inthefuture,IEDswillpresentoneofthebiggestchallengesfor

combatengineers.Currenttechnologyincludesradiojammers

and ground penetrating radar systems, but these solutions

havetheirshortcomings.Futuredevelopments in theareaof

IED neutralisationwould likely include high energyweapons

suchashighpoweredmicrowavesand lasers todisrupt the

electronicsanddetonatorsinIEDs.

CONCLUSION

There have been great technological advances in combat

engineering equipment over the years. These advances,

particularlyinautomation,havebeenleveragedinthesystems

used by the SAF. The benefits include a leaner operating

force,more ergonomic systems, aswell as safer and faster

completion of combat engineering tasks. Besides utilising

technology, innovations in design have also played a role

in reducing the logistic requirements of the systems. While

technology continues to advance, it would take time for

these advancements to be adopted in combat engineering

equipment, due to the difficulties and high cost associated

withimplementationcurrently.

ACKNOWLEDGEMENTS

The authors would like to express their appreciation to

MrPatrickDRozario andMr Teo Tiat Leng for their patient

guidanceandinputsinpreparationofthisarticle.


BIOGRAPHY

PHUA Zhengqi Daryl is a Project

Manager (Land Systems) involved in

the acquisition of combat engineering

systems. He was formerly involved in

armoured vehicle projects and was a

System Manager in Headquarters,

Maintenance andEngineeringSupport in

theArmy.Daryl graduatedwith aMaster

ofEngineering (MechanicalEngineering)degreewithFirstClass

Honours from Imperial College London, uK, in 2009 under the

DSTAundergraduateScholarshipprogramme.

TANChun isaProgrammeManager(Land

Systems) overseeing the acquisition of

combat engineering systems for theArmy.

He has extensive experience in managing

projects in the areas of military bridges,

mine clearing vehicles, engineering plants

and lightseavessels.TanChungraduated

withaBachelorofEngineering(Mechanical

Engineering) degree from the National university of Singapore

(NuS) in1993.HefurtherobtainedaMasterofScience(Military

Vehicle Technology) degree from the Royal Military College of

Science,uK, in2001undertheDSTAPostgraduateScholarship

programme.

WONG Yee Yin Kimberly is an Engineer

(Land Systems) who is currently involved

in the acquisition management of combat

engineer systems. Kimberly graduated

withaBachelorofEngineering(Mechanical

Engineering)degreefromNuSin2013.



DELIVERINgNEWMINECouNTERMEASuRECAPABILITIESToTHERSN

INTRODUCTION

AnondescriptcargoshiptraversestheStraitofSingaporeand

releasesacylinderdiscreetly intothewater.Later,aRepublic

ofSingaporeNavy(RSN)MineCountermeasureVessel(MCMV)

RSSBedokisperformingaroutinesurveywhenitsminehunting

sonar1 detects a cylindrical object lying on the seabed. A

lightweightunderwaterinspectionvehicleisdeployedremotely

to investigate and visual confirmation via its fibre-optic link

showsthattheobjectisindeedaseamine.

ThisinformationissentbacktotheMaritimeSecurityTaskForce

Headquarters and patrol vessels are dispatched immediately

tocordonoff theaffectedarea.RSSBedok then launchesa

lightweightexpendableminedisposalvehiclewhichdetonates

theminesafely.

ThisisahypotheticalbutpossiblescenariothattheRSNMine

CountermeasureSquadronmayfaceintheirmissiontokeep

theStraitofSingaporesafeforshipping.

TheMCMVsplayanimportantroleinthemaritimesecurityof

Singaporebyensuringthatitssurroundingsealanesandthe

GOHYongHan,LAMSuYingAudrey

ABSTRACT

AnymineincidentintheSingaporeStraitwouldseverelyimpactSingapore’seconomy.Assuch,theRepublicofSingaporeNavy(RSN)MineCountermeasureVessels(MCMV)formthemainstayoftheSingaporeArmedForces’underwaterminecountermeasurecapability.TheBedok-classMCMVsoftheRSNwereoperationalisedin1995.In2009,amodernisationprogrammewasintroducedtoupgradetheMCMVs’minecountermeasurecapabilitiessoastoenhancetheirminehuntingandneutralisationrate.Thisarticlediscussesthethreatswhichseaminescanpose,theprojectandtechnicalchallengesindeliveringtheminecountermeasurecapabilitytotheRSN,andthesystemengineeringbasedapproachadoptedbytheprojectmanagementteaminovercomingthesechallenges.

Keywords:minecountermeasure,minehunting,minedisposal

SingaporeStrait are freeandopen to international shipping.

Thelayingofminesbypotentialadversariesorterroristsinthe

SingaporeStraitor inoursealinesofcommunicationscould

leadtoportclosure,whichwouldresultindirecttradelosses

amountingtomorethanS$2billiondaily2.

THE SEA MINE THREAT

Seamines are explosivedevicesplaced inwater todestroy

surface ships or submarines (Wikipedia, n.d.). They range

frombottommines,mooredmines,driftingorfloatingmines

to limpetminesattacheddirectly to thehullsof targets.Sea

minesare laidand left towaituntil theyare triggeredby the

approachof,orcontactwith,anenemyvessel.Theseamine

is a lethalweapon that datesback to themid-19th century.

They are low-cost weapons which are extremely difficult to

detect, identify and destroy, presenting a significant threat

eventothemostsophisticatedwarshipstoday.Figure1shows

apictureofaBritishMk14seamine.Seamineshaveevolved

over time from the early low-cost contactmines tomodern

influencemineswithmagnetic,acousticandpressuresensors.

Advanced influence mines with modern signal processing


capability can be triggered by pre-determined logic or pre-

programmed characteristics of a particular class of ship or

submarine’ssignature.

TheCommitteeforMineWarfareAssessmentoftheuSNaval

StudiesBoard(2001)describesthestrategicuseofmines in

thedenialofpassagethroughconfinedwatersandtheentryor

exitofportsofcoastalnations.Minesareasymmetricweapons

and can influence a war campaign greatly. For example, in

WorldWarII,miningbytheAlliesachievedsomeremarkable

successes. During the Atlantic War lasting five years, the

RoyalAirForce(RAF)flew20,000mine-layingsorties,sinking

638 shipswith the loss of 450 aircraft. In comparison, RAF

bombsand torpedoessank366shipsover thesameperiod

withthelossof857aircraft.DuringtheTankerWar3in1988,

theguidedmissilefrigateuSSRobertswasheavilydamaged

byadrifting Iranianmineand theuSNavyspentmore than

uS$90 million to repair the damage (see Figure 2). In the

1991gulfWar, Iraqimines hindereduS amphibious assault

planningandheavilydamaged twouSwarships,preventing

them from further operations. For inferior forces, mines are

particularlyvaluabletodefendagainstasuperiornavalforce.

Sea mines are available widely and are often more difficult

Figure1.BritishMk14SeaMine(©oxyman/File:BritishMk14SeaMine.jpg/WikimediaCommons/CCBy-SA3.0)

andtime-consumingtoneutralisethanairandmissilethreats.

SinceWorldWarII,14uSNavyshipshaveeitherbeensunkor

damagedbymines,ascomparedtoonlytwowhichhavebeen

damagedbyairormissileattacks.

DELIVERING NEW MINE COUNTERMEASURE CAPABILITIES TO THE SINGAPORE NAVY

The RSN’s four Bedok-class MCMVs were acquired from

Sweden and commissioned in 1995. In view of their ageing

systems and the advent of new technologies, DSTA

embarked on a modernisation programme for the MCMVs.

Thisprogrammecommenced in2009with the installationof

an advanced and integrated mine countermeasure combat

system,comprisingaMineInformationSystem,HullMounted

MineHuntingSonar (MHS),TowedSyntheticApertureSonar

(TSAS) and Expendable Mine Disposal System (EMDS)

(see Figure 3). The approaches taken by the DSTA project

management team (PMT) to deliver these new capabilities

successfullyaredescribedinthefollowingsections.


Figure3.Newminecountermeasuresystemsinstalled

Figure2.Theeight-metreholeinthehullofguidedmissilefrigateuSSRobertscausedbyanIranianM-08mine(©Mussi/File:Ffg58minedamage2.jpg/u.S.Navy/ID:DNST8902266/PublicDomain)


ManagingaComplexUpgradingProjectWithaLeanPMT

The conventional approach to managing a Life Extension

Programme (LEP)ofanaval vesselof suchcomplexity is to

form a core integrated PMT of more than 10 engineers to

overseemajorcombat,platformandshoresystems.ThePMT

adoptedaprimecontractorapproachtominimisethesizeof

theteamafteracarefulstudy.Theprimecontractorsupplied

themajorityofthesystems,installedandintegratedthemwith

existingsystems,andwasresponsiblefortheperformanceof

thetotalsystem.ThisallowedacorePMTofhalf thetypical

sizetomanagetheentireLEP.

ManagingDevelopmentalItemsDuringContracting

During the tender exercise in 2008, all submitted proposals

had some key systems that were still in the high risk

development phase due to the demanding technical

performance specifications of the tender. By applying the

procurement principles of competition and value formoney,

thePMTemployedcompetitivebiddingexercisesandincluded

contractual clauses to protect our interests in the event of

possible failure of the developmental systems. This ensured

thetenderreturnswouldbecosteffective,withacceptablerisk

managementmeasuresputinplacebyeachofthetenderers.

AchievingCostEffectivenessDuringContracting

ThePMThadoriginallymandatedalltendererstoengagethe

originaldesignerof theMCMVas theplatformconsultant to

oversee theplatformmodificationworksasa riskmitigation

measure. Subsequently, the PMT conducted a thorough

technicalriskassessmentandexploredengaginganalternate

platform consultant with the tenderers to achieve greater

costeffectiveness.ThePMTconducteddetailedshipsurveys

on each MCMV, reviewed the existing documentation and

drawings,anddeterminedthatminimalplatformmodifications

wererequired.Allrequiredinformationcouldalsobeobtained

throughmeasurements.By systematically going through the

risks of modification and integration, the PMT selected an

alternate platform consultant with experience in managing

MCMVplatformupgradingandachievedfurthercostsavings.

Withtheaddedriskassessmentandmanagementprocesses

putinplacecontractuallyandthroughprojectmilestonereview

meetingsandprogressivemonitoring,thisapproachledtothe

effectiveandsuccessfulexecutionoftheprogramme.

ManagingIntegrationRiskWithLegacySystems

The delivery of the upgraded programme capability was

heavilydependentonthesuccessfulintegrationoftheexisting

systemswiththenewsystems.This isamorecomplextask

comparedtonew-builtprogrammesassomeoftheinformation

required for integration is not available for some legacy

systems. To mitigate this risk, pre-condition assessments

(PCA) were performed to establish and record the baseline

configuration of the ship through a series of inspections

and tests.Thisenabled the reconstructionandextractionof

missinginformation.Atthesametime,thePCAservedtoverify

thelegacysystems’performanceandinterfacespecifications

tofacilitateintegrationwiththenewsystems.

DeliveringImprovedMineHuntingCapability

underwater mines are located using sonar which is

traditionally a slowand tediousprocess.With the advent of

new technologies, e.g. the synthetic aperture sonar (SAS),

minehuntingcanbeperformedbetterandfaster.Theprinciple

ofSASistocombinesuccessivepings4alongaknowntrack

coherently inorder to increase the resolutionof theazimuth

direction(along-track).Hansen(2011)explainedthatwiththis

increased “synthetic aperture” length, the sonar is able to

obtainhigher resolution imageswith respect toconventional

sonarprocessing.

ThecoveragerateforaTSASisaboutfivetimesfaster than

the legacy hull-mounted MHS. This is achieved due to a

highersurveyspeedandwidersonarswathwidths(seeFigure

4). Being hull-mounted, the one-sidedMHS array limits the

MCMV speed during survey,while the TSAS is a two-sided

arrayabletocovermorearea,andcanbetowedatahigher

speed to achieve amuch higher coverage rate. In addition,

theTSASprovidessignificantlyhigherresolutionforimproved

classification capability5. The new TSAS also offers an

automaticdetectionandclassificationcapability to ease the

operator’sworkloadinminedetectionandclassification.



Compared to the previous sonar system which was

hull-mountedandnottowed,thePMTconductedanextensive

safetyreviewontheproceduresprovidedbythecontractors

for the launching, recovery and towing operations. All the

emergencysafety featuresof theTSAS, suchasemergency

surfacing, cable breaking tensions and emergency stops,

were individually analysedduringdesign reviewsand tested

thoroughly during sea acceptance tests to ensure safe

operations. The launch and recovery procedures were also

improvedandsimplifiedthroughnumerousseatrials.

DeliveringImprovedMineNeutralisationCapability

Theminedisposalsystem(MDS)hasbeenusedbytheRSN

for mine neutralisation since 1995. The vehicle used in the

MDSweighsabout900kgandrequiresacraneandhandling

systemforlaunchingandrecoveryduringmineneutralisation

missions.AspartoftheMCMVmodernisationprogramme,a

newEMDSwasacquiredandinstalledonboardtheMCMVs.

TheK-Ster6EMDS is capableof identifyingandneutralising

mine-like objects to support the mine clearance operations

of the RSN. It is a remotely operated vehicle that consists

of a lightweight vehicle and supporting shipboard systems.

The vehicle has two configurations – the K-Ster Inspection

for identification of mine threats, and the K-Ster Combat

for neutralisation of mines. The expendable K-Ster Combat

vehicleisdesignedtoneutraliseaminewithasingleshot(see

Figure5).

This vehicle has led to vast improvements in mission

effectivenessasitislightweight,simpletooperateandeasyto

deploy.At50kg,itislessthan10%theweightoftheprevious

MDSvehicle,and its lighterweightsimplifiesthe launchand

recoveryprocess. It isestimated that theoperation timeper

mine is reduced by about half. Equipped with just a small

charge,thevehicleisdesignedwithatiltablewarhead,sonar,

sighting laser, video camera and searchlights to locate and

attackminesaccuratelyandefficiently.

TheK-SterCombatvehiclesarestoredintheEMDSmagazine

onboardtheMCMVs.Tominimisemanualhandlingofvehicles,

thePMTworkedcloselywiththeprimecontractortodesigna

setofcustomisedjibsandfixturestofacilitateamoreefficient

transferofK-SterCombatvehicles.

Figure4.(a)MHStransmissionpattern(b)TSAStransmissionpattern

(a) (b)


TheRSN is the first navy in theworld to conduct live-firing

using this vehicle. As this is a new weapon system, there

were no previous firing templates or references. The PMT

collaboratedwiththeRSNtodeveloptestscenariosandsafety

firing templates. Subsequently, with the knowledge gained

fromthefirstfiring,thePMTworkedoutanewweapondanger

area template which significantly reduced the safety radius

comparedtothefirstfiring.Thisachievedfurthercostsavings

intermsofassetsandtimerequiredforsafetyclearance.

Inaddition,overtheseveralseatrialsandlive-firing,thePMT

enhanced the preparation procedures progressively, and

implemented additional instrumentation to further automate

thepre-launchprocess.Theseservedtoreducethepreparation

timeneededbeforeeachfiring.

CONCLUSION

Through the application of the system engineering based

approach, the PMT had successfully completed the

MCMVmodernisation programme for the RSN in 2014 in a

cost-effectivemanner.Thishasresultedinnewandenhanced

mine countermeasure capabilities to keep Singapore’s sea

lanesmine-freeandsafe.

Figure5.K-Stervehicleapproachingatarget

REFERENCES

Bedokclassminecountermeasurevessels.(n.d.).InWikipedia.

Retrieved July 14, 2014, from http://en.wikipedia.org/wiki/

Bedok-class_mine_countermeasures_vessel

Committee for Mine Warfare Assessment, Naval Studies

Board,DivisiononEngineeringandPhysicalSciences,National

ResearchCouncil.(2001).Navalminewarfare:operationaland

technicalchallengesfornavalforces.Retrievedfromwww.nap.

edu/openbook.php?record_id=10176

Hansen,R.E.(2011).Introductiontosyntheticaperturesonar.

InN.Z.Kolev(Ed.),Sonarsystems(pp.3-27).Rijeka,Croatia:

InTech.doi:10.5772/23122

Mussi, C., PH1. (1988, May). File:Ffg58minedamage2.jpg.

InWikimediaCommons.RetrievedJuly14,2014,fromhttp://

commons.wikimedia.org/wiki/File:Ffg58minedamage2.jpg

oxyman.(2007,December1).File:BritishMk14SeaMine.jpg.

InWikimediaCommons.RetrievedJuly14,2014,fromhttp://

commons.wikimedia.org/wiki/File:British_Mk_14_Sea_Mine.jpg



ENDNOTES

1 Sonar stands for sound navigation and ranging, and is a

techniquethatusessoundpropagationtodetectobjectsonor

underthesurfaceofthewater.

2 Singapore’stotaltradein2013wasS$980billionbasedon

theMinistryofTradeandIndustryfigures,ofwhichabout80%

orS$784billionwastransportedviaseabornemeans.

3 TheTankerWarreferstotheanti-shippingcampaignsduring

the Iran-IraqWar (1980-1988). In1981, Iraq initiatedattacks

onshipstoweakenIran’swarfightingcapability,startingwith

ships carryingmilitary supplies to the groundwar front and

subsequentlyattackingmerchantshipscarryingIran’sexports.

Iranretaliatedinasimilarfashion,attackingshipsbelongingto

Iraq’stradingpartnersandtocountriesthatsupportedIraq’s

wareffort.In1987,theuSjoinedthewaruponKuwait’srequest

toofferprotectiontoKuwait’stankerfleet.WithuSwarships

patrolling thegulf, Iranstarted tosowthegulfwithanti-ship

mines. This resulted in severaluS ships beingdamagedby

Iranianmines,includingtheguidedmissilefrigateuSSSamuel

B.Roberts(FFg-58)inApril1988.

4 Thetransmissionofsoundwavesunderwateriscommonly

referredtoaspings.

5 Withthehigherresolution,betterclassificationisachieved

as there are more pixels associated with the object under

investigationtocomputeitssizeandshapemoreaccurately.

6 TheK-SterEMDSisaproductofECARobotics,France.

BIOGRAPHY

GOHYongHan is aProgrammeManager

(Naval Systems) managing the Mine

Countermeasure Vessel (MCMV) upgrade

programme. He has worked on the

Challenger-class submarine upgrade,

undertaken defence R&D at DSoNational

Laboratories and managed research and

technology projects at the Ministry of

Defencebeforeassuminghiscurrentrole.ArecipientofthePublic

Service Commission Scholarship, yong Han graduated with a

Bachelor of Engineering (Electrical Engineering)with FirstClass

HonoursfromtheNationaluniversityofSingapore(NuS)in1997.

HefurtherobtainedaMasterofEngineering(ElectricalEngineering)

degreefromNuSin1998undertheNuSResearchScholarshipas

wellasaMasterofScience(ElectricalandComputerEngineering)

degreefromtheuniversityofCalifornia,SanDiego,uSA,in2007

undertheDSTAPostgraduateScholarship.

LAM Su Ying Audrey is a Systems

Architect (DSTA Masterplanning and

Systems Architecting) currently supporting

themasterplanningandoperationsconcept

formulation with the Singapore Armed

Forces. She was previously a Senior

Engineer at Naval Systems Programme

Centre,wheresheledthedeliveryofseveral

naval systems, including naval guns on the Formidable-class

frigatesandthenewminecountermeasurecombatsystemsinthe

MCMVupgradeprogramme.Audrey graduatedwith aBachelor

ofEngineering(MechanicalEngineering)degreeandaMasterof

Science(IndustrialandSystemsEngineering)degreefromNuSin

2004and2007respectively.




EWoRKPLACE:EVoLVINgDSTA’SKNoWLEDgEMANAgEMENTJouRNEy

INTRODUCTION

Since embarking on its Knowledge Management (KM)

journey in 2003, DSTA has attainedmajormilestoneswhich

includedthecreationoftheeHabitatIntranetPortal,theDSTA

Extranet platform, as well as the Content and Document

ManagementSystem–DSTA’srecordsrepositorythatarchives

enterprisecontent,collaborationtools,andActionandIssues

ManagementSystemmeetingtools.

From2003 to 2009, thegrowthofKM inDSTAwasevident

from the usage of different KM applications and tools, to

thecontributionofdocumentsandsharingof information.A

community of Knowledge Managers, who acted as change

agentsfortheirrespectiveentities,wasalsoestablishedtohelp

proliferatetheground-upadoptionofKMintheorganisation.

ThefollowingphaseofDSTA’sKMjourneywasintransforming

its eHabitat Intranet platform from an information-centric

to a social-centric workplace. This involved implementing

enterprise social collaboration through the use of modern

web technologies and user-centric design. Enterprise

social collaboration describes the technologies and

processes used to boost collaboration in the workplace.

KOOYihLiangKevin,LIMLayHarEvon,HOWeiLingAngela,SOHYunLinJason

ABSTRACT

It has been more than a decade since DSTA embarked on its Knowledge Management (KM) journey in 2003.The next step in DSTA’s KM journey lies in the evolution of its Intranet portal from an information-centric to asocial-centric workplace.

ThisarticledescribeshowDSTAimplementedenterprisesocialcollaborationinitsIntranetportalthroughtheuseofmodernwebtechnologiesanduser-centricdesign.Thishascreatedaconduciveenvironmentto improveengagement, learning,collaborationandproductivitywithintheorganisation.

Keywords:collaboration,knowledgemanagement,sharepoint,socialcollaboration

Thisnewwaveshifted theparadigmofhowpeoplecreated,

stored and shared knowledge by leveraging Web 2.0 and

socialcomputingtools.ItwasthustimelyforDSTAtorideon

theseemergingtechnologiestoaddressexistinggapsinKM,

redefineitslandscapeandtransformthewaystaffwork.

At the same time, there were significant changes in the IT

landscape of DSTA and the Ministry of Defence (MINDEF).

These changes included network separation1, heightened

security concerns and an increased need for cross-domain

collaborationtosolvecomplexandlarge-scaleproblems.

These were the driving factors behind the transformation

of DSTA’s eHabitat Intranet platform into its current form –

eWorkplace.

eWorkplace is DSTA’s next-generation digital workplace

designed to transform how people create, organise, store,

search,useandshareknowledge.Asaresult,itwillenhance

workeffectiveness,knowledgesharingandengagementwithin

theorganisation.


EWORKPLACE STRATEGY

ThebasisoftheeWorkplacetransformationinvolvedamove

fromacontentstrategytoaconnectionstrategy(seeFigure1).

ContentstrategyfocusesoncodifyingknowledgeinDSTAinto

reusableknowledgeassetsthatarestoredintorepositoriesto

beretrievedatalatertime.Italsofocusesonthemanagement

ofenterpriserecords.Codifyingknowledgeassetsmaximises

knowledge retention and the efficient reuse of knowledge

assets.

Figure1.DSTA’sknowledgestrategyshift

Connection strategy focuses on connecting people to

collaborate and solve problems by tapping their collective

wisdom. Italso involvesconnectingpeopletocontent inthe

repositoriesthroughefficientsearchandretrieval.Connection

strategyemphasises thecreationofnewknowledgethrough

the building ofDSTA’s enterprise social network to facilitate

people-to-peopledialogueandcollaboration.

EWORKPLACE PARADIGM OF SPACES

The basic tenet of eWorkplace is thus centred on DSTA’s

paradigmofthreemainspaces:corporatespace,teamspace,

andpersonalspace(seeFigure2).

Figure2.DSTAeWorkplaceparadigmofthreespaces


Corporate space is the authoritative source of all official

enterprisecontentandan importantcommunicationchannel

for staff. In addition to information dissemination, it should

leverage social computing tools to provide a role-based,

seamlessandmultimodalworkexperiencefortheenduser.

Team space is a contextualised one-stop page for staff

to collaborate, store, access and work on shared content

together.

Personal space is a personalised page for staff to manage

theirownproductivity.Italsoservesasaone-stopchannelto

providespecific informationtostaffbasedontheiractivities,

memberships and subscriptions within the workplace. In

addition,staffusethisspaceasaplatformforsocialnetworking.

DESIGN PRINCIPLES

To ensure a simple and secure platform, eWorkplace was

designedwiththreemainprinciples:

a) One-stopSecured and Integrated Front – Staffwould

be able to find information they need quickly and easily via

aone-stop integratedfrontwhichpools together information

fromdisparatesources.

b) Person-centric – The emphasis would be on the

individualandhowheor sheworks.Staffwouldexperience

ahighlypersonalisedexperience via intuitiveuser interfaces

andreceivecontextualised informationthrough, forexample,

subscriptionsandnotifications.

c) Process-centric – As a large part of work in DSTA

revolvesaroundprocesses,therewasaneedforeWorkplace

toenablestafftocomplywithprocessesviaacontextualised

andpersonaliseduserinterface.

TRANSFORMING DSTA’S WORKPLACE USING ENTERPRISE SOCIAL COLLABORATION

Locating a subject matter expert, social commenting and

accessingnewsfeedsandenhancedsearchfeaturesaresome

of the newnorms that have changed thewaypeoplework,

shareinformationandincreaseproductivityinanorganisation.

Therewas a lackof social tools inDSTA’sprevious Intranet

platform.Theprojectteamthusevaluatedseveraltechnologies

and selected a reliable and powerful platform that could

providecollaboration,socialmedia,search,enterprisecontent

managementandbusinessintelligencefeatures.

Collaboration,LearningandSharing

Communitiesofpracticeandteamsitesallowstafftoshareand

acquirenewknowledgeandinsightseasily.WithineWorkplace,

staff can share tacit knowledge through noteboards, blogs

anddiscussionforumswithinthecorporateandteamspaces.

An examplewouldbe the eLibrary, a central repository that

consolidates and shares technical documents authored by

DSTAstaff2.Itenablesstafftoconductwork-relatedresearch

into technicaldocuments thatareeitherproducedbyastaff

through thecourseofaprojectorby independent research.

eWorkplace provides a proper underlying file structure and

classification system for the sharing of these technical

documents.

AseWorkplace isan integratedenterpriseproductivitysuite,

the user experience in performing co-authoring and review

duringcollaborationisenhancedgreatly.Withofficewebapps

and different service applications powering eWorkplace and

the enterprise productivity suite, multiple authors are now

able to review and edit different segments of a document

simultaneously.

For traditional users accustomed to checking in and out

files, eWorkplace has transformed the way they work.

Building on existing versioning conventions, eWorkplace

allows differentiation between published and draft versions

of documents. Workflows can also be put in place if there

is a need to set up a formal approval process for sensitive

documentsthatrequireapprovalbeforepublication.

Staff can also keep up to date on documents that are of

importance to them by subscribing to alerts via e-mail and

newsfeedspushedtotheirpersonalpagesandteamsites.

Wikipage isanothersocialcollaboration toolwhichcollates

authoritativeandinformativecontenttopromotestafflearning.

TheeaseofuseofaWikipageasacontentpublishingtool

encouragesuserstocreateandsharecontentwiththerestof

theorganisation.


StaffEngagement

onefundamentalshiftforeWorkplacewasfromaninformation-

centrictoasocial-centricparadigm.Managementblogswere

developedtoallowmanagementtosharetheir thoughtsand

engagestaffinaninvitingandaccessibleenvironment.Through

social networking capabilities, staff are able to comment on

theseblogpostsandcontributeopinionsandfeedback.They

canalsosubscribetotheseblogstokeepabreastofanylatest

development.

Communitydiscussion forumscanalsobeusedtogenerate

social conversations and feedback related to specific

organisation initiativesandevents.Forexample,prior to the

DSTAStaffConference, a community discussion forumwas

usedtogatherfeedbackfromstaffonpertinentengagement

issues which were subsequently addressed during the

conference.

Thisallowsmanagementtosensethepulseoftheorganisation

constantlyandreceivefeedbackdirectlyfromtheground.

PersonalProductivity

one key innovation and a cornerstone of eWorkplace was

the consolidation of disparate information into a one-stop

personalpage.Previously,usershad tonavigate todifferent

pages of interests to retrieve information. Now, they can

receiveupdatesfromthesepagesinstantlythroughnewsfeeds

andnotifications.

The new personal page feature in eWorkplace also serves

as a directory that profiles each individual staff within

the organisation. Staff can list down their expertise and

competencies,whicharecombinedwithpertinentinformation

suchassecondaryappointments.Thisenablesstafftolocate

subjectmatterexpertsandalsoprovidesaplatformforstaffto

developsocialnetworkswithintheorganisation.

The newsfeed and notifications features allow staff to stay

updatedonlatesthappeningsinaproductiveandinteractive

manner.Staffmembersareable to subscribe to siteswhich

theyareinterestedinandreceivenewsfeedsonupdatesmade

to these sites. Staff can also receive notifications regarding

eventstheyhaveregisteredfor,taggeditems,socialactivities,

upcomingmeetingsandassignedtasks.

Furthermore,eWorkplaceprovidesTwitter-likefeaturestostaff

microblogs. Inaddition toallowing imagesandvideos tobe

postedwithstatusupdatesforricherinteraction,eWorkplace

alsointroducedhashtagsandmentionstocreateanimmersive

digitalequivalentofa real-timeconversationandnetworking

session.

When staff follow a hashtag, they will be notified via their

newsfeedoncurrentconversationsofinterestforthemtojoin.

New anddirect connections can also bemade among staff

throughtheuseofmentionswithinconversations.

Tofurtheraugmentproductivity,eachstaffwasprovidedwith

apersonalrepositorytostoreprivatefiles.Thisallowsstaffto

accesstheirfilesfromanywherewithinDSTA.

CorporateinformationontheIntranetwasreorganisedintoa

newlydefinedinformationarchitecture.Policiesandguidelines

were consolidated into a one-stop site instead of disparate

corporate portals. Access to services, like transactional

applications and online forms, was also centralised. This

improved information architecture enhanced information

findabilityandoveralluserexperienceandproductivitygreatly.

InnovationandCrowdsourcing

Socialcollaborationtoolsprovidecrowdsourcingcapabilityto

complementproblem-solvingandidea-generatingprocesses.

Through the use of discussion forums, tags and ratings

features, staff can contribute ideas and opinions on various

contentthroughouttheworkplace.Thisinfluencesthevisibility

of salient information and contributes to the innovation

process.

Introduction of implicit (hashtags) and explicit (mentions)

waysforstafftonetworkandengagewithoneanotherwithin

eWorkplacealsoallowsmulti-dimensionalchannelstopromote

crowdsourcingandinnovationintheorganisation.

ReliableInformationatOne’sFingertips

ThesearchfeatureineWorkplaceusesanewrankingmodelto

determinetheitemstodisplayandtheorderinwhichresults

aredisplayed.Behindthescenes,theanalyticscomponentin

eWorkplace tracks and analyses how content is connected

continuously,howoftenanitemappearsinsearchresultsand

whichsearchresultspeopleclickedtodetermineandimprove

the relevance of search results. Previously searched results



Though thepreviously released versionwouldbe supported

until 2020, the project team recognised the importance

of harnessing advances in technology to optimise current

investments and future-proofDSTA’s IT landscape. Thenew

product version would also improve user experience and

provide better performance. Hence, the decisionwasmade

toupgradeto thenewversionandreuseexistingdeveloped

modules.

ChangeManagementandUserEngagement

With a large-scale implementation catering to more than

3,000users,userengagement throughchangemanagement

was key to the success of the project. The team was able

to solicit feedback on eWorkplace quickly and improve the

systemdesignthroughthehelpofKnowledgeManagersand

AssistantDirectors(PlanningandControl)aschangeagents.

Walk-inclinics(insteadoftheusualclassroom-basedtraining)

wereconductedaftereWorkplacewentliveandservedasan

avenueforuserstorequestforassistanceorprovidefeedback

easily.

Itwasalso important topromote theuseof thesystem ina

funandinformalway.Quizzes,logodesigncompetitions,blog

articles, feedback sessions and various other engagement

campaigns were organised throughout the first year of

implementation. This encouraged users to discover more

about eWorkplace and familiarise themselves with its new

features.

GOING FORWARD

Since its implementation,eWorkplacehasachievedsuccess

as DSTA’s next-generation information work space. This is

evident fromthecontent thathasbeenaccumulated–more

than1,000teamsites,7,000wikipages,1.4milliondocuments,

46,000photosand25,000announcements.

eWorkplace implemented a secure and robust content

infrastructureandleveragedWeb2.0technologiestoachieve

thegoalofconnectingpeopletopeopleandpeopletocontent

via person-centric workspaces. These include a one-stop

personalised space, team spaceswithworkflows for simple

approvals and proper filing of project documents, and a

reorganised corporate Intranet based on the information

architectureandtaxonomydevelopedinPhase1.

are also displayed as query suggestions at the top of the

page.Staffcanfurtherrefinetheresultsbytogglingbetween

thedifferentsortoptionssuchas relevance,date (newestor

oldest),lifetimeviewsandrecentviews.

Crawlerswere also configured to look for entities (words or

phrases)within titles, headers,metadata and contentwithin

documents. Dictionaries can be created and deployed to

searchandstoretheseentitiesasmanagedpropertiesinthe

searchindexeswhichcanbeusedtocreatesearchrefinersfor

moregranularfiltering.

Visual improvements such as displaying an application

icon next to the title of search results and previewing of

documentcontentwhenacursorisplacedonasearchresult

is “highlighted”, also improved the speed inwhich staff can

locatecontentrelevanttothem.

CHALLENGES

ShorteningofITImplementationCycle

A typical IT implementation cycle in DSTA would require at

least18to24monthstodeliverthefirstmilestone.Theteam

was faced with the challenge of delivering key eWorkplace

capabilities in less than a year. The team’s strategy was to

tap an internal pool of resources to prioritise and proceed

with the system implementation for key collaboration and

engagementcapabilities.ThisalsoallowedDSTAtobuildup

and sharpen its competencies in KM technologies quickly.

Spiral developmentwas another strategy undertakenby the

teamtoshortentheprojectdeliverytime.Insteadofthetypical

waterfall model3 to deliver all capabilities in a ‘big bang’4

approach,theteamdeliveredeWorkplaceusingacommercial

off-the-shelf software with minimal customisations. This

helped to reduce the risk of implementation caused by

evolvingbusiness requirementsandallowedrequirements to

bevalidatedthroughaprototypingapproach.

Future-proofingAgainstTechnology

In themidstof theproject, the teamhadtodecidebetween

upgradingthesoftwaretothenextproductversion,whichwould

provideenhancedenterprisesocialcollaborationfeaturesand

in-memorycapabilities,orstayingwithapreviously released

version.


Figure3.ThefutureofeWorkplace

HIGHLY-CONTEXTUAL CONTENT

INTRICATELY-NETWORKED STAFF LIMITLESS PRODUCTIVITY

PEOPLE-TO-CONTENT

PEOPLE-TO-PEOPLE

NETWORKED-PEOPLE TO

INTELLIGENT WORKSPACE

“Building a robust content infrastructure”

“Constructing staff-centric workspaces”

Corporate Intranet

“Intelligently connecting people to achieve shared goals”

Project-focused Workspace

Community-driven Workspace

Personal Space Information Governance/Architecture

Contextual Search

Extensive Repository

Enterprise Social Network Analytics

External Collaboration

Focus for the next phase will be on the infrastructure,

processes and strategies needed to build social channels

for communication and collaborationwithin andbeyond the

organisation. Enterprise social technologies combined with

dataanalysisandmobiletechnologieswillconnectpeopleto

spottrendsandleveragetheknowledgeoftheorganisationto

completetasksrapidlyandsurfacepreviouslyhiddenpockets

ofvaluableinformation(seeFigure3).

ENDNOTES

1 Network separation refers to the initiative to secure the

organisation’s work environment by means of information

segregation.

2 DSTA-authored technical documents include trip and

studyreports,conferencepapers,post-gradthesesandcase

studies.

3 The waterfall model is an approach used in software

development.Itisasequentialdesignprocessthatcomprises

thephasesof Initiation,Design,Testing, Implementationand

Maintenance.

4 Big bang adoption is a software release method that

involvesgettingridoftheexistingsystemsandtransferringall

userstothenewsystemsimultaneously.



SOHYun Lin Jason is a Senior Engineer

(EnterpriseIT)fromtheBusinessInformation

Analytics and Database Services team.

He isresponsibleforthedevelopmentand

adoptionofsentimentanalysis,datamining

andvisualisation.Hewaspreviouslypartof

theEIT-DSTASystemsteaminvolvedinthe

supportofoperationsforeHabitatandops-

SupportNet,andtheireventual transformation intoeWorkplace.

JasonobtainedaBachelorofComputing (InformationSystems)

degree,withspecialisationinServiceScience,Managementand

Engineering,fromNuSin2011.

BIOGRAPHY

KOO Yih Liang Kevin is the current ops

Manager of eWorkplace in DSTA’s Chief

Informationoffice (CIo).He is responsible

fordrivingtheadoptionofeWorkplaceand

thecontinual improvementoftheplatform.

Hewasalsopartof theEnterprise IT (EIT)

teamthattransformedthehumanresource

IT landscape for the Ministry of Defence

and the Singapore Armed Forces. Kevin obtained a Bachelor

of Science (Electrical andComputer Engineering) degree and a

Master of Engineering (operations Research and Information

Engineering) degree from Cornell university, uSA, in 2002 and

2003respectively.

LIM Lay Har Evon is Assistant Director

(Planning and Control) from Enterprise

IT Programme Centre. She led a team

to implement eWorkplace and other

Knowledge Management (KM) projects in

DSTA to deliver the nextwave ofKMand

social enterprise capabilities. She was

previously involved in business development and planning as

well as in competency development programmes to drive the

development of SAP Enterprise Resource Planning capabilities

andcompetencybuildupwithinEnterpriseITProgrammeCentre.

EvonobtainedaBachelorofScience(ComputerandInformation

Sciences)degreefromtheNationaluniversityofSingapore(NuS)

andaMasterofTechnology (SoftwareEngineering)degree from

theInstituteofSystemsSciencein1999and2004respectively.

HO Wei Ling Angela is Head Design

Thinking(EnterpriseIT).Shewaspreviously

theopsManagerofeWorkplace inDSTA’s

CIo. She co-created DSTA’s KM vision

and strategic roadmap, and oversaw

the successful roll-out of eWorkplace.

A recipient of the DSTA overseas

undergraduate Scholarship, Angela graduated with a Bachelor

of Science (Computer Science) degree with Honours, with a

focus on Human-Computer Interaction from Carnegie Mellon

university,uSA,in2004.ShefurtherobtainedaMasterofScience

inTechnologyandPolicy,withafocusonAeronauticsandHuman

Factors,fromtheMassachusettsInstituteofTechnology,uSA,in

2006.




MoDEL-DRIVENARCHITECTuREAPPRoACHFoRENTERPRISESySTEMS

INTRODUCTION

The Model-driven Architecture (MDA) is a software design

approach defined by the object Management group

(oMg). TheoMg is an international, openmembership and

non-profit computer industry standards consortium that

developsenterpriseintegrationstandardsforawiderangeof

technologiesandindustries.

Modelsarekey toMDA inasoftwaredevelopmentprocess.

TheMDAapproachusesPlatform-IndependentModels(PIM)

whichincludebusinessprocessestodefinethefunctionalities

of a system. A strong foundation and institutionalisation of

Enterprise Architecture (EA) practice within the organisation

enforcesacommonmodelling language tocaptureallPIMs.

These PIMs are then translated and linked electronically

to Platform-Specific Models recognised by computers for

execution. This process of planning, design, development

and testing is termed theApplicationLifecycleManagement

(ALM). Tools that integrate across the ALM processes help

ensureclear traceability,controlofchangesandassessment

ofchangeimpact.

LAIKokKee,NGWendy,LOWKweeBoon

ABSTRACT

TheModel-drivenArchitecture(MDA)approachhasbeenrecognisedasamethodologythatcanhelpenhanceagilityandspeed in the implementationof enterprise IT systems. This article introduces the concept ofMDAandhow it helps inmanagingthecomplexitiesofintegrationandimprovingbusiness-ITalignment.IthighlightshowDSTAusesvariousMDAtechniquestoachieveaprocess-orientedsoftware implementationparadigm,wherebusinessrequirements (captured inprocessmodels)are linkedtoactual ITsystemconfigurationstoshorten implementationcycles.ThisarticlealsoshareshighlightsandlessonslearntintheMDAjourney.

Keywords:model-drivenarchitecture,softwaredevelopment,applicationlifecyclemanagement

DEFINING MODEL-DRIVEN ARCHITECTURE AND APPLICATION LIFECYCLE MANAGEMENT

Model-drivenArchitecture

InDSTA’scontext,MDAisdefinedasanapproachtoorganise

andmanagebusinessrequirementsinPIMs,whereautomated

tools can be reused for design and implementation into

IT systems. The principle is to separate the specification of

functionalityfromthatofimplementation.

Thisapproachleveragesthestrongfoundationandmaturityof

EApractice.Since2006,EAhasbeenadoptedasameansto

strengthenbusiness-ITintegrationinDSTA.TheEAframework

was developed to ensure that business requirements

(operationalview)andITimplementation(systemandtechnical

views)arereflectedaccurately inanarchitecturalmodel (see

Figure1).

TheMDAapproachentailstwospiralsofEAdevelopment(see

Figure2).Thefirst isat thebusinessparadigmlevel (what is

modelledversuswhat isbuilt). It isabout re-engineering the


businessparadigmintoprocess-orientedmodelsaccordingto

EAstandards.Thekeybenefitofthisstandardisedapproach

isthatfinalbusinessmodelswillbesignificantlysimplerand

refined, yet functionally richer than traditionalmethods. The

secondisattheITparadigmlevel (what isbuiltversuswhat

is used). It is about enabling the re-usability of business

models for design and implementation into IT systemswith

appropriate toolsandservices.This integrationenforces the

mappingofrequirementscapturedinthebusinessmodelonto

actual implementation.Withsuchconsistency, thealignment

betweenbusinessandITisalsoenhanced.

The MDA concept also allows process implementations,

systemconfigurationsaswellastestscenariostobegenerated

automatically through business models. Through early

Figure1.EAview

Specifies

system

capabili/es

required to

sa/sfy

informa/on

exchange

System View (SV)

Relates systems to Operational views

Operational View (OV)

Identifies what needs to be accomplished and

who does it

Technical View (TV)

Prescribes standards and conventions

Figure2.ConsistencythroughanMDAapproach

Models Business Process Models

Logis1c + Finance + Human Resource

What is Modelled

What is Built

What is Used

prototyping and better communication across stakeholders,

issues and conflicts are reduced during implementation.

overall, systemdevelopmenteffort and timeare reducedas

compared to conventional systems development methods.

This enhances the pace and agility of how systems are

designed, built and tested in the application lifecycle of the

system.

ApplicationLifecycleManagement

ALM facilitates the coordination between the business

and development teams (see Figure 3). This includes the

managementofrequirements,building,testinganddeployment

sothatapplicationscanbemanagedeffectivelythroughoutthe

applicationlifecycle.


ThemainstagesofALMareasfollows:

a) Requirements –The requirements foranewapplication

are gathered and expressed as business processes, events

andactionstakenbyvariousstakeholders.

b) Design – The requirements are translated into

specificationsfortheITcomponents.Theyincludethedesign

of the application or any customisation to the standard

packagedsoftwareaswellasthedesignoftheenvironmentor

operationalmodelthattheapplicationhastorunon.

c) Build – Both the application and the operational model

aremadereadyfordeployment.Applicationcomponentsare

codedoracquired,andthenintegratedandtested.Foroff-the-

shelfsoftware,requiredcustomisationswillbedoneduringthis

phase.

d) Deploy – Both the operational model and applications

are moved from the development environment to a test

environment, and finally to the production environment.

Applica'on Lifecycle Management Phases:

Requirements Design Build & Test Deploy

Business Business Process

Expert Development

Team

Figure3.ALM

Requirements

Design

Build and Test

Deploy

Operate

Op7mise

Discover and Evaluate Business Func3on Predic3on

Prepara3on and Blueprint

Landscape Verifica3on Iden3fy Technical Usages Maintenance Op3miser

Realisa3on – Install, Ac3vate

Installa3on Switch Framework Configura3on + Customisa3on Predefined Test Content

APPLICATION LIFECYCLE

MANAGEMENT

Figure4.ALMphases

The operational model is incorporated into the existing IT

environment and the application is installed on top of the

operational model. This is also typically governed through

a release and deployment management process to ensure

properconfigurationcontrol.

e) Operate – The IT services organisation operates the

applicationaspartofthedeliveryofaservicerequiredbythe

business.Theperformanceoftheapplicationinrelationtothe

overallservice ismeasuredcontinuallyagainstservice levels

andkeybusinessdrivers.

f) Optimise – The results of the service level performance

measurements are measured, analysed and acted upon.

Possible improvements are discussed and developments

initiatedifnecessary.Thetwomainstrategiesinthisphaseare

tomaintainorimproveservicelevelsandlowercost.

The key phases and the stakeholders involved in ALM are



THE EVOLUTION OF MDA

MDA grew in sophistication with the maturity of EA, tools

andpeople.Itisusedto:harmoniseprocessestofacilitateIT

implementation, transform business processes, and drive IT

implementation.

MDAtoHarmoniseProcessesandFacilitateITImplementation

The MDA approach is largely driven by the successful

implementationoftheEnterpriseSystem(ES)intransforming

the logisticsandfinanceoperationsof theSingaporeArmed

Forces(SAF).

Initiated in 2003, the ESwas the first large-scale system in

whichbusinessprocessmodelsweredevelopedextensively

toanalyseandestablishbusinessrequirements.Theapproach

proved crucial in facilitating the analysis, harmonisation and

integration of the diverse business processes found in the

differentservicesandLinesofBusiness(LoB).Withoutthese

business models, it would have been extremely difficult to

visualise and understand the complexity of the business

operations. In all, theproject harmonisedand standardised

more than 90% of some 600 processes defined across the

Singapore Army, the Republic of Singapore Navy (RSN),

the Republic of Singapore Air Force (RSAF), Joint and the

Ministry of Defence (MINDEF). Based on these business

process blueprints, ES(Logs) was implemented in phases

startingwiththeRSNin2005,theSingaporeArmyandJoint

in2006,followedbytheRSAFin2007–allontimeandwithin

budget. The ability to reuse processes also resulted in cost

savings of some S$80million in systems implementation. It

alsotransformedthewaylogisticsandfinancialoperationsare

carriedouttoday(Lim,Ham,Heng,&Koh,2010).

Thiseffortwasextendedtothedomainsofplatforms,buildings

and infrastructure,medical logistics, IT andR&D, leading to

furtherbenefits.

MDAtoTransformBusinessProcesses

The Business Process Management (BPM) Department

under the MINDEF Chief Information office was set up in

February2008tofacilitateanddrivebusinesstransformation.

Inadditiontoworkingwithbusinessownerstoleadbusiness

transformationprojects,itisalsoresponsibleforbuildingupthe

DefenceBusinessMap1andfacilitatingenterpriseintegration.

The business transformation initiative is an enterprise-wide

effort. For it to succeed, strong commitment and support

from senior leadership is crucial. The existing IT steering

committee,chairedbyseniormanagementfromMINDEFand

theSAF,extended its termsof referenceto includebusiness

transformation initiatives. The steering committee, together

with LoB leaders, provide overall leadership in transforming

business capabilities. It reviews and endorses business

transformation proposals, and also plays the critical role of

identifyingandresolvingownershipissuesforbusinessareas

whereclearownershipislacking(Limetal.,2010).

A four-phase Integrated Methodology for Business

Transformation was also established and practised. This

approach starts off with the prioritisation and selection of

business functions for process mapping, followed by the

developmentoftargetbusinessarchitecture,andsubsequently

the development of the conceptual solution. The last phase

involvesimplementationofthesolution.

usingthisapproach,variousbusinessprocesstransformation

projects have since been successfully implemented with

enhancedcapabilitiesacrossthemanagementofareassuch

ashumanresource,buildingandinfrastructure,transportand

ammunition.

MDAtoDriveITDevelopment

Withthematurityofapplicationsinthemarket,itisnowpossible

to translate requirements frommodels intoapplications (see

Figure5).ThetrendofadoptingMDAtodriveITdevelopment

hasbeenconsistentacrossotherdefenceagenciesfromother

nationsaswell.

For commercial off-the-shelf (CoTS) products such as the

SAP2EnterpriseResourcePlanning (ERP)System, theMDA

approach leverages theALM functionofSAP,wheremodels

aresynchronisedintoaSAPapplicationlifecyclemanagement

toolforapplicationdevelopment.

Forcustomisednon-CoTSsystems,theMDAapproachutilises

BPMSuite(BPMS)totranslatebusinessprocessmodelsinto

executableapplications.



ThethreekeyareasofMDAimplementationareasfollows:

Model-driven Documentation and Development

DSTAhastakenona leadershiprole in thetechnicalareaof

EA to ensure architectural alignment and sound technology

implementation. In 2008, DSTA successfully delivered a

central repository of businessmodels called AVATAR3 using

the Architecture of Integrated Information Systems (ARIS)

platform.Thesemodelswerecreated inaccordancewithEA

modellingstandardsdefinedbyaDSTAteam.

Inearly2011,aproof-of-conceptwasconductedtoascertain

that the existing business process models captured in

AVATAR could be synchronised with SAP SolutionManager

(SAP SolMan). The synchronisation mechanism was tested

to work both ways. First, business requirements defined

as process models in AVATAR should be transferred and

translated automatically to SAP systems so that developers

canbeginconfiguration.Second,SAPreferencemodelsand

implementationinSAPSolManshouldalsobeimportedback

intoARIStoolstojump-startorupdatethebuild-upofunique

businessrequirementsinAVATAR.Thiswouldhelptoreduce

thedevelopmentleadtimeandallowchangestobemanaged

properly.

Figure5.MDAtodriveITdevelopment

Model-driven Testing

Extensivetestingisrequiredforeverypieceofsoftwareduring

the various phases of software development. The scope of

testing usually involves conducting some form of change

impact analysis which requires a good knowledge of the

business,softwarearchitectureanddesigndetails.

Model-driven testing is a new and promising approach for

software testing as it reduces effort and turn-around time

significantly. It enables efficient test scope planning by

accurately identifying the affected business models due to

changes introduced. This is made possible because of the

betterbusiness-ITalignmentoftheMDAapproach.

Withbusinessprocessesdocumentedintheformofbusiness

processmodels,softwaredevelopmentteamscannowmake

useofthesemodelstocreatetestcasesbasedonstructured

scenarios.Eachtestscriptproposedwillcorrespondtoagiven

scenario,thusenablingeasytrackingandverification.Thisnew

methodology and toolset was pioneered in the Centralised

Corporate Services (CCS) BPMS project and has proven to

reducethetimetorolloutnewapplications,whilemaintaining

highstandardsinsoftwarequality.

Business Model

Synchronisation and generation of system

configuration specifications

Model-‐Driven Development

Generation of business Process documents

Model-‐Driven Documenta3on

Generation of business scenarios for testing

Model-‐Driven Tes3ng

Analysis of business process performance

Process Intelligence and Performance Monitoring


Process Intelligence

CCS comprises a suite of services in the areas of human

resource, corporate finance and budgeting, logistics and

procurement and IT support. The aim of CCS is to bring

together common corporate functions performed across

entities to create synergies and leverage the competencies

of respective professional agencies for greater efficiency.

Theintentisalsotoautomateasignificantnumberofmanual

processessuchasassetstocktakingandcondemnation.

To help CCS overcome the need to automate manual

processes, DSTA piloted a BPMS tool that enables timely

andaccuratemonitoringandanalysisofbusinessprocesses,

thus enhancing efforts in streamlining, tuning and exception

handling. The BPMS platform also serves as a common

workflow tool to integrate existing disparate processes and

reducemanualprocesshand-offs.

usingtheMDAapproach,theBPMStoolwasabletoaccept

existing business processes already mapped out under the

EA framework and translate them into executable business

workflows with minimum development effort. This helped

to reduce the implementation lead time for ITsystemswhile

meetingbusinessrequirements.

CHALLENGES

Although the MDA approach is still at its infancy, several

challengeshavebeenencounteredsofar.

PracticalityofaSingleModellingStandard

TheMDAapproachwaspiloted inthe implementationofthe

EnterpriseSystemandCCSBPMS.Forbothplatforms,there

were significant efforts to ensure that the models captured

in AVATAR were usable by SAP and BPMS for subsequent

application development. Through the pilot tests, it was

concludedthatasingleprescribedEAstandardandmodelling

toolmaynotbeable tomeetbothSAPandBPMSplatform

specific requirements. Hence, a hybrid model will have to

be developed to better meet the needs of both CoTS and

bespokedevelopments.

IndustryCompetencyandReadiness

When the MDA approach is fully operationalised, all

IT implementations will need to comply with the MDA

requirements. This means that System Integrators (SI)

undertakinganyprojectimplementationwillneedtohavethe

necessarycompetencyinMDA.AstheadoptionoftheMDA

approachisstillnewintheITindustry,thismayposeariskto

the projects’ timeline and cost.DSTAhas taken a proactive

approachtoaddressthisbyprovidingthenecessarytraining

and guidance on the MDA approach to SIs engaged in IT

projects.ThecompetenciestotakeontheMDAapproachis

alsoexpectedtomatureasmoresoftwarefirmsandSIsadopt

theMDAapproachacrosstheITindustry.

MOVING FORWARD

ThefirstphaseadoptionofMDAhasmetits intendedgoals.

The next phase will be to proliferate the practice across

CorporateIT(CIT)systemsprogressively.Todothis,thecurrent

MDAapproachwillbereviewedinordertobemoreeffective

andefficientincateringtodifferenttypesofITsystems.

The focus will also be centred on further enhancing the

integrationofAVATARwithSAPERPapplications.Thiscould

be achieved through further automation of themanagement

oftheapplicationlifecycle,aswellastheenhancementtothe

qualityofexistingmodelscapturedwithinAVATAR.

Beyond the objectives of achieving business agility and

shorteningthedurationfromapplicationdevelopmenttoroll-

out,theMDAapproachalsoensuresbusinesscontinuityinthe

eventthatthereisaneedtore-platformtheSAPERPsystem.

CONCLUSION

The concept of using the MDA approach to automate the

integration of business requirements captured in business

processmodels into actual application development is both

desirableandexciting forbusinessusersand IT teams.The

MDA approach has demonstrated the ability to reduce the

duration from application development to roll-out, for IT

systems through the use of models that drive application

development.



REFERENCES

Lim,H.C., Ham,y. F., Heng,C.C. L., &Koh,C. y. (2010).

Driving Business Transformation through a Process-centric

Approach.DSTAHorizons,28-43.

objectManagementgroup.(n.d.).objectmanagementgroup

terms and acronyms. Retrieved from http://www.omg.org/

gettingstarted/save_terms_and_acronyms.htm.

ENDNOTES

1 The Defence Business Map illustrates the CIT lines of

business (such as finance management, human resource

management)andrespectivebusinesscapabilitiestosupport

the enterprise’s strategic outcomes and objectives. The

DefenceBusinessMapwillbefurthercolourcodedtoindicate

theIT-enablementstatusofeachbusinesscapabilitiesandfor

planningoffutureITenablement.

2 Systems Application and Product in Data processing,

or SAP, is an enterprise resource planning software which

comprises a number of fully integrated modules covering

virtuallyeveryaspectofbusinessmanagement.

3 AVATAR stands for “The Actionable, CollaboratiVe and

AlignedEnTerpriseArchitectureRepository”.

BIOGRAPHY

LAI Kok Kee is a System Manager

(Enterprise IT) who currently oversees the

operations and support management of

enterprise SAP systems. Kok Kee is also

a SAP Certified Technology Consultant.

He graduated with a Bachelor of Science

(Information Systems) degree from the

ThamesValleyuniversityofLondon,uK,in

2003.

NG Wendy is a Principal Engineer

(Enterprise IT) who is currently involved

in realising the Model-driven Architecture

infrastructure through the implementation

of SAP Application Lifecycle Management

(ALM)fortheMinistryofDefence(MINDEF)

Enterprise Systems(Logs). She was

previously involved in the development

of Corporate IT’s Enterprise Architecture for MINDEF and

the Singapore Armed Forces (SAF). Wendy is a certified SAP

consultant inSAPapplications and a TogAF (Theopengroup

Architecture Framework) Certified Enterprise Architecture

Practitioner. She graduated with a Bachelor of Science in

(ComputerandInformationSciences)degreewithMeritfromthe

NationaluniversityofSingapore(NuS)in1998.

LOW Kwee Boon is a Senior Systems

Architect(EnterpriseIT).SincejoiningDSTA

in2000,hehasbeeninvolvedinvariousSAP

implementation projects for MINDEF and

the SAF in areas ranging from functional,

technical tocross-applicationmodules.He

is well versed in the latest development

andevolutionoftechnologiesintheareaof

EnterpriseResourcePlanning.Hehaschartedoutstrategiesand

implementation roadmaps in areas includinggovernance,Risk

andComplianceaswellasALMforMINDEFandtheSAF.Kwee

Boon isalsoacertifiedSAPconsultant inSAPapplications.He

graduatedwithaBachelorofEngineering(MechanicalEngineering)

degreewithFirstClassHonoursfromNuSin1997.




DATAANALyTICSFoRoPTIMISINgCyBERANDDATACENTREoPERATIoNS

INTRODUCTION

Data analytics is about deriving insights fromdata. Through

mathematical, statistical and machine learning methods,

patternscanbediscovered topresentbusinessesandother

operationswithamorerelatableviewoftheirdatafordecision

making.

DSTA implements data analytics for theMinistry of Defence

(MINDEF)inanumberofareas.Intheairandmaritimedomains,

analyticsisusedtoaugmentsituationalawarenesswithreal-

time interpretation of movement patterns and detection of

anomalous activity. Analytics is also applied in a variety of

Enterprise IT areas such as finance, procurement, logistics

and human resource. The rich data sets accumulated over

extendedperiodsofoperationareanalysedtoprovidesupport

inareaslikebudgetoptimisation,frauddetection,supplyrisk

managementandstaffengagement.

CHANGXuquanStanley,SIMSzeLiang,WONGMingQian

ABSTRACT

Awealthofdata isgeneratedby theMinistryofDefence’s ITnetworkswhichcanbeanalysed to improvecyber threatdetectionanddatacentreoperations.

Incyberdefence,detectionalgorithmshaveadvancedfromstaticrulestomachine-learningalgorithmsthatcanharnesstherichamountofnetworkdataavailabletodetectlowsignatureanomaliesintheenvironment.Similarly,statisticalanalysisoninfrastructurelogscanderivepatternsinsystemutilisationanduserbehaviourtogaininsightsintoincreasinglycomplexoperatingenvironments,pre-emptincidentsandoptimiseresourceallocation.

ThisarticleshareshowDSTAappliesdataanalytics toenhanceefficiencyandeffectiveness incyberdefenceanddatacentreoperations.

Keywords:dataanalytics,cyberdefence,datacentre,IToperationsanalytics

This article describes how DSTA uses data analytics in the

appliedareasofcyberdefenceanddatacentreoperations.

DATA ANALYTICS IN CYBER AND DATA CENTRE OPERATIONS

Cyber defence and data centre operations are becoming

increasingly challenging to manage due to: (a) increasing

criticality of IT; (b) growth in complexity and number of

systems;and(c)increasinglysophisticatedcyberadversaries

andthreats.

Traditionalincidentandeventmanagementtoolshaveserved

adequately in the detection, notification and reporting of

events.However,theyneedtobecalibratedmanuallytodetect

anomaliesandtoanalysecorrelatedeventsandtrends.


Dataanalyticscanbeimplementedtoautomatethederivation

ofsuchinsights.Itscapabilitiesincludeadvancedsearchand

indexingtechniquesthatalloweffectivecorrelationofevents

andanalysisofstatisticalpatterns.Machinelearningmethods

are also applied to discover topological relationships and

establishbehaviouralbaselines.

Assuch,theexistingwealthofdatacanbemined.Thedata

made available for analytics are: (a) machine data such as

utilisation and activity logs from servers and networked

devices;(b)networkdata;and(c)syntheticdata,whichisdata

generatedbyprobingthesystemwithsimulatedtestcases.

The application of these data in cyber defence and data

centreoperationscanbecategorisedinto:anomalydetection,

discoveryofhiddenpatternsandinsights,andoptimisationof

resources.

AnomalyDetection

Statistical andmachine learningmethods are used to trend

and forecast system utilisation and behavioural patterns

continuously,andbuildbaselinesthatalsoenablethedetection

oflow-signatureanomalieswithhighfidelity.

Bycorrelatingsuchanomaliesacrosssystemconfigurations,

compliancechecksandsecurityevents,preventivemeasures

can be taken to avert the potential onset of performance

degradationorhalttheprogressofcyberattacks.Theusage

of analytics thus reduces the reliance on tacit knowledge

and individual competencies to identify risks, and construct

an action planwhichwould be unsustainable in the face of

growingsystemcomplexitiesandmanpowerconstraints.

Systemutilisationandwebaccesspatternsarealsocorrelated

tounderstand,andsubsequentlyanticipatetheimpactofuser

activitieson theperformanceofapplications.These insights

allow data centre operations to prepare for planned user

activities effectively or determine the possible causes of an

unplannedsurgeinutilisation.

Anomalies can also be indicative of new exploits or freshly

compromised assets in the networks that warrant further

investigationby incident response teams.Examplesof such

anomalouseventsincludereconnaissanceactivitiesbyexternal

entitieswhoareattemptingtogaininsightsintothenetwork,

and computers infected with viruses that are attempting to

performunauthorisedactions.

Machine learning algorithms, both supervised and

unsupervised, are also being applied to pick up events that

exhibit similar behaviours from past anomalous events or

deviations. Supervised learning algorithms are used to pick

updomainnameresolution requests tosuspiciousdomains.

unsupervisedlearningalgorithmssuchask-meansclustering

areusedtocategorisenetworkandmachinedatainto“normal”

and“anomalous”clusters.Thesemachinelearningalgorithms

are commonly used in the cyber domain, especially in the

detectionofnewthreatsasthere isnoknowninformationof

thethreatthatcanbeusedtoidentifyitwithcertainty.

Alternatively,knownabnormalbehaviourscanalsobeusedin

analyticsto facilitatethecategorisationofobservedpatterns

into“normal”and“anomalous”groupsbasedonthepatterns’

similarities toknownattributesofabnormalbehaviours.This

approachcanbeusedtosupplementanomalydetectioninthe

initialphaseofdefiningthebaselineoftheenvironment,where

either insufficienttimehas lapsedtobuildareliablebaseline

orwhereitisundesirabletoassumethatlearnedbehaviouris

normalbydefault.

DiscoveryofHiddenPatternsandInsights

Advanced analytics algorithms are used for indexing,

searchingandcorrelating largedatasets todiscoverhidden

patterns, relationshipsand insights thatarenormallydifficult

forahumantoperceive.

one such example is the analysis of time intervals between

Internetrequestsoriginatingfrommachineswithinthenetworks.

Malwareoftenneedstocontactitscommandandcontrol(C2)

serverson the Internet to receive further instructions.These

requests usually occur in very regular intervals as they are

controlledbyaprogramme.Thisregularpatternisillustratedin

Figure1.Bycomparison,human-initiatedwebsurfingrequests

arerandomwithirregularintervalsbetweenrequestsasshown

in Figure 2. The standard deviation and entropy between

the differences in the time intervals are calculated, and low

standarddeviationsandentropyvaluesgive indications that

theserequestsareoccurringinaperiodicmanner.usingthis

analysis,machinesinfectedwithmalwarethatwerecontacting

C2serversperiodicallyhavebeendetectedinthepast.


Analytics algorithms are also used to identify cyber attack

campaignsbyanalysingtheirintrusionindicators,suchasthe

sourceoftheattacksandmalwareused,againstanintrusion

attributes database. Hutchins, Cloppert, and Amin (2011)

defined a kill chainmodel1 to describe the intrusionphases

of a cyber attack and identify indicators that link individual

attacks to a campaign. using information derived from the

analysis,cyberdefenceteamscanalsodeterminethetactics,

techniques and procedures (TTP) of the attackers, allowing

themtostayaheadoftheattacks.

Fordatacentreoperations,advancedsearchingandindexing

capabilities are used to perform root cause analysis for

incidents.Thisisacriticalyettime-consumingactivityduring

theresolutionofincidents.Acommonproblemisthedifficulty

indeterminingtherootcauseeventamongaclusterofevents

thatfollowafterit.

For example, the failure of a common email gatewaywould

eventually result in transactional failures being reported by

most applications that serve user workflows. That would

generateonealert foreachapplicationservice foreachtime

itattempts tosendanemail.Amongallof thosealerts,only

onealertwouldhavebeengeneratedfortherootcause.Data

Figure1.Regularnetworktrafficpatternsofmalwarebeaconing

Figure2.Irregularnetworktrafficpatternofwebsurfingbehaviour

analysiscouldbeusedtoestablishsimilaritiesintime,volume

and textual patterns between alerts, thus removing large

amountsofnoiseandmaking iteasierto identifyanomalous

events that are unique to the time period leading up to the

incident.

Analytics is also being used to discover new relationships

between events. For example, an application error and a

shortageof disk spacemaybe considered as two separate

events that are resolved independently. yet, the shortage of

diskspacemayhavebeencausedbyapplicationerrorswhich

generatelargeamountsoflogswithinashortperiodoftime.

Analytics is being applied to discover such correlations and

causalrelationships,improvingbothassessmentandreaction

toevents.

OptimisationofResources

Analytics has become more pertinent in optimising data

centre resources in recent years. This can be attributed to

the increasing use of virtualisation technologies that enable

resourcesharingandlivemigrationofworkload.


Within the data centre, a requirement placed on server

virtualisationandstorageplatformsistoprovidetheanalytics

andautomationrequiredtoredistributeworkloaddynamically

for optimal system resource usage. This also enables the

overall distribution of virtualised application servers to

adapt constantly to changes in utilisation patterns and find

placementonphysicalserversthatcanbestservetheirneeds.

Fordatastorage,analyticsidentifiesfrequentlyaccesseddata

continuouslyforplacementonstorageresourceswhereitcan

beservedwiththebestthroughput.

Analytics also supports “right-sizing” by utilising historical

and projected patterns to anticipate the correct quantities

of resource allocation, enable reclamation and redistribution

of resources to copewith short-term surges, andmoderate

changes in demand by optimising resources within existing

capacity.

Beyondshort-termoptimisations,predictiveanalytics isalso

used tomodel the impact that projected requirementsmay

haveonexistingcapacityandanticipategrowthrequirements.

Duringthisanalysis,availabilityandperformancerequirements

are considered alongside anticipated changes in resource

demands as a result of the utilisation and growth patterns

of existing deployments. With the inclusion of planned

deploymentswithinthecapacityplanningmodel,anyresulting

shortageincapacityandtimeframecanguidethepreciseand

timelyacquisitionsofadditionalcapacity.

CHALLENGES

The following points present challenges that need to be

addressedinordertorealisethepotentialofanalyticsforcyber

defenceanddatacentreoperations.

DerivingRelevantInsights

Before analytics can be implemented, a problem statement

has to be defined to determine the appropriate data and

methods to use. It is usually not straightforward to frame a

problem statement for analysis. For example, questions

like “how to reduce operating costs” or “which servers are

infectedwithmalware”arenaturalquestionsthatarisebutare

toogeneric to initiate immediateanalysis.Apart fromhaving

familiaritywiththecontextofthequestion,framingaproblem

statementrequiresstafftobetechnicallyproficientindiverse

domains such as IT infrastructure, cybersecurity, statistics

and mathematics in order to identify suitable metrics and

methodsthatcanderivetherequiredinsights.Acquiringsuch

competenciesrequiresanorganisationalcommitmentoftime

andresources.

Inpractice,whereadesiredinsightisframedatalevelthatis

tooobscureto initiateanalysis,a thoughtprocesstoreduce

theoriginalquestion recursively intosmaller intrinsicqueries

helpstomaketherelevantdataandapproachmoreapparent.

ReliabilityofAnalyticResults

The reliability of both descriptive and predictive analysis

remainsfundamentaltotheeffectiveuseofanalytics.Indata

centreoperationswheretheproblemismoredefined,numerous

analyticaltoolsareavailableinthemarketwhichprovidenon-

traditionalmeans to collect relevant data and analyse them

forcommonly required insights.Thisnecessitatesevaluation

of the results produced by these products aside from their

technicalspecificationsorcapability.However, thechallenge

liesinvalidatingtheaccuracyoftheseresults.

In practice, the validation of descriptive analytics results

is straightforward as in the case of incident resolution and

avoidance, as erroneous analysis is usually obvious and

remediable on hindsight. However, in the case of predictive

analyticssuchasresourceoptimisationorcapacityplanning,

analytics recommendations are used to serve as inputs to

thetraditionalplanningprocessforaperiodoftimebefore it

maybedeemedsufficiently reliable to replacethe traditional

processitself.

CONCLUSION

Thisarticlehashighlightedsomewayswhichdataanalyticsis

used tooptimisecyberdefenceanddatacentreoperations.

Dataanalyticsisasignificantgame-changerthatallowsmore

effective application of insights. From the strengthening of

securityposturetoenhancementofuserexperience,atangible

impacthasbeenmadeonMINDEF’snetworks.

Dataanalyticshasalsoresultedinlessdowntimeduetobetter

predictivecapabilities,while improved insightshaveenabled

swifterandmoreeffectiveactionsagainstcyber threatsand

serviceoutages.

In addition, data analytics augments the optimisation of

resourceallocation indatacentreoperations, enabling rapid

applicationdeliveryinamorecost–effectiveapproach.



Tostayaheadofsophisticatedcyberattacks,dataanalytics

has become an important tool to pick up hidden indicators

of these threats. As hackers come up with new TTPs, new

measureshavetobedevelopedandthisismadepossibleby

thevaluableinsightsgainedfromanalytics.

Together, a more secure yet effective IT experience can be

deliveredtoMINDEFwhileincurringlesseffortandpotentially

loweroperatingcoststhroughtheuseofdataanalytics.

REFERENCES

Dua,S.,&Du,x.(2014).Dataminingandmachinelearningin

cybersecurity.CRCPress:BocaRaton,FL.

Hutchins, E. M., Cloppert, M. J., & Amin, R. M. (2011).

Intelligence-driven computer network defense informed by

analysisofadversarycampaignsand intrusionkill chains. In

J.,Ryan(Ed.),Leadingissuesininformationwarfare&security

research (pp. 78 – 104). Reading, uK: Academic Publishing

InternationalLtd.

Jacobs,J.,&Rudis,B. (2014).Data-drivensecurity:analysis,

visualizationanddashboards.JohnWiley&Sons:Indianapolis,

IN.

Münz, g., Li, S., & Carle, g. (2007, September). Traffic

anomalydetectionusingk-meansclustering.Proceedingsof

Leistungs-, Zuverlässigkeits- und Verlässlichkeitsbewertung

von Kommunikationsnetzen und Verteilten Systemen, 4. gI/

ITg-WorkshopMMBnet2007,Hamburg,germany.

Singhal, A. (2007). Data warehousing and data mining

techniquesforcybersecurity.Springer:Newyork.

ENDNOTES

1 Akillchainisasystematicsequenceofeventsperformed

by an adversary to select and engage a target to achieve

a desired effect. In describing a cyber attack, Hutchin,

CloppertandAmindefinedthekillchainassevenphasesof

activity comprising: (a) reconnaissance to select the target;

(b) weaponisation of the attack payload; (c) delivery of the

payload;(d)exploitationofthetarget;(e)installationoftrojanor

backdoorintothetarget;(f)establishingcommandandcontrol

channel tocontrol thetarget;and (g)executionofactionsto

achieveobjectives.

BIOGRAPHY

CHANG Xuquan Stanley is a Manager

(Cybersecurity)workingonthedevelopment

and application of analytics for cyber

threat detection. Stanley graduated with

a Bachelor of Engineering (Computer

Engineering) degree with First Class

Honours from the National university

of Singapore (NuS) in 2006. He further

obtainedaMasterofScience(DefenceTechnologyandSystems)

degreefromTemasekDefenceSystemsInstitute in2010aswell

as a Master of Science (Computer Science) degree from the

NavalPostgraduateSchool,uSA,in2011.

SIM Sze Liang is a Manager (InfoComm

Infrastructure) overseeing data centre

virtualisation and private cloud computing

initiatives for the Ministry of Defence’s

(MINDEF)Corporate IT (CIT) infrastructure.

Sze Liang has also been actively involved

inthemodernisationofITinfrastructureand

leading implementations for infrastructure

consolidation and system automation. In

addition,heisexploringtheuseofanalyticstoaugmentincident

managementandcapacityplanning forvirtualised infrastructure.

Sze Liang graduatedwith aBachelor of Engineering (Computer

Engineering)degreewithHonoursfromNuSin2009.

WONG Ming Qian is a Senior Engineer

(InfoComm Infrastructure) managing

and implementing analytics capabilities

across MINDEF’s CIT infrastructure. He

is also involved in the enhancement of

existing analytics dashboards to improve

functionalityandusabilityofanalyticaltools

and systems.MingQian graduatedwith a

BachelorofComputing(ComputingScience)degreefromNuSin

2006.




CHALLENgESANDDESIgNCoNSIDERATIoNSFoRRADARoPERATIoNINLoCALLITToRAL

INTRODUCTION

The Naval Doctrine of the uS Navy (2010) defines a littoral

regionastheportionof theworld’s landmassesadjacentto

theoceanswithindirectcontrolofandvulnerabletothestriking

power of sea-based forces. Singapore, being one of the

busiestportsintheworld,issurroundedbybusyandnarrow

water passages, where large numbers of vessels of varying

sizes pass through. This results in a very complex littoral

environmentfor localradaroperations,posingamultitudeof

uniquechallenges for radarsystemsto track targetsquickly,

accurately and reliably. Hence, it is important that these

challengesare identifiedand tackled throughupfrontdesign

considerations,iterativesystemtestingandoptimisation.

Thisarticlepresentsthechallengesfacedandtheconsiderations

involvedindesigningaradarforoperationinthelocallittoral

environment.Thefirstsectionofthearticledescribesthekey

characteristicsof the local littoral environmentandhow it is

differentfromanopensea.Thisisfollowedbyanoverviewof

actualobservationsfromlocalradartrialsanddemonstrations.

LOManLing,LOKEMunKwong

ABSTRACT

Due to Singapore’s geographical locality, its naval and airborne maritime surveillance radars operate in a uniquelittoral environment that poses a plethora of challenges to their system design. This article briefly describes thesechallenges, and focuses on relating experiences gained from operationalising radars that perform well in thelocalenvironment.Theseexperiencesincludethestringentprocessoffocusingonthesystemarchitecturedesignduringthe front-end definition phase, simulation and testingwith environmental data during the development phase, and theeventual fine-tuning andoptimisationphase through trials.With advancements in technology, this article alsoprovidesanoverviewofpromisingadvancedradardevelopmentsandtheexpectedbenefitsinthefuture.

Keywords:radar,littoral,surveillance

Next, thearticlepresentsbestpracticesof radardesignand

testingdevelopedfromtheseexperiencestomakethesystems

morerobustforlittoralsurveillance.Finally,itconcludeswitha

glimpseofpromisingtechnologieswhichhavethepotentialto

improveradarperformanceinthelocalenvironment.

OPERATIONAL ENVIRONMENT

Whendesigningandevaluatingasensorsystem,athorough

understanding of the chief design drivers – mission profile,

area of operation and targets of interest – is essential. This

is especially critical in a complex littoral environmentwhere

thereisalargevarietyanddensityoftargetsunderanomalous

propagation effects, multipath and radio frequency (RF)

interferences.Figure1showsasatelliteimageofthecongested

Singaporeharbour.


UrbanCoastlineandNarrowPassageways

In the open sea, target returns are large compared to

background sea and weather clutter. As such, sufficient

target strength for detection can be accomplished easily to

obtainagoodsurveillancepicture.on thecontrary, a target

hastocompetewithlandclutterandmanyothertargetsina

littoralenvironment.According to theuSEnergy Information

Administration (2014), the Strait of Malacca is one of the

world’smostsignificanttrafficchokepoints,withthePhillips

Channelnarrowingdownto1.7mileswideclosetothesouth

of Singapore. This is exacerbated by coastlines lined with

buildings and man-made structures which typically have

strong radar reflections. In addition, the presence of targets

at close proximity decreases the amount of reaction time

available. This implies a heavier demandon the radar tobe

reliableintargetdetectionandextraction.

DiversityofTargets

Due to the proximity to land, radars operating in a littoral

environmentalsoneedtocopewithagreatervarietyoftargets

whichcanbeairborne,surfaceorpop-uptargetsfromnearby

landareas.Examplesincludesmallfastcraft,helicopters,low

flyingunmannedaerialvehiclesandsubmarineperiscopes,all

ofwhichpossessverydisparatekinematicsandphysicaltraits

and are used for different missions. Figure 2 illustrates this

diversitybasedon typicalRadarCrossSection (RCS)values

andvelocitiesatmicrowavefrequencies(Skolnik,2002).

LocalPropagationConditions

Chia, Khan and Chou (1988) highlighted that the equatorial

locationofSingapore results inanabsenceofstrongstorms

and typhoons.Thewindspeeds inandoutofSingaporeare

atalowaverageofabout10knots,leadingtocalmconditions

in the surroundingwaters. The low sea states translate into

reflectiveseasurfaces,whichcouldresultinmultipatheffects.

Another propagation effect affecting radar performance

is ducting. Although young, Loke, Shui, and Chen (2010)

found thatducting is not unique to the local landscape, this

phenomenon, ifnotproperlytreated,maybeexacerbatedby

strongurbanclutterbeyondtheradar’sinstrumentedrangein

alittoralenvironment.

Figure1.Singaporeharbouron27June2013(LC81250592013178LgN01courtesyoftheu.S.geologicalSurvey)


Interference

Compared to the open sea, a radar system operating in a

littoral environment iswithin the range of interferences from

shore-basedemitters.Figure3showsa frequencyallocation

chart that depicts a crowdedemission spectrumdue to the

proliferation of commercial communication networks for

aeronautical,landmobile,meteorologicalandsatelliteservices.

POSSIBLE OCCURRENCES

This section will provide insights into some possible

observations due to effects of littoral environment on

surveillance radar systems, the design optimisation required

andtheproposedbestpracticesthathavebeenadoptedfor

front-enddesigndefinitions.

FalseTracks

onemainchallengeofalittoralradarsystemistomaintaina

largedatabaseoftrackswhilereportingataverylowfalsetrack

rate.Forautomatictrackinitiation,averyfrequentoccurrence

is theformationof falsetracksonunwantedtargetssuchas

Figure2.TargetsandtheirtypicalRCSvaluesandvelocities

oil rigs and buoys that are swarming the already saturated

surveillancepicture.Theseeffectsareseentobemoresevere

for areas near urban coastlineswith strong reflective points

suchasbuildings.

To overcome these effects, both plot and track formation

processes have to be examined. Figure 4 is a generic

representation of a track-while-scan (TWS) radar’s tracking

flowwhereeachstepcouldbeafactorcontributingtothefalse

trackperformanceof the system.A systemcanerroneously

initiate a trackwhen there ispoorqualityof inputplots and

a lackofstringentcoherencycheckspriortoassociationsof

plotstotracks.Thisisbecausethequalityofatrackisbased

onthequalityofthetargetkinematicsmeasurementandplot

formation process which includes detection, validation and

unfolding.False trackscanalsobe formedwhen there isan

impropertreatmentofdetectionsofmajorscatterersacrossan

extendedtarget.

Shortsystem latency isoftendesired for fast trackupdates.

However,areductioninsystemlatencyalsotradesoffwaiting

timeessential forcorrectplot-trackassociation,especially in

an environmentwith high target density in adjacent azimuth

sectors.


Figure3.Spectrumallocationchart(InfocommDevelopmentAuthority,2014)

Typically, radar systems possess track elevation accuracies

intheorderofmilli-radians.However,whentargetsareflying

at lowelevationsof lessthanonebeamwidth,fluctuationsin

elevation accuracies have been observed to be as large as

twice the actual flight level. In addition, the signal to noise

ratio can be so low that there are no target detections in

the multipath nulls. Such erratic elevation measurements

decreasetheoperator’sconfidenceintargetidentificationand

engagement.Figure5showsanexampleof thepropagation

losses due to multipath effects, with increasing values of

attenuationfrombluetored.

DegradedTargetCharacteristics

Inadditiontodetectionandtracking,surveillanceradarsystems

often record the targets’ kinematics and RF characteristics.

An example is the use of the Doppler spread spectrum of

helicopterhubsandblades.TheseRFreturnsaretypicallyvery

weak,withatleast20dBlesssignalstrengthcomparedtothe

TargetMaskingandTrackLoss

It is common tohave large surface vessels in the vicinity of

one another in a littoral environment, possibly with smaller

targetsweavingamongthem.Whenasmallboatapproachesa

largersurfacetarget,thesmallertargetismaskedbythelarger

targetanditstrackdrops.Asradarsystemsarethe‘eyes’of

surveillance ships, such track loss events could place itself

orothers inperilous situations. Ingeneral, highRCS targets

can easily cause a saturation of the radar, masking targets

over an extensive range. This is also an indication of high

time sidelobes1, insufficient dynamic range and poor clutter

rejectiontechniques.

FluctuationsandDegradationsinTrackAccuracies

Propagationlossesandfluctuationsinelevationmeasurements

are the two major effects arising from multipath effects.



mainbodyRCS.Spread spectrumdetectionbecomesmore

challengingwhen thehelicopter is hoveringover landmass.

Additionally,ifthepresenceoftheseDopplermodulationsisa

prerequisiteforexternaltrackreporting,theremightnotbea

helicoptertrackgenerated.

DESIGN BEST PRACTICES

With the accumulation of experiences and identification of

possible areas of improvement, the following best practices

andimportantwatchareashavebeenestablishedtoimprove

front-endradarsystemdefinitionanddevelopment,sothatthe

radarismoresuitedforlittoralsurveillance.

InherentFeatures

Inagoodlittoralradardesign,robustclutterrejectionandfalse

alarm control techniques are essential. To prevent receiver

saturationandhandlestrongclutter,thereshouldbeadequate

dynamic range, sensitivity and gain control. Traditional

methodsofgaincontrolthatapplysimilarsuppressionlevels

over the full radar scan will not be able to deal adequately

with the non-homogenous clutter in a littoral environment.

Adaptiveandsector-basedgaincontrolmethodsmaybemore

effectivesolutions.Similarlyforconstantfalsealarmrate,more

sophisticatedandrigorousmethodswillbeneededtoadaptto

backgroundnoiseandclutterstatistics.

Asmediumpulse repetition frequency is often the preferred

waveformschemeforlittoralradars,thesystemdesignshould

alsoincludethetransmissionandprocessingoffill-inpulses.

Withoutthesepulses,theDopplerresponseswillbewidened,

whichcouldpossiblyleadtomorefalsealarms.However,by

assigningadefaultnumberoffill-inpulses,itispotentiallyusing

upmoreradarresourcethannecessaryforthesuppressionof

clutterreturns.

Figure4.AgenericTWSflowchart


Doppler measurement is often regarded only as a tool for

determiningatarget’sradialspeed.Infact,Dopplerinformation

can be harvested for target discrimination, false track rate

control and clutter rejection, all of which are indispensable

properties of a littoral radar. For example, in the Doppler

spectrum, surface clutter is characteristically located in the

zeroDopplerbin.Withwell-designedDopplerfiltersandgood

systemstability,surfacecluttercanbesuppressedeffectively.

Inaclusteroftargetswheredetectionsmightbewithinsimilar

range,azimuthandelevationcells,Dopplercanbe themain

discriminator and help lower the probability of track swaps

or splits. Furthermore, ensuringDoppler coherency in signal

processing makes the system more robust against active

jammers.

High tracking accuracy is highly desirable for target

engagementasitimprovestheprobabilityofkillforweapons

using radarplotsor tracksas theirprimary input forballistic

calculations.However,hightrackaccuracycouldalsosignify

adeeprunningofasingletrackmodelfilterwhichisunableto

copewith targetmanoeuvresandsustain trackcontinuity.A

goodsurveillanceradarshouldhaveanimplementationwhich

isalsoabletoprovidehightrackmaintainability.

Tocounteract the increasedriskof interference, littoral radar

systems should have adequate self-protection measures.

Thesemeasures can reside in the front-enddesign such as

low antenna sidelobes, and in signal processing techniques

suchasasynchronouspulserejection,sidelobeblankingand

frequencyagility.

DedicatedTechniquesandArchitecture

With thematurityofsolidstate transmitters, there isamove

towardsActiveElectronicallyScannedArrayradars.Thisclass

of systems is usually associated with high volume search,

flexiblewaveformmultiplexingandgracefuldegradation. This

architecturecanalso improve thesystem’sdynamic range.

Aconventionalanaloguephasedarrayperformsbeamforming

by means of phase shifters before conversion to digital

signal. As such, the analogue to digital converters have a

highriskofsaturationfromthegainoftheantenna.Fordigital

beamforming,beammanipulationisperformedonlyafterthe

signalsateveryelementaredigitallyconverted.Thisresultsin

adynamicrangegainashighasthegainoftheantenna,which

isbeneficialforhandlingstrongclutterreturns.

Littoralradarsystemsshouldalsohavewaveformstocopewith

ad-hocevents.Theclosenessoftheplatformtosurrounding

coastalareascausesittobemorevulnerabletopop-upairand

surfacetargets.Itisthereforedesirableforthesetrackstobe

initiatedwithasfewplotsaspossible,whileretainingalowfalse

trackrate.Tofurtherreducereactiontime,thereshouldalsobe

ahighdegreeofautomationintheoperationoftheradar.As

much as possible, operator actions should be required only

whentheyhaveadditionalthirdpartyinformationwhichcanbe

usedasinputstosupplementtheradar’sperformance.

Figure5.PropagationlossesindBduetomultipatheffects



SIMULATIONS, TESTS AND FINE-TUNING

Systemdesignreviewsformthebaselinetheoreticalanalysis

of the radar’s capabilities. In order to determine the actual

integrated radar performance, different types of tests from

controlledlaboratorysetupstolocalon-sitetrialsaretypically

conducted.Toevaluatetheeffectivenessoftheimplemented

signalprocessingtechniques,simulationsofRFreturnscanbe

injected in the radar signal processing units. Depending on

thetargetscenariosimulated,parameterssuchasreporting

thresholds and classification criteria can be checked and

furtheroptimised.Fullloadscenarioisoneofthevitalsoftware

teststobeperformedforlittoralradarsystems.Inadditionto

thelargenumberofreportedtracksduetoanexpectedhigh

target density, the radar should also be able to handle the

voluminousinternaldetectionsandunreportedtracksincurred

inalittoralenvironment.

Software simulators are unable to emulate the operational

environmentwithfullfidelityasclutterandpropagationeffects

are often omitted. oneway of filling this gap is to use raw

data collected from similar systems. However, in the usage

ofrawdatafromothersystems,severalareasmustbetaken

careofbyanalysisorscalingtoensurethevalidityofoutput

results.This includes theactual testsetup fromaltitudeand

grazinganglestothescalingofRFfront-endparameterssuch

asantennapatterns,effectiveradiatedpowerandattenuation

settings.

ultimately, local radar testing is the most robust method

of performance validation. Therefore, from a project

management perspective, ample time, sufficient amount of

upfrontplanningandavailabilityofamultitudeoftesttargets

should be catered to allow for comprehensive testing and

fine-tuningof the radar. In general, performance testing can

bebrokendown into: trial planning, conducting of trial, and

analysisofcollecteddata.

In the local environment where both ducting and multipath

can be expected, learning to recognise such propagation

effects and having an in-depth understanding of the impact

ofenvironmentalconditionsarecrucial for trialplanningand

performance analysis. False alarm performance trial is also

verychallenginginalittoralenvironmentcomparedtoanopen

seaasafalsetrackcannotbeverifiedeasily.Hence,variedand

reliablesourcesofgroundtruthneedtobemadeavailableto

validatetheperformanceoftheradar.

LOOKING AHEAD

With improvements in antenna technology and software

processingcapability,manytechniqueswhichwerepreviously

deemedascomputationally intensivehavebecomepractical

toimplementinrealtime.

CognitiveRadar

Haykin(2006)identifiedthreeingredientsofacognitiveradar–

signalprocessingthatbuildsonlearningthroughinteractions

with the surrondings, feedback from receiver to transmitter,

and preservation of collected information. Limited degrees

of cognition are already manifested in many existing radar

modes.AnexampleisLeastJammedFrequency,whereradar

systemssurveythesurroundingspectralenvironmentanduse

these findings to automatically select frequencieswhich are

leastinterferedwithfortransmission.Anotherexampleisthe

mappingofclutterandplotdensitylevelstoprovidefeedback

to the detection process of the radar. For a dynamic littoral

environment,cognitioninradarsystemswillfacilitatecontinual

estimationsandadaptationstotheenvironment.Thiscanlead

tobetterperformanceandreactiontimeinthefuture.

MultipleInputMultipleOutput

Melvin and Scheer (2012) defined a Multiple Input Multiple

output(MIMo)radarasasystemthathasmultipletransmitters

emitting independent waveforms and observes the returns

of the scene of interest with multiple receivers. In a widely

separatedconfigurationasexplainedbyHaimovich,Blumand

and Cimini (2008), MIMo is able to better exploit target

reflectivity using disparate aspect angles, provide enhanced

resolutionofcloselyspacedtargetsandevenimproveDoppler

estimationswiththediversityofreceivedwaveforms.Hence,

the realisation ofMIMosystemscanbring about significant

gainsfortargetdetectionandtrackinginalittoralenvironment.


CONCLUSION

The complex littoral environment imposes a unique set of

challenges for radar systems. The accrual of experiences in

thisdemanding landscapehas resulted in the formulationof

numerous design best practices such as accurate Doppler

measurements,hightrackmaintainabilityandflexibleresource

allocation.Inadditiontotheserequirements,itisalsopivotal

to understand radar behaviour under local conditions and

validate system performance via simulations and trials

comprehensively. Looking ahead, the advent of new

technologies will enable radar systems to incorporate new

techniques that could further improve performance in the

locallittoralenvironment.

ACKNOWLEDGEMENTS

The authors would like to thank the radar community for

sharingtheirinvaluableknowledgeandwealthofexperience.

Theauthorswouldalsoliketoexpresstheirgratitudetowards

MrTngyanSiongforhiscontinualreviewofthearticle.

REFERENCES

Chia, L. S., Khan, H., & Chou, L. M. (1988). The coastal

environmentprofileofSingapore.Retrievedfromhttp://books.

google.com.sg/books/about/The_Coastal_Environmental_

Profile_of_Sin.html?id=qSIJ2uKKs88C

Dawber, B., &Branson, J. (2005). use of site specific radar

modelling to improveCFARperformance in the littoral. IEEE

InternationalRadarConference,Arlington,VirginiaUSA,161-

166.doi:10.1109/RADAR.2005.1435812

deJongh,R.V. (2005).Naval radar ina littoralenvironment.

IEEE MTT-S International Microwave Symposium Digest,

Amsterdam, The Netherlands, 1457-1460. doi: 10.1109/

MWSyM.2005.1516964

Haimovich,A.M.,Blum,R.S.,&Cimini,L. J. (2008).MIMo

radarwithwidelyseparatedantennas.IEEESignalProcessing

Magazine,25(1),116-129.doi:10.1109/MSP.2008.4408448

Haykin, S. Cognitive radar: a way of the future. IEEE

Signal Processing Magazine, 23(1), 30-40. doi: 10.1109/

MSP.2006.1593335

Infocomm Development Authority of Singapore. (n.d.).

Singapore Spectrum Allocation Chart. Retrieved September

25, 2014 from http://www.ida.gov.sg/~/media/Files/

PCDg/Licensees/SpectrumMgmt/SpectrumNumMgmt/

SpectrumChart.pdf

LC81250592013178LgN01. (2013). In U.S. Geological

Survey. Retrieved from http://earthexplorer.usgs.gov/form/

metadatalookup/?collection_id=4923&entity_id=LC81250592

013178LgN01&pageView=1

Martin,J.,&Mulgrew,B.(1990).Analysisoftheoreticalradar

returnsignalfromaircraftpropellerblades.RecordoftheIEEE

1990InternationalRadarConference.Arlington,VA,569-572.

doi:10.1109/RADAR.1990.201091

Melvin,W.L.,&Scheer,J.A.(Eds).(2012).Principlesofmodern

radar:advancedtechniques.Herts,uK:SciTechPublishing.

Myers, H., & Jarrett, R. (1995). Processing techniques for

surfacesurveillanceradarsinlittoralenvironments.Recordof

theIEEE1995InternationalRadarConference.Alexandria,VA,

33-38.doi:10.1109/RADAR.1995.522515

Skolnik,M.I.,(2002).Introductiontoradarsystems.Newyork,

Ny:Mcgraw-HillHigherEducation.

u.S.EnergyInformationAdministration.(2014,November10).

WorldOilTransitChokepoints.Retrievedfromhttp://www.eia.

gov/countries/analysisbriefs/World_oil_Transit_Chokepoints/

wotc.pdf

u.S.Navy.(2010).Navaldoctrinepublication1:navalwarfare.

Retrieved from https://www.usnwc.edu/Academics/Maritime-

-Staff-operators-Course/documents/NDP-1-Naval-Warfare-

(Mar-2010)_Chapters2-3.aspx

youngK.C.,LokeM.K.,ShuiR.C.&Chen,L.(2010).Ducting

phenomenaandtheirimpactonapulsedopplerradar.DSTA

Horizons,88-99.

ENDNOTES

1 Timesidelobesaretheresponsesfromtheoutputofpulse

compression, which is a technique used to improve radar

rangeresolutionandsignaltonoiseratio.



BIOGRAPHY

LOManLingisaSeniorEngineer(Advanced

Systems). She is currently involved in

the development of sensor systems for

naval platforms. Man Ling graduated

with a Bachelor of Science (Electrical and

Computer Engineering) degree from the

universityof IllinoisaturbanaChampaign,

uSA, in2009.ShealsoobtainedaMaster

ofScience (ManagementScienceandEngineering)degree from

Stanforduniversity,uSA,in2010.

LOKE Mun Kwong is DSTA’s Deputy

Director (Technology)attached to theJoint

Plans and Transformation Department

where he is involved in the strategic long-

term capability development planning

for the Singapore Armed Forces. He

has vast experience working on sensor

system applications for ground, air and

navalplatformsaswell as large-scale systems integration.Mun

KwongwasappointedSeniorAdjunctFellowofTemasekDefence

Systems institute at the National university of Singapore in

2006andhassincebeen lecturingonradarsystems.underthe

Defence Technology Training Award (predecessor of the DSTA

Postgraduate Scholarship), he also graduated with aMaster of

EngineeringdegreewithFirstClassHonoursfromImperialCollege

London,uK,in1995.




KABANDSATELLITECoMMuNICATIoNSDESIgNANALySISANDoPTIMISATIoN

INTRODUCTION

Various types of satellites, including geosynchronous Earth

orbit(gEo),MediumEarthorbitandLowEarthorbitsupport

beyondlineofsightcommunications.Thelinkbudgetanalysis

inthisarticleisbasedongEosatellites.AgEosatelliteorbits

atafixedlongitudinallocationatanaltitudeofabout36,000km

abovetheequator.Thetranspondersonthesatelliteprovidea

signalboostandfrequencytranslationofsignalsfortheground

terminals.Theantennasonthesatellitearedesignedtoprovide

therequiredcommunicationscoveragetotheterminalsonthe

ground.Thegroundsegmentcomprisesthehubandremote

terminals of different sizes and transmission powers. The

remote terminalscanbehostedondifferentstaticormobile

platforms.

operating intheKabandofferssomesignificantadvantages

overconventionalsatellitenetworksoperatingintheKuband

and lower frequencies.Notonly ismorebandwidthavailable

at the higher Ka band frequencies, Ka band antennas have

higher gain than antennas of comparable size operating at

lowerfrequencies.However,thedisadvantageofusingtheKa

bandisthatadverseweatherconditions impacttheKaband

LEONGSeeChuan,SUNRu-Tian,YIPPengHon

ABSTRACT

Kabandsatellitecommunications (SATCoM) frequenciesprovidenewopportunities tomeethighbandwidthdemands,especiallyforsmallaerial,maritimeandmobilelandplatformssupportingbeyondlineofsightrequirementsfornetwork-centricoperations.Thisispossibleduetotheavailabilityof3.5gHzofbandwidth,andalsobecauseKagroundsegmentcomponentsaretypicallysmallerindimensioncomparedtothoseofKuband.However,KabandlinksexperiencemuchhigherrainfadesintropicalregionsascomparedtoKubandandCband.Inthisarticle,variousfactorsinthelinkbudgetareexploredtodeterminetheirimpactonoverallsignalstrength.ThesefactorscanbetradedoffandoptimisedtoenabletherealisationofaKabandsolutionforSATCoM.

Keywords:Kaband,satellitecommunications,linkbudget,trade-offanalysis,mitigationtechnique

muchmorethanatlowerfrequencies.Itisthereforeimportant

that there is appropriate planning for the implementation of

well-designed ground systems, network links reliability and

resources so as to mitigate these adverse weather effects

(Petranovichl, 2012) (Abayomi Isiaka yussuff, & Nor Hisham

Khamis,2012)(Brunnenmeyer,Milis&Kung,2012).

This article presents a design approach and analysis of key

satellite communications (SATCoM) network parameters

for a Ka band network. Various trade-offs and optimisation

betweenoperationalparameters(e.g.linkavailability),ground

segment(e.g.poweramplifierratingsandantennasizes)and

space segment (e.g. transponderpower andbandwidth)will

be considered. In addition, mitigation techniques such as

hubsitediversity,adaptivecodingandmodulation(ACM)and

uplinkpowercontrolareexploredtomitigatetheincreasedrain

fadesatKabandandimprovetheoveralllinkavailability.This

analysisdemonstratesthatitisfeasibletousetheKabandto

supportmissioncriticalSATCoMoperationsinourregion.


KA BAND DRIVERS

TheKabandisattractiveasaSATCoMsolutionduetoafew

reasons.

AvailabilityofSpectrumandHigherThroughput

Substantially more spectrum bandwidth is available at the

Ka band than at the Ku band and other lower frequencies.

For example, Ku band allocation is around 2gHz for uplink

and 1.3gHz for downlink with actual contiguous bandwidth

allocation of less than 0.5gHz per satellite. In comparison,

the Ka bandSATCoMhas a bandwidth of 3.5gHz for both

uplinkanddownlink.Table1illustratesthemilitaryandcivilian

frequency allocation.With thewider spectrum availability at

theKaband,highertrafficthroughputcanbesupported.Full

motionvideoforexample,hasbeenidentifiedasakeydriver

inthedemandforbandwidththatcanberealisedbyKaband

satellites (Northern Sky Research [NSR], 2012). In addition,

as theKabandhascommercialandmilitarybandsadjacent

toeachother,commercialservicescanalsocomplementthe

militaryband’scapacity.

GreaterCostEfficiency

Ka band satellites feature narrow spot beams (0.5° to 1.5°

at 3dB beamwidth) which support greater frequency reuse

in geographically isolated spots. With larger allocation and

frequency reuse capabilities, using the Ka band translates

toat leasta1 to2ordermagnitude increase in transponder

throughput,thereforereducingleasingcostperunitbandwidth.

SmallerTerminals

At higher frequencies, wavelengths are smaller, allowing

proportionally smaller, lighter weight and probably less

Band Receive(GHz) Transmit(GHz)

Military 20.2-21.2 30.0-31.0

Civilian 17.7-20.2 27.5-30.0

Table1.FrequencyallocationwithintheKaband

expensiveterminalstoberealised.Thereductionofphysical

dimensions therefore allows Ka band SATCoM to bemade

available for new markets such as manpacks and mobile

platforms.TheuseofmorefocusedandnarrowKabandspot

beams provides higher equivalent isotropic radiated power

(EIRP),signalgain(g/T)andthereforebettersignallinkquality

orhigherdataratesforthesesmallerterminals.Comparingthe

KabandtotheKuband,theimprovementinoveralllinkquality

withtheuseofnarrowspotbeamsis intherangeof6dBto

10dB.

GreaterResiliencytoInterference

WithwiderKabandbandwidth,betterinherentanti-interference

propertiescanbeachieved(e.g.frequencyhoppingordirect

sequencespreadspectrum).WithKabandtranspondersizes

of 125MHz ormore over 54MHz at Ku band, the additional

interferencemarginwithtwicethespreadingcanbeimproved

byatleast3dB.

KA BAND CHALLENGES

WiththeintroductionofsmallermobileterminalsforKaband

SATCoM,morestringentlinkrequirementswillneedtobemet.

Thedesignchallengesareasfollows:

MeetingAdjacentSatelliteInterferenceRegulations

The regulatory bodies governing satellite communications

include the International Telecommunicationunion (ITu) and

the Federal Communications Commission. With the high

densityofsatellitesinorbitandmanymoreKabandsatellites

plannedforlaunch,adjacentsatelliteinterference(ASI)willbe

akeyconcern.Satellite terminals thatwish to transmitmust

meet the emission regulations. ASI is more challenging for


smallterminalswheretheantennasidelobepowersarelarge

withrespecttotheirmainlobes,therebylimitingthemaximum

power they are allowed to transmit. When these terminals

areonthemove,allowableemissionsareconstrainedfurther

as the mechanical antenna pointing accuracy experienced

duringshockandvibrationneedstobeaccountedforduring

movementthroughland,variousseastatesorairturbulence.

LargeRainAttenuation

The SATCoM link that passes through the atmosphere is

degradedbyrain,fog,cloud,ice,snowandhail.Thebiggest

challenge in using the Ka band is the high rain attenuation

compared with the Ku band and higher rainfall rates in

the tropics. Since the electromagnetic wave absorption

componentisincreasedatKaband,theamountofattenuation

per unit length is also increased (see Figure 1). Additional

margin isneededtoensurehighsystemavailabilityortrade-

off in link availability. However, adding an additional margin

may be impractical for remote terminals with small antenna

andlowpoweramplifierthatoperatesinhighrainfallregions.

Forexample,collectedrainstatistics inSingaporegenerated

by Leong and Foo (2007) show a higher rain rate than ITu

specifications (International Telecommunications union –

RadiocommunicationsSector[ITu-R],2012).Thisresultsina

downlinkrainlossof12dBattheKabandversus2.6dBatthe

Kubandtoachieve99%linkavailability.Inadditiontohigher

attenuation,therainfaderateattheKabandwillbeverymuch

higherthanattheKuband.Thehighrainfaderatewillimpact

theoperationofmitigationmeasuressuchasACMalgorithms

builtintothesatellitemodem.

MITIGATION TECHNIQUES

The large rain attenuation at the Ka band may not be

compensatedfullybytheimprovementinKabandnarrowspot

beamsandbetter interferenceenvironment.Degradations in

linkqualitycanbefurthermitigatedbyemployingthreemain

techniques.

HubSiteDiversity

Sitediversityisafademitigationmeasurethatinvolvestwoor

morehubterminalssetuptotransmitorreceivethesignalin

realtimebyusinganalgorithmtochoosetheleastamountof

linkdegradationamongallthehubsitesatanyoneinstance.

Whenonehubexperiencesrainanddetectsthatthelinkmay

becut, thealgorithmcalls foraswitchover to theotherhub

wherethereareclearskies(seeFigure2).

Forsitediversitytobeuseful,therearetwomainconsiderations.

First,hubsitesmustbesufficientlyseparatedtoachievethe

required diversity gain or diversity improvement factor. It is

shownthatdiversitygainimproveswithdistancebutthegain

tapersoffatdistancesmorethan11kmasitcanbetreatedas

asinglesitefadeevent(Leong,Loh,Chen,yip,&Koh,2012).

Table2showsthatthediversitygain isnot justafunctionof

distancebutalsotheorientationofthelineconnectingthetwo

Figure1.Rainattenuationstatisticsat30degreeselevation


sites.ThediversitygainforSentosa-Woodlands(South-North

direction)isalmostequivalenttotheTuas-Changi(West-East)

sitecombinationalthoughthedistancebetweeneachpairof

sitesisquitedifferent.Second,whenasitediversitydecision

ismade,thedowntimeincurredfromthehubswitchoverand

the predicted duration of rain outage must both be taken

intoaccount.Due to thecomplexityofsitediversityand the

resultingcostofimplementation,itwillbemorecosteffective

touseKabandsatellitenetworks.

The hub diversity concept can similarly be extended to

remoteterminals.Inabentpipelink,whenthetransmitterand

receivers are locatedat adistanceapart, the twositesmay

notexperiencethesameamountofrainfallbuttherainfallat

thesitesmaybecorrelated.Therefore,inatypicallinkbudget

planning, the dual rain fade conditions for both the uplink

and downlink are considered when the distance between

the transmitter and receiver is less than 3km. For distances

greaterthan50km,asinglerainfadecondition,usuallyonthe

uplinkside,isconsidered.Inthesetwoplanningmethods,the

rangeofrainattenuationat99%totallinkavailabilityattheKa

bandvariesfrom12dBto39dB.Duetothislargeattenuation

range, it is therefore important toplan the attenuation value

accuratelysoastomeettheenduserservicelevelagreement

while optimising the entire ground and space resources

(Leong,2012).

Figure2.Illustrationofhubsitediversity

Primary Hub

Remote Terminal

Satellite

Secondary Hub

SelectionCombination DivGain/dB Dist/km

Tuas-Sentosa 11.2 22.72

Tuas-Woodlands 10.1 24.40

Sentosa-Woodlands 13.9 23.62

Sentosa-Changi 8.8 23.13

Woodlands-Changi 12.0 27.49

Tuas-Changi 14.8 42.44

Table2.Diversitygainimprovementoverasinglesite



Figure3.Majorlinkparametersusedinlinkbudgetanalysis

Hub + Antenna size/gain + Power -‐ Backoff -‐ Losses -‐ Intermodula<on -‐ ASI regula<ons

Remote terminal + Antenna size/gain + Power -‐ Backoff -‐ Losses -‐ Intermodula<on -‐ ASI regula<ons

Outbound Link Inbound Link

Legend:

Satellite •  Satura<on Flux Density •  Transponder linearity •  Transponder bandwidth •  G/T, EIRP •  Intermodula<on

Link parameters •  data rate •  MODCOD scheme •  Link availability

Propaga6on models •  Free space •  Precipita<on •  Cloud

AdaptiveCodingandModulation

InACM,themodulationandcoding(MoDCoD)ofthecarrier

is altered within the modem in step sizes to increase the

survivabilityof the transmission link.Bydecreasing thedata

rate, thesignal tonoise ratio required fora lowerMoDCoD

is reduced and therefore the carrier becomesmore resilient

to rain fade. To support a varying data rate transmission

during dynamic rain conditions, the video codec running in

the application layer should allow a seamless reduction in

videoqualityorresolutiontoensurethattherecipientisable

toreceiveit. Inotherwords,byadjustingtheMoDCoD,it is

possible tooptimise the trade-offbetweenperformanceand

survivability.Applications thereforeneed tobedesignedand

testedaccordinglytotakefulladvantageoftheACMcapability.

ACMtypicallyprovides15dBofmarginacrossthefullrangeof

MoDCoDs.

AutomaticUplinkPowerControl

Automatic uplink Power Control (AuPC) is implemented by

increasing carrier power at the transmit end to ensure link

survivability. When a rain fade event is encountered, more

powerisdrawnfromthehighpoweramplifier(HPA)tomaintain

the carrier to noise ratio. Due to the need for additional

equipment, AuPC is usually employed only at larger hub

stationssincethesmallerremoteterminals’HPAmayalready

beoperatingwithnegligiblebackoffduringclearsky.AuPCat

hubstationstypicallyprovide15dBofpowercontrolmargin.

DESIGN ANALYSIS AND OPTIMISATION

Takingintoconsiderationspacesegmentparameters;ground

segmentmitigation techniques that improve the link quality;

environmentfactorsthatdecreasethelinkqualitysignificantly;

and the increased use of high bandwidth demand video

application, a more stringent design analysis approach for

link budget calculations is required. The approach will also

requireasensitivityanalysis,wherevarioustrade-offsbetween

operationalparameters(e.g.desiredlinkavailabilityforcontrol

andmissionlinks),groundsegment(e.g.poweramplifierratings

and antenna sizes) and space segment (e.g. transponder

power and bandwidth) can be analysed and optimised.

Throughthesetrade-offanalyses,thefeasibilityofusingtheKa

bandtosupportmissioncriticalmilitaryaeronautical,maritime

andlandSATCoMoperationscanbedetermined.

Therearemanyparameterstoconsiderinthelinkbudget.The

primaryparametersareasshowninFigure3.


It is recommended to start the satellite network design by

first identifying the design boundaries –which are themost

constraining factor(s) and which are the parameters that

arewithin and outside of the designers’ control. The typical

constraintsareasfollows:

Satellites

usually,theareaofoperationswilldefinethechoiceofsatellites.

If two or more satellites are able to provide the required

coverage, thenparameters suchas theavailablepowerand

bandwidthonthetransponder,receiverg/T,saturationpoints

ofthereceiversandsaturationfluxdensity(SFD)canbeused

forthetrade-offanalysis.The linearityof thetransponders is

also an indicator of their performance. Themore linear they

are,thelowertheintermodulationnoiserelativetothecarrier

will be produced, and therefore the better the output signal

whichcanbeachieved.

RemoteTerminalsandHub

Constraints for remote terminals include the infrastructureor

platformtheywillbehostedin.Iftheterminalsaretobeused

onthemove,theplatformwillverylikelylimittheantennasize/

weight, position, minimum/maximum elevation angles and/

orpoweramplifiersize.Ifthehubhasbeenimplemented,its

fixedinfrastructuresuchasantennasizeandpoweramplifier

size may be constraining factors. Transmit power back-off

(reduction in the transmit power level) and intermodulation

noiseshouldbecatered for ifmultiple frequencycarriersare

transmitted from a common power amplifier. Losses due to

cablesandinterconnectorsaswellasinaccuraciesinantenna

pointingshouldalsobetakenintoaccount.

Besides these technical parameters, the satellite network

designershouldalso takemarketavailabilityof theproducts

intoconsideration.

CommunicationLinks

a) Outbound Link - The outbound link is the overall

communicationslinkfromthehubtotheterminal.Itconsistsof

thehubuplinkandtheterminaldownlink.Theoutboundlinkis

generallyengineeredsothattheterminaldownlinkdominates

performance. Since the hub services many terminals, it is

generallycosteffectivetomakethehubantennalargeenough

toprovideextratransmitpowermarginonthehubuplink.

b) Inbound Link - The inbound link is the overall

communicationslinkfromtheterminaltothehub.Itconsistsof

theterminaluplinkandthehubdownlink.Theinboundlinkis

alsogenerallyengineeredsothattheterminaluplinkdominates

performance, since the large hub antenna provides extra

receivegainonthehubdownlink.

c) MODCODScheme-ThechoiceofMoDCoDisrelatedto

thesignaltonoiseratiorequiredbythemodemtodemodulate

the signal successfully as well as the carrier bandwidth

required. These parameters are usually referenced from the

modem specifications. The available transmit power or the

receiversensitivitymaylimitthechoiceofMoDCoDscheme.

OperationalInputs

The operational inputs consist of the information exchange

requirements, data rates and link availability required for

themission. Depending on the application andmission, the

end user may have minimum data rate and link availability

requirements.Thesewouldthenbesetasdesigntargetsand

inputs to the link budget analysis. They impact the satellite

transponderresourcesdirectlysuchaspowerandbandwidth

requiredtosupportthelink.

CASE STUDY: SATCOM ON THE MOVE

A remote terminal antenna size of 0.45m or 0.6m, power

amplifierofupto20Wandaninboundlinkofupto5Mbpsare

usedastheinputparametersinthiscasestudy.Ifthechoiceof

satelliteisstillopen,thedesignershouldlookforonewithhigh

g/Tandhighlinearitytransponderinordertomeetthedesired

link availability for the mission and minimise the resources

required.

SensitivityAnalysis

With numerous link budget parameters, sensitivity analysis

is needed to determine the critical trade-offs between size,

power, bandwidth and link availability. The key findings are

highlightedasfollows:

a) Increasing remote terminal antenna size from 0.45m to

0.60mallowsa reduction in the required transponderpower

equivalent bandwidth (PEB) by 20% to 40% per 64Kbps

link, leading to long-term savings in operating expenses. At

thesametime,itallowstherequiredpoweronthehubtobe

reducedby30%to40%.Bothdirectlycontributetoanincrease

inthenumberofremoteterminalsthatcanbesupported.



b) Itisestimatedthatasingletranspondercansupportabout

9x5Mbpsor16x3Mbpsmissionlinks.Forthemissionlink,

satelliteSFD–aparametercontrolledbythesatelliteservice

provider–andtheEIRPcontourinwhichthehubis located,

are themajor factors influencing the number of links which

can be supported per satellite transponder. Increasing the

SFD sensitivity level by 6dBW/m2 reduces the transponder

PEB requiredby60%to70%, leading tosignificantsavings

in operating expenses. It is therefore important to choose,

negotiateandestablishaservicelevelwiththesatelliteservice

providerwhichmeetuserrequirements.

c) Foramissionlinkwithhighdatarate(3Mbpsto5Mbps)but

smallantenna(0.45mto0.6m)andlimitedpower(upto20W),

themaximumlinkavailabilityisonly96%to97%.Withlower

datarates(below1Mbps),ahigher linkavailabilityofat least

98%canbeachieved.

ApplicationofMitigationTechniques

Hub Site Diversity

Hub site diversity provides a means to overcome rain fade

onthepathbetweenthehubandthesatellite.Consequently,

whenthereisnorainattenuation,thenumberoflinksthatcan

be supported per transponder/hub increases. In essence,

this increases the total capacity of the satellite network in

termsof increasing thenumberof remote terminals thatcan

besupportedper satellite transponder.For remote terminals

equipped with 0.45m antenna and up to 20W power, hub

site diversity can increase the number of remote terminals

supportedpertransponderbyupto18%.

Adaptive Coding and Modulation

ThemissionlinkavailabilitywillbeimprovedifACMisapplied.

Duringraineventswhenthelinkfunctionsindegradedmode,

forexampleata lowerdatarate,videosaretransmittedata

lowerresolution.Bydecreasingthedatarate from1Mbpsto

512Kbpsor256Kbps,thelinkavailabilityisincreasedfrom98%

to98.5%.This translatestoareduction indowntimeof43.8

hoursperyear.Commercial-off-the-shelfsatellitemodemsare

usuallyequippedwithACMthatenablethelinktobesustained

aslinkconditionsdeteriorate.

OperationalConsiderations

Besidesdesigninganetworkwiththerequiredlinkavailability,

data ratesandpower, it isnecessary toaddressoperational

concernsandplanforcontingencies.

Impact of Loss of Mission Link and Mitigation

Alinkof64Kbpscouldbelostinrainexceedingapproximately

20mm/hr.Theimpacttothemissiondependsonfactorssuch

as theperiodof linkoutageand latency requirementsof the

data.Mitigatingmeasuresforlinkoutagecanincludeastore-

and-forwardmethodwherebythedataisstoredonboardthe

platformuntilacommunicationslinkisre-established.

Link Resiliency

Thelinksshouldbedesignedtoberobustagainstintentional

or unintentional interferences. The communications security

andtransmissionsecurityfeaturesoftheSATCoMlinkdepend

toalargeextentonthemodemcapabilitiesandwaveform.The

accuracyoftrackingandpointingaswellasthedesignofthe

SATCoM antennas, especially on side lobe emissions, also

playapartinreducinginterferencesinthenetwork.


CONCLUSION

The use of the Ka band in SATCoM has allowed for new

and smaller mobile terminals that utilise high throughput

applications as compared to the Ku band to be feasible

options in operations. However, with significantly larger rain

attenuation to overcome, the Ka band link budget design

analysis ismore complex than in lower frequency bands to

achievecomparablelinkavailability.Theuseofsensitivityand

trade-offanalysisintheillustratedSATCoMonthemovecase

studydemonstratesthefeasibilityofKabandSATCoMinour

region.otherKabandoperationalconsiderations–suchasthe

possibilityof fallback to lower frequencybandduringsevere

fade conditions and change in transmission plans required

whencrossingovermultiplespotbeamstocovertheareaof

operation–mayalsobeincludedaspartofthedesignanalysis

uponfutureexploration.

REFERENCES

Abayomi Isiaka yussuff, & Nor Hisham Khamis (2012). Rain

attenuationmodellingandmitigationinthetropics:briefreview.

InternationalJournalofElectricalandComputerEngineering,

2(6),748.

Brunnenmeyer, D., Mills, S., Patel, S., Suarez, C., & Kung,

L. (2012, october). Ka and ku operational considerations

for military SATCoM applications. Military Communications

Conference,2012–MILCOM2012.

International Telecommunications union –

Radiocommunications Sector. (2012). Characteristics of

precipitationforpropagationmodelling(P.837-6).geneva.

Leong, S. C. (2012). Extraction of distance-dependent rain

rate distributions for satellite links calculation. Progress In

ElectromagneticsResearchSymposiumProceedings,Russia.

1571–1575.Retrievedfromhttp://piers.org/piersproceedings/

piers2012MoscowProc.php?start=300


Leong,S.C.,&Foo,y.C. (2007,December).Singaporerain

ratedistributions.6thInternationalConferenceonInformation,

Communications&SignalProcessing,2007.

Leong,S.C.,Loh,W.J.,Chen,y.J.,yip,P.H.,&Koh,B.T.(2012).

Evaluationofsitediversityeffectivenessusingweather radar

data for Singapore. Progress In Electromagnetics Research

SymposiumProceedings,Malaysia,620–625.Retrievedfrom

http://piers.org/piersproceedings/piers2012KualalumpurProc.

php?start=100

NorthernSkyResearch.(2012).Government&militarysatellite

communications(8thed.).Cambridge,MA:NSR.

Petranovich, J. (2012). Mitigating the effect of weather on

ka-band high-capacity satellites. Retrieved from https://

www.viasat.com/files/assets/Broadband%20Systems/

Mitigating%20the%20Effect%20of%20Weather%20on%20

Ka-Band%20High%20Capacity%20Satellites.pdf


BIOGRAPHY

LEONG See Chuan is a Development

Manager (C4I Development). He has

designed,developedandmanagedcomplex

software based command and control

systemsincludingsatellitecommunications

(SATCoM). He has published numerous

academic publications, some of which

are related toSATCoM,withabestpaper

presentationawardinanIEEEconference.ArecipientofthePublic

Service Commission Scholarship, See Chuan graduated with

a Bachelor of Engineering (Electrical Engineering) degree from

the National university of Singapore (NuS) in 1999. He further

obtainedaMasterofEngineering (ElectricalEngineering)degree

fromNuSin2002.

SUN Ru-Tian is a Principal Engineer

(Advanced Systems). She was a project

manager for ground SATCoM systems

and is currently involved in the front-end

planning in the domain of communication

switching systems and radios. A recipient

of the DSTA undergraduate Scholarship,

Ru-Tian graduated with a Bachelor of

Engineering(ElectricalEngineering)degreefromNuSin2005.She

further obtained aMaster of Science (SatelliteCommunications

Engineering) degree from the university of Surrey, uK, in 2009

undertheDSTAPostgraduateScholarship.

YIP Peng Hon is a Senior Principal

Engineer (Advanced Systems) who has

many years of experience managing

large-scale communications network

projectsforgroundandnavalplatforms.He

iscurrentlyinvolvedinthefront-endplanning

and systemsarchitectingof theSingapore

Armed Forces’ SATCoM capabilities.

Peng Hon graduated with a Bachelor of Engineering (Electrical

Engineering) degree and aMaster of Science (Communications

andComputerNetworking) degree fromNanyang Technological

universityin1993and2000respectively.




PERFoRMANCECHALLENgESFoRHIgHRESoLuTIoNIMAgINgSENSoRSFoRSuRVEILLANCEINTRoPICALENVIRoNMENT

INTRODUCTION

In the modern battlefield, high resolution imaging sensors,

typicallyoperatinginthevisibleorinfrared(IR)electromagnetic

(EM) spectrum, provide a distinct advantage by detecting

targets and offering an unambiguous means of target

identification.However,environmentalconditionscanlimitthe

performanceofsuchsystemsduetotheeffectsofatmospheric

gasesandaerosolsonEMwaves.Specifictothelocaltropical

environment,keyfactorsaffectingsensorperformanceinclude

humidity, rain,cloudsandhaze.Thesefactors, togetherwith

thebasicattenuationmechanisms,arecoveredindetailinthe

followingsections.

TYPES OF IMAGING SENSORS

Imaging sensors can be broadly classified into passive and

active imagingsensors.Apassive imagingsensor intercepts

andcollectssurroundingEMradiationreflectedofforemitted

byobjectsfoundwithinthefield-of-viewofitsdetectortoform

imagesofitssurroundings.Anactiveimagingsensoriscoupled

with an illumination source to illuminate the objects to be

LEECheowGim,EEKokTiong,HENGYinghuiElizabeth

ABSTRACT

Electro-optical(Eo)sensorsofferanunambiguousmeansoftargetidentification,albeitwithlimitationsinrangeperformanceduetoenvironmentalconditions.Specifictothelocaltropicalenvironment,factorsaffectingsensorperformanceincludehumidity,rainandclouds.Anotheremergingkeyenvironmentalfactor isthepresenceofhaze.Thisarticleexaminesthephysicsbehind theseenvironmental factorsand their impacton theperformanceofdifferentsensors. ItalsodiscusseshowspecificEocharacteristicscanbeleveragedtoenhancesensors’performanceinthelocalenvironmentandhighlightsemergingtechnologiesforfutureconsideration.

Keywords:electro-optical,targetidentification,imagingsensor,rangeperformance,attenuation

observed.Acompatibledetectorwillthencollectthereflected

energy for extended surveillance range performance and

betterspatialresolutionunderlowlightandindarkconditions.

Different detectormaterials are sensitive to radiation energy

fromdifferentportionsoftheEMspectrum.

Intheday,thedominantsourceofvisiblespectrumradiation

is thesun,whileatnight,asignificantamountofnightglow

comes from the moon and stars. Detectors that operate

in the visible spectrum typically make use of either the

charge coupled device or the complementary metal oxide

semiconductortechnologiesforimagingindaytimescenarios

when ample ambient light exists. Detectors using image

intensifier(II)technologymakeuseofphotonamplificationto

generateimagesinlow-lightscenarios.

on the other hand, thermal imagers (TI) are used to detect

heatorradiationemittedfromhotobjectsinthemidwaveIR

(MWIR) or longwave IR (LWIR) spectrums to generate two-

dimensionalimages,basedontheblack-bodyradiationcurve

and thecontrastbetween the temperaturesof theseobjects

and theirbackground.These imagersoperate independently


from ambient lighting condition and even in complete

darkness. TI detectors are further classified into two types

– one that relies on changes to the temperature-dependent

property of itsmaterial, and another that determines the IR

photon flux by measuring the electrical current due to the

generation of electron-hole pairs caused by the absorption

of incidentphotons.Commonmaterialsused to fabricateTI

detectorsareindiumgalliumarsenide(IngaAs),leadsulphide,

indiumantimonide,mercurycadmiumtellurideandvanadium

pentoxide.

BASIC ATTENUATION MECHANISMS

Atmosphericattenuationiscausedbyboththeabsorptionand

scatteringofEM radiation fromaerosolparticlesormoisture

dropletssuspendedintheatmosphere.EMradiationismainly

absorbed by gaseous agents such aswater vapour, carbon

dioxide,nitrousoxide,ozone,molecularandatomicoxygen,

andnitrogen.Theabsorptionisgenerallynegligibleinthevisible

regionandataminimuminafewatmosphericwindowsinthe

IRregion.ThewindowbandsinthevisibleandIRspectrumare

summarisedinTable1andillustratedinFigure1.

AtmosphericscatteringistheprocessbywhichEMradiation

is redirected by gaseous molecules and aerosol particles

suspendedintheatmosphere.Itisgovernedbytherelationship

betweentheradiiofthescatteringmoleculesorparticlesand

thewavelengthoftheincidentradiation.

Atmosphericabsorptionandscatteringcreateatwo-foldeffect

on luminanceandcontrast transmittance.First,EMradiation

WindowBandsinEMSpectrum Wavelength

Visible 0.4µmto0.7µm

NearIRandShortIR 1µmto2µm

MidIR 3µmto5µm

FarIR 8µmto12µm

Table1.WindowbandsinthevisibleandIRspectrum

Figure1.AtmosphericwindowsinthevisibleandIRregions

froma targetand its immediatebackground isprogressively

scattered out of the viewing path. Some will be absorbed

andwillnotreachthesensor.TheattenuationofEMradiation

followsanexponentiallawinhomogeneousair.AbeamofEM

radiationcontainingafluxFoatthetargetofrangeRfromthe

sensorwill have a residual fluxF received by the sensor in

homogeneousair,givenby:

3

Atmospheric scattering is the process by which EM radiation is redirected by gaseous molecules and aerosol particles suspended in the atmosphere. It is governed by the relationship between the radii of the scattering molecules or particles with the wavelength of the incident radiation.

Atmospheric absorption and scattering create a twofold effect on luminance and contrast transmittance. First, EM radiation from a target and its immediate background is progressively scattered out of the viewing path. Some will be absorbed and will not reach the sensor. The attenuation of EM radiation follows an exponential law in homogeneous air. A beam of EM radiation containing a flux Fo at the target of range R from the sensor will have a residual flux F received by the sensor in homogeneous air, given by:

)exp(])(exp[ RFRkbFF eoo

where b and k are the scattering coefficient and absorption coefficient respectively, and σe is the extinction coefficient of the atmosphere. Second, EM radiation, which does not come directly from the target or its immediate background, is scattered into the viewing path. This additional radiation is called airlight or path radiance, and varies with the scattering angle.

ATMOSPHERIC ATTENUATION IN THE LOCAL ENVIRONMENT

The key atmospheric elements in our local environment which attenuate the sensors’ performance in the visible and IR spectrum are rain, clouds, fog and haze. The first three elements are related to the presence of water vapour, which is the most influential absorbing gas as well as the most variable. Relative humidity and absolute humidity are the two environmental parameters associated with water vapour content.

Rain Attenuation

Rain effects are difficult to estimate because of the variation in droplet size, density, drop size distribution, phase function, raindrop shape and the different effects that the water refractive index has on different waveband portions of the EM spectrum. Rain attenuates EM radiation through absorption and scattering, with the relative amounts dependant on the ratio of raindrop radius to wavelength. In the visible and IR spectrums, attenuation by rain is independent of wavelength because the raindrop radius (typically from 0.5mm to 5mm) is much larger than the wavelength. The extinction coefficient, which measures how strongly the EM radiation is absorbed by the medium (rain), indicates that the greater the rainfall rate, the higher the absorption.

Rainfall is the most significant atmospheric element affecting sensors operating in the visible and IR portions of the EM spectrum in Singapore, where rainfall occurs every month of the year. The two main wet seasons are the Northeast monsoon season from late November to March, and the Southwest monsoon season from late May to September, which account for 48% and 36% of the annual rainfall respectively. The type of rainfall varies from drizzle, which has a rain rate of up to 1mm/hr, to


whereb and k are the scattering coefficient and absorption

coefficient respectively, and σe is the extinction coefficient

of the atmosphere. Second, EM radiation, which does not

come directly from the target or its immediate background,

isscatteredintotheviewingpath.Thisadditionalradiationis

calledair-lightorpathradiance,andvarieswiththescattering

angle.

ATMOSPHERIC ATTENUATION IN THE LOCAL ENVIRONMENT

Thekeyatmosphericelementsinthelocalenvironmentwhich

attenuate the sensors’ performance in the visible and IR

spectrumarerain,clouds,fogandhaze.Thefirstthreeelements

arerelatedtothepresenceofwatervapour,whichisthemost

influentialabsorbinggasaswellasthemostvariable.Relative

humidity and absolute humidity are the two environmental

parametersassociatedwithwatervapourcontent.

RainAttenuation

Raineffectsaredifficult toestimatebecauseof thevariation

indropletsize,density,dropsizedistribution,phasefunction,

raindrop shape and the different effects that the water

refractiveindexhasondifferentwavebandportionsoftheEM

spectrum. Rain attenuates EM radiation through absorption

and scattering, with the relative amounts dependant on the

ratio of raindrop radius towavelength. In the visible and IR

spectrums,attenuationbyrainisindependentfromwavelength

becausetheraindropradius(typicallyfrom0.5mmto5mm)is

much larger than thewavelength. The extinction coefficient,

whichmeasureshowstronglytheEMradiationisabsorbedby

themedium (rain), indicates that thegreater the rainfall rate,

thehighertheabsorption.

Rainfall isthemostsignificantatmosphericelementaffecting

sensors operating in the visible and IR portions of the EM

spectruminSingapore,whererainfalloccurseverymonthof

theyear.ThetwomainwetseasonsaretheNortheastmonsoon

season from late November to March, and the Southwest

monsoonseasonfromlateMaytoSeptember,whichaccount

for48%and36%oftheannualrainfallrespectively.Thetype

of rainfall varies fromdrizzle,whichhas a rain rate of up to

1mm/hr,tothunderstormswithrainratesexceeding50mm/hr.

Thunderstormsareobservedduring78%of thosedayswith

rainfall.ToillustratetheimpactofrainonvisibleorIRsensors,a

moderaterainrateof10mm/hrwouldallowonlyapproximately

6.7%oftheEMradiationtopassthrougha1.8kmpath.Along

a10kmpath,thetransmittanceeventhroughalightrainwith

rain rate of 2.5mm/hrwouldbeonly 0.1%.Thismeans that

visible and IR sensors are renderedalmost useless in either

case.Table2showsthevariousrainfallstatisticsfordifferent

inputparameters(Bernardetal,2013).

Rainrate 5mm/h

(lightrain)

10mm/h

(moderaterain)

20mm/h

(heavyrain)

Extinction 0.5km-1 0.7km-1 ~1.2km-1

WaterVolume/RainVolume 320mm3m-3 588mm3m-3 1070mm3m-3

Numberofraindrop 3600m-3 4800m-3 6400m-3

Table2.Variousrainfallstatisticsfordifferentinputparameters,computedusingMarshall-Palmerdistribution


CloudAttenuation

Similar torain,cloudeffectsaredifficult to forecastbecause

of the variation inparticle size.Attenuation isdependenton

thecloudtypeandwatervapourcontent.Withinthecloud,the

attenuationofEMradiationbywaterdroplets isdue toboth

absorptionandscattering.

CloudsinSingaporetendtohavehigherwatercontentbecause

warmerairholdsmoremoisture thancolderair.Themedian

cloud cover in Singapore is 90% (mostly cloudy) and does

not vary significantly. The typical cloud forms in Singapore

are cirrus clouds (high clouds with bases above 20,000ft),

altocumulus clouds (medium clouds with bases between

7,000ft to 20,000ft) as well as cumulus and cumulonimbus

clouds (low clouds with bases below 7,000ft). The typical

cloudthicknessrangesfrom1kmto6kmwiththeradiiofthe

cloudwaterdropletsrangingfromabout0.5µmto80µm.The

extinction coefficient increases by an order of 10 for every

kilometreofcloudthickness.Hence,theverticaltransmittance

through such clouds would be less than 0.005%, and the

cloudswouldbeopaquetovisibleandIRradiation.

FogAttenuation

Water vapour in the air usually condenses high in the

atmospheretoformcloudsbutitcanalsocondensecloseto

thegroundtoformfog.Fogformswhenthedifferencebetween

atmospherictemperatureanddewpointislessthan2.5°Cand

occursatrelativehumiditynear100%,whichmeansthatthe

airwillbecomesupersaturatedifadditionalmoistureisadded.

Whentheairisalmostsaturatedwithwatervapouratrelative

humidityofcloseto100%,fogcanforminthepresenceofa

sufficientnumberofcondensationnucleiwhichcanbesmoke

or dust particles. The reason for degradation of visibility in

a foggy atmosphere is the absorption and scattering of EM

radiation by fog particles. The degradation is dependent on

thedropletsizeanditsdistribution.Foghassimilareffectsto

thatofclouds.

In Singapore, fog is most likely to occur during the inter-

monsoon periods, at times when winds are light and on

cloud-free nights with high humidity. on the average, the

dewpoint forSingaporevariesfrom22°Cto27°C,whilethe

atmospheric temperature varies from a low 25°C to 27°C

during the monsoon seasons and inter-monsoon periods

respectively.Thefogobserved inSingapore ismainlydueto

radiative cooling of the land,which causes the temperature

of the air near ground level to fall within 2.5°C of the dew

point, resulting in thecondensationofwaterdroplets. It can

beobservedthatattenuationduetoanevolvingfogdecreases

rapidlywithincreasingwavelength.Hence,IRsensorsperform

betterthanvisiblesensorsinfoggyconditions.Thesameisnot

truefortransmissionthroughstablefog,whereIRattenuation

issoseverethatIRsensorshavelittleadvantageovervisible

sensors.

DeterminingVisibleandInfraredSensorPerformanceUsingTransmittance

Visualrangeisameasureofvisibility(Malm,1999).Thelarger

thevisualrange,thebetterthevisibility.Visibilityiscalculated

fromameasurementofEMradiationextinctionwhichincludes

thescatteringandabsorptionoflightbyparticlesandgases.

Innaturalweatherscenarios,transmittancealongtheline-of-

sightbetweenthesensoranditstargetisdegradeduniformly.

Thisdegradationischaracterisedbyanextinctioncoefficient

αkm-1.Theextinctioncoefficientquantifieshowthepassage

oflightfromascenetoanobserverisaffectedbyairparticles

(Malm,1999).Theextinctionisdependentonparticlemassand

chemicalcomposition.Toestimatetheperformanceofsensors

inman-madeinducedobscurants,transmittancedegradation

in the form of a mass extinction coefficient α’ m2/g for the

obscurantisaddedtonaturalweatherdegradation.Hence,to

determine the sensorperformance, its transmittancecanbe

calculatedusingtheequation:

Transmittance,T=exp(-αR-α’CL)

Whereα = ambientatmosphereextinctioncoefficient(km-1)

R = range(km)fromtargettosensor

α’ = inducedobscurantmassextinctioncoefficient (m2/g)

CL= induced obscurant concentration pathlength

(g/m2)

Inotherwords,rangecanbecalculatedas:

R=-[ln(T)+ α’CL]/α

Therangeequationshowsanegativelogarithmicrelationship

to transmittance. However, the equation does not show a

direct relationship towavelength as it is affected to varying

degreesbyambientatmosphereandobscurant.

Whiletheaboveequationsprovideageneralguideonpredicting

sensorperformance,itmustbenotedthattheindustryemploys

more complex simulation software such as the TACoM

Thermal ImageModel, NVThermIP withMoDTRAN, NVESD



TimeDependentSearchParameter searchmodel for search

and detection predictions, andmost recentlyNVLabCap for

detailedanalysisandpredictionofsensorperformanceunder

differentenvironmentalconditions.Thesesoftwareareableto

evaluateimagingsensorsystemperformancecomprehensively

based on multiple factors including system design and

tolerances which cannot be adequately represented in the

equations mentioned. The NVESD Time Dependent Search

Parameter search model, NVThermIP and NVLabCap are

authoritative simulation systemsdevelopedby theuSArmy

NightVisionandElectronicSensorsDirectorateovertheyears

to improve the accuracy and repeatability of measurement

techniques,intheirefforttocharacteriseandevaluateavariety

ofEoimagingsystemsfortheuSArmy.

ImpactofHazeonSensorPerformance

Haze refers to small particles dispersed throughout the

atmospheric aerosol (Chen, 1975). Biomass burning is a

global phenomenon that releases large quantities of gases

and aerosol particles into the atmosphere, affecting the

atmospheric chemistry and climate on a large scale via

the scattering and absorption of solar radiation (Li, Shao, &

Buseck, 2010). Aerosols in regional hazes are contributed

largelybyanthropogenicsourcessuchasindustrialemissions,

coalpowerplantoperations,vehicularfossilfuelcombustion,

andagriculturalbiomassburning(ABB).Academicresearchers

in China who conducted studies about the severe haze

situation in Beijing and northern China, have defined seven

majorfineaerosolparticlescommonlyfoundinhaze–namely

mineral,soot,organicmatter,flyash,K-rich,S-richandmetal

particles (Lietal,2010).Notably, theirstudies indicated that

theageingofsootparticlesfromABBinahighrelativehumidity

environmentincreasedtheabsorptionofvisiblesolarradiation

ascomparedtosootalone.

Attenuationduetohazeisverycomplexbecauseofdiversity

intheparticletype,size,shapeandsizedistributioninahazy

atmosphere.Ingeneral,hazeattenuatesvisibleradiationmore

than it attenuates IR radiationdue to the small sizeof haze

particles.This isbecause thesmalldiameterof theparticles

typically coincides with the short wavelengths found in the

visibleportionoftheEMspectrum.Scatteringisthedominant

attenuation factor. However, in regions with high relative

humidity, aerosol liquid water absorption can increase as

moisturecancondenseontheparticles.Experimentalstudies

haveshownthatattenuationatlongerwavelengthsislessthan

thatatshorterwavelengths.Hence,IRsensorscanpenetrate

further through haze than visible optical sensors. Figure 2

showsthephotographstakeninSingaporeonacleardayand

onahazydayin2013.

Specific to Southeast Asia (SEA) is the recurring biomass

burning-inducedsmokehaze.TheparticlesintheSEAhazeare

contributedprimarilybyforestfiresburninginIndonesia.This

isdifferentfromtheparticlescontributedbyanamalgamation

of sources, including industrial emissions, coal power plant

operations and vehicular fossil fuel combustion on top of

Figure2.DaytimephotographshowingBedokSouthAve1,lookingwesttowardsMarineParadeat12:04pmon24June2013(left)(©Wolcott/File:MarineParadeRoad(2).jpg/WikimediaCommons/CCBy3.0)andviewfromthesame

vantagepointon21June2013at10:35am,whenthehazewasathazardouslevels(right)(©Wolcott/File:HazeobscuringMarineParade.jpg/WikimediaCommons/CCBy3.0).


ABBinnorthernChinathatresulted inthepersistenthaze in

Beijing.oneoftheSEA’sworstairpollutioneventstookplace

inJune2013whenthe three-hourPollutionStandards Index

(PSI) inSingapore reacheda recordhighof401on21June

2013.Figure3illustratestheseverityofthehaze’simpacton

visibilitywhenthePSIreached312.

Studies by local academic researchers noted that the

concentrations of particulate matter of 2.5µm size (PM2.5)

increasedmore than15 timesduring thehaze inJune2013

ascomparedwithnon-hazeperiods.Toexploretheparticles’

interactionwithlight,theopticalpropertiesofambientaerosols

wereexaminedintermsofbap,thelightabsorptioncoefficient

ofparticles,aswellasbsp, the lightscatteringcoefficientof

particles. Itwas found that PM2.5wasmore correlatedwith

bsp than bap. This implied that attenuation by PM2.5 was

mainly due to the scattering of light by the particles (See,

Balasuhramanian,&Wong,2006).

The studies also reported that fine particles were generally

found ingreatermassconcentration thancoarseparticles in

ourlocalenvironment.Aninterestingobservationmadeduring

hazydayswas thatwhile themass concentration increased

across theentiresize range, the increase incoarseparticles

was larger than in fine particles. A possible reason for this

couldbethatwhilebiomassburninginIndonesiaemittedmore

fineparticlesthancoarseparticles,someofthefineparticles

amalgamated to formcoarseparticlesduring the long range

transporttoSingapore.

Figure3.Daytimephotographofatrafficjunctiontakenfrom1km(left)andcloserangeof50m(right)whenthePSIreached312inJune2013.

In summary, the increase inparticles,particularly thosewith

diameter similar to thewavelength of light, is thought to be

responsibleforvisibilityimpairmentonhazydays.Sincevisible

light would be more strongly attenuated, IR sensors would

performbetterundersuchconditions.However,thescattering

andabsorptioncoefficientsofhazeparticlesincreaseastheir

concentrations increase. This could reduce theperformance

of even IR sensors, as the wavelengths at which they are

operatingcouldalsobeseverelyattenuated.

EMERGING TECHNOLOGIES FOR FURTHER CONSIDERATION

Localenvironmentaleffectsaffectsensorperformance,albeit

to different extents on different wavelengths. Multispectral

or hyperspectral detection technologies could potentially

reducesensorperformancedegradationbyapplyingspectral

signature analysis across multiple wavelengths. In addition,

sensingintheshortwaveIR(SWIR)portionoftheEMspectrum

(wavelengthsfrom0.9to1.7microns)hasrecentlybeenmade

viable with the maturity of IngaAs sensors. Image fusion

and enhancement techniques are examined in the following

sections.



MultispectralandHyperspectralSensors

Multispectralsensorsdetectthewavelengthsacrossdifferent

bands in the EM spectrum. The spectral signature of the

detectedtargetisthencomparedtothoseofknowntargetsto

determineifthereisamatch.Similarly,hyperspectralsensors

create a larger number of images from contiguous, rather

than disjointed regions of the EM spectrum, typically with

muchfinerresolution.Thefinerresolutionprovidesadditional

informationforthespectralsignatureanalysis(Renhornetal,

2013).Studieshaveshown that suchsensorscouldprovide

significant improvement in target detection performance as

well as improve false alarm performance over single-band

sensors.Thechallenge,however, is the requirement tobuild

acomprehensivedatabaseofthetargets’spectralsignatures

forcomparisonduringtheanalysis.onedevelopmentthathas

potentiallyspedupthedevelopmentofviablemultispectraland

hyperspectralsensors istheQuantumWell IRPhotodetector

(QWIP).

The QWIP typically consists of multiple quantum wells

sandwiched between its emitter and collector contacts

(Ting et al, 2014). By adjusting the width and depth of

the well, the sub-band transition energy and absorption

wavelength can be adjusted. The gallium arsenide-based

or aluminium gallium arsenide-based QWIP stands out

as a prime candidate for the development of Focal Plane

Arrays (FPA)due to its favourable inherentproperties.These

include its ease of fabrication, ruggedness, pixel-to-pixel

uniformity, high pixel operability, temporal stability and its

abilitytobetailoredtotheselectedwavelengths.Todate,the

industry has demonstrated themegapixel single band LWIR

QWIP FPA, simultaneous dual-band megapixel QWIP FPA,

640x512 formatspatiallyseparated four-bandFPA,aswell

asthedevelopmentofasuper-pixelQWIPFPAforanimaging

multiple wave band temperature sensor. The next bound of

development is expected to focus on harnessing the QWIP

FPAtechnologyforacompact7.5µmto12µmhyperspectral

IRimager.

ShortwaveInfraredSensors

In recent years, SWIR has grown to become a viable low

light imagingoptiontoovercomeenvironmentaleffectssuch

as haze, fog and dust for mission scenarios, where target

detailsarehighlyvaluedbutdonotshowuponMWIRorLWIR

sensorsduetonon-representationbytemperaturedifferences.

ASWIRsystemcantypicallyprovidebetterspatialresolution

thanasimilarclassMWIRorLWIRsystem,enhancingtarget

recognition and identificationwhich are critical functions on

thebattlefieldtominimisecollateraldamage.Whenoperated

at night, SWIR can also take advantage of an atmospheric

phenomenon called night sky radiance or night glow,which

emitsfivetoseventimesmoreilluminationthanstarlight,and

nearly all in the SWIR wavelengths. These wavelengths are

relativelycovertastheyareundetectablebyvisiblespectrum

cameras, II-basednightvisiondevicesaswellasMWIRand

LWIR cameras. In addition, SWIR also allows surveillance

throughglasswindowswhichMWIRandLWIRcamerasare

unabletopenetrate.

ImageFusionandEnhancementTechniques

Recent global geo-political activities have escalated the

urgent need to deploy multiple imaging sensors operating

across different spectral bands to provide timely anomaly

detection. Surveillance agencies now require situation

pictures from multiple sensors to reach their command

centres simultaneously. The use of effective image fusion

techniques would: (a) maximise the amount of relevant

informationthatisdeliveredtotheoperator;(b)cutdownthe

timespentonirrelevantdetails(e.g.falsealarms),uncertainty

and redundancy in the output; (c) prevent the occurrence

of artefacts or inconsistencies in the fused image; and

(d)suppress irrelevant featuressuchasdistortioncausedby

noisefoundinthesourceimages.Inaddition,fusedimagerythat

optimallyagreeswithhumancognitionprocessesallows the

humanoperatortograspthegistofthedisplayedinformation

quicklyandexecuteefficientandeffectiveresponsesfortime-

critical surveillancemissions.over the years,multiple smart

processing and target tracking algorithms, including edge

detection,autonomousvideomotiondetectionand tracking,

havebeendevelopedandintegratedwiththecentraldisplay

interfaceofmultiplesensorfeeds.Thishasgreatlyenhanced

collectivesurveillanceefforts.

Fused imagery has traditionally been represented in grey or

monochromatictones,duepartlytotheoutputdisplaysofthe

IRimagingsensors.Whilethehumaneyecanonlydistinguish

about100shadesofgreyatany instant, IR imagingsensors

produced in the industry today can discriminate between

severalthousandsofcolours.Theuseofcolour imagesmay

improve feature contrast and reduce visual clutter, enabling

better scene segmentation, object detection and depth

perception. This will also yield a more complete mental

representationoftheperceivedsceneforenhancedsituational

awareness.


Researchinthisfieldisongoing.Intime,thesedevelopments

will mature and systems could be deployed in the field for

improvedwideareasituationawareness.Thehigherprobability

of targetdetection, recognitionand identificationwouldalso

drivetheoptimisationofautomaticsurveillancesystemsand

alleviatemanpowerconstraintsfacedbynationalsecurityand

defenceentitiesregionallyandglobally.

CONCLUSION

Eo sensors play a critical role in detecting and identifying

targets in modern day battlefields. However, they can be

affectedbyenvironmentalconditions.Inourlocalenvironment,

the presence of water vapour due to high humidity, rainfall,

cloudsorfog,canseverelyimpactthesensors’performance.

Itisthususefultounderstandtheoperationalrequirementsof

the sensors and the environmental conditions inwhich they

operate, inorder to select themostoptimumsensor for the

situation.

REFERENCES

Anon. (1993). Protecting visibility in national parks and

wildernessareas.NationalResearchCouncil,Washington,DC.

Bernard,E.,Riviere,N.,Renaudat,M.,guiset,P.,Pealat,M.,

& Zenou, E. (2013). Experiments and models of active and

thermalimagingunderbadweatherconditions.Proceedingsof

SPIE,Electro-OpticalRemoteSensing,PhotonicTechnologies,

andApplicationsVII;andMilitaryApplicationsinHyperspectral

Imaging and High Spatial Resolution Sensing, 8897. doi:

10.1117/12.2028978

Chen,C.C. (1975).Attenuation of electromagnetic radiation

byhaze,fog,clouds,andrain.(RANDReportNo.R-1694-PR)

Retrieved from http://www.rand.org/content/dam/rand/pubs/

reports/2006/R1694.pdf

Eismann,M.T.,Schwartz,C.R.,Cederquist,J.N.,Hackwell,

J.A.,&Huppi,R.J. (1996).Comparisonof infrared imaging

hyperspectralsensorsformilitarytargetdetectionapplications.

Proceedings of SPIE Imaging Spectrometry II, 2891(91).

doi:10.1117/12.258056

Leong,S.C.,&Foo,y.C. (2007,December).Singaporerain

ratedistributions.6thInternationalConferenceonInformation,

Communications & Signal Processing, 2007. doi: 10.1109/

ICICS.2007.4449534

Li,W. J., Shao, L.y., & Buseck, P. R. (2010). Haze types in

Beijing and the influence of agricultural biomass burning.

AtmosphericChemistryandPhysics,10(17),8119–8130.doi:

10.5194/acp-10-8119-2010

Malm, W. C. (1999). Introduction to Visibility. National Park

Service and Colorado State Institute for Research on the

Atmosphere,FortCollins,Colorado.

Renhorn, I. et al. (2013). Hyperspectral reconnaissance in

urbanenvironment.ProceedingsofSPIE,InfraredTechnology

andApplicationsXXXIX,8704.doi:10.1117/12.2019348

See,S.W.,Balasuhramanian,R.,&Wang,W.(2006).Astudy

of the physical, chemical, and optical properties of ambient

aerosolparticles inSoutheastAsiaduringhazyandnonhazy

days. Journal of Geophysical Research Atmospheres,

111(D10S08).doi:10.1029/2005JD006180

Ting, D. Z., Soibel, A., Keo, S. A., Rafol, S. B.,Mumolo, J.

M., Liu, J. K., … gunapala, S. D. (2014). Development of

quantumwell,quantumdot,and type II superlattice infrared

photodetectors. Journal of Applied Remote Sensing, 8(1).

doi:10.1117/1.JRS.8.084998

Wolcott. (2013). File:Haze obscuring Marine Parade.jpg. In

WikimediaCommons.RetrievedMarch10,2015,fromhttp://

commons.wikimedia.org/wiki/File:Haze_obscuring_Marine_

Parade.JPg#mediaviewer/File:Haze_obscuring_Marine_

Parade.jpg

Wolcott. (2013). File:Marine Parade Road (2).jpg. In

WikimediaCommons.RetrievedfromMarch10,2015,http://

commons.wikimedia.org/wiki/File:Marine_Parade_Road_(2).

JPg#mediaviewer/File:Marine_Parade_Road_(2).jpg



BIOGRAPHY

LEE Cheow Gim is a Project Lead

(AdvancedSystems)whomanageselectro-

optical(Eo)projectsfortheSingaporeArmy.

HegraduatedwithaBachelorofEngineering

(Electrical and Electronic Engineering)

degree from Nanyang Technological

university(NTu)in2003.

EE Kok Tiong is a Project Manager

(Advanced Systems) who manages Eo

projectsfortheRepublicofSingaporeNavy.

HegraduatedwithaBachelorofEngineering

(Electrical Engineering) degree from the

National university of Singapore (NuS) in

2002 and a Master of Science degree in

ElectricalEngineering(ComputerNetworks)

fromtheuniversityofSouthernCalifornia,uSA,in2008.

HENGYinghui Elizabeth is a Programme

Manager (Advanced Systems) who

overseesEoprojectsacrosstheSingapore

Armed Forces. She graduated with a

Bachelor of Engineering (Electrical and

Electronic Engineering) degree from NTu

in 2003. She also obtained a Master of

Science(ElectricalEngineering)degreefrom

theNaval Postgraduate School, uSA, and aMaster of Science

(DefenceTechnologyandSystems)degreefromTemasekDefence

SystemsInstitutein2013.




SAFETyMANAgEMENToFNATIoNALDAyPARADEFIREWoRKSDISPLAy

INTRODUCTION

SIMGimYoung,LEEChungKiat,OEISuCheok,ME5ONGWoeiLeng

ABSTRACT

ThefireworksdisplayisahighlightofeveryNationalDayParade(NDP)inSingapore.Thedriveforamorespectacularandintimatefireworksexperience increases the importanceplacedon thesafetyofperformersandspectators.Thisarticledescribesthechallengesinmanagingthesafetyoffireworks.Itstartsbyintroducingthecharacteristicsoffireworksanditssafetymanagementapproaches,followedbyanexplanationofhowsafetyisaddressedthroughthefireworkslifecycle–fromproductdesigntotransportation,storage,installation,initiationanddisposal.Finally,thearticlesharesinnovativesolutionsused tocontrol the safetydistanceof fireworks.Theeventualoutcomeof theseapproaches is a safeandspectacularfireworksdisplayforNDP.

Keywords:fireworks,safety,NationalDayParade,NDP

Figure1.NDPFireworksDisplay,2014


on 9 August each year, Singaporeans come together to

celebratethenation’sindependence.TheNationalDayParade

(NDP)featuresmassperformances,aceremonialparadeand

multimedia displays that depict Singapore’s cultures and

values. The fireworks display is a highlight of theNDP (see

Figure1).

At everyNDP, a FireworksCommittee comprisingpersonnel

from theSingaporeArmedForcesAmmunitionCommand is

formedunder theNDPExecutiveCommittee toorganise the

fireworks display. This committee specifies the performance

andsafety requirements tobe implementedby thefireworks

contractor.

While the displays are highly entertaining, fireworks are

explosives and need to be treated with the appropriate

safetyprotocol.Assuch, theFireworksSystemSafetyTeam

comprising members from the Fireworks Committee and

DSTA’s armament safety specialistswork together to review

and ensure fireworks safety duringNDPs. The team adopts

a systemsafetyperspective1 toassess thehazardsof each

fireworkactivityandrecommendssafetymeasurestominimise

therisks.Thisconceptwasinitiatedintheearly1960sinthe

aerospace industryand isaneffectivewayofanalysingand

managingsafetyrisksholistically.

FUNDAMENTALS OF FIREWORKS

Fireworkswere invented inChina in the seventhcentury. Its

applications have since spread to other parts of the world

andhavebeenintegralinmanycelebrations.Despiteitslong

history, fundamentals of fireworks displays have remained

largely thesameover time.Aerialshells,mines,cometsand

variantsformthemaindisplaycomponents.Fireworkscanbe

classifiedbroadlyintotwomaingroups.

The first group, aerial shells, is builtwith a time fuse, lifting

charge, bursting charge and pyrotechnic chemicals. Aerial

shells are launched from mortar tubes into the sky using

gunpowderasaliftingchargebeforeburstingintodisplaysof

brilliant andcolourful lights.Thebrilliant coloursarecreated

from the combustion of metallic powders. The appearance,

component and effects of an aerial shell are illustrated in

Figure2.

Figure2.Aerialshell

AerialShell CrossSection Effects


The second group comprises only lifting charge and

pyrotechnic chemicals. The effects are ejected from the

ground upon ignition. Examples are comets and mines as

showninFigure3.Bothtypesoffireworkscanbecustomised

toproducedifferentcolours,patternsandbrilliance.

CHALLENGES

The NDP fireworks display is the largest fireworks event

in Singapore and an important component of the parade’s

choreography.Thereareseveralrehearsals leadinguptothe

final event, and anymishaps would have significant impact

duetothefireworks’proximitytoperformersandspectators.

Therefore, the safety review of NDP fireworks display is

important to ensure safety of the public, performers and

fireworkstechnicianswhilemeetingthedemandsoftheparade.

Thenumerousstakeholders, regulationsandsiteconstraints

(includingweatherconditions)addtothecomplexity.Fromthe

start,theteamaddressestheintrinsicsafetyofthefireworks

productanditscompliancewithregulatoryrequirements.The

teamreviewstheindustrysafetypracticesconstantlyduringthe

preparation,setupanddisplayofthefireworks.Theestablished

safety codes of theUSNational Fire Protection Association

for Display of Fireworks and British Pyrotechnic Association

serveasbaselinesafetyrequirements.TheSingaporeArmed

Forces(SAF)andDSTAalsoengageestablishedpractitioners

to provide training to reinforce understanding and ensure

alignmentwithbestpractices.

Figure3.Cometsandmines

CometsandMines EffectsofComet EffectsofMine

ENSURING PRODUCT SAFETY

Fireworks must be designed and constructed to be safe

duringhandling,transportationanduse.InMay2000,afatal

explosion occurred during a fireworks display at Bray Park,

Australia.Threegroundfireworksexplodedand ruptured the

steellaunchertube,leadingtoonefatalityandseveninjuries.

Theinvestigationconcludedthatthefireworksmalfunctioned.

Theexplosioneffectwasworsenedbyitsconfinementwithin

thesteeltubes.

The teamensuresproductsafetybysourcingfireworks from

reputable manufacturers with strong safety track records

of supplying fireworks to major international events. The

team also requires themanufacturer to be certified by their

local authority and registeredwith auSbased independent

auditor – the American Fireworks Standards Laboratory

(American Fireworks Standards Laboratory, 2011). The

manufacturer must also be ISo 9001 Quality Management

System compliant. Furthermore, the team conducts factory

assessmentstoobservetheproductionprocessandtestthe

productperformanceandsafetyfeaturesofthefireworks.


HAZARD CLASSIFICATION OF FIREWORKS

In 2000, the Netherlands had a major fireworks accident

within the city of Enschede. The fireworks explosions were

equivalentto4000kgto5000kgofhighexplosives2.Itresulted

indamageamountingtomorethan€450million.Thisincident

was a stark reminder that fireworks could behave like high

explosives,andresultedinareviewofthehazardclassification

offireworks.Theuseofcorrecthazardclassificationenables

appropriate safety protocols to be enforced during storage

andtransportation.In2005,theunitedNations(uN)published

amethod todetermine thehazardclassificationoffireworks

basedonchemicalcomposition.

The fireworks classification system assigns the hazard

categoryoffireworksbasedontwokeycriteria:thediameter

ofthefireworksandtheproportionofflashcompositionused

inthefireworks.Flashcompositionisamixtureofpyrotechnics

chemicals that reactsmorevigorously toproduceabursting

orexplosioneffect.Ittypicallyconsistsofanoxidiser,anon-

metallic fuel and ametallic fuel. An excessive proportion of

flashcompositioninfireworkswillresultinaviolentexplosionin

theeventofanaccident.Thediameteroffireworksdetermines

the mass of pyrotechnics chemical that will combust

instantaneously. As the diameter increases, the combustion

effect becomesgreater and can increase the likelihoodof a

violentexplosion.

Figure4.Bargeforfiringaerialfireworks,2014

usingtheuNfireworksclassificationsystem,theteamreviewed

the chemical composition of all fireworks used in NDP and

identified thosewhichcanpotentiallyexplode if anaccident

occurred during storage or transportation (Russell, 2009).

These items were isolated and stored with sufficient safety

distancestothesurroundingsites.Theywerealsotransported

separatelyandprepared topreventsympatheticexplosions3

shouldanaccidentoccur.Atthefiringsite,theywereinstalled

into the launching tubes at the earliest opportunity so that

accidental ignition would launch the fireworks into the sky

instead of causing an explosion of stacked fireworks (see

Figure4).

SAFE DEPLOYMENT

Fireworks are stored in licensed explosive storehouses in

Singapore.Beforeeachdisplay, thefireworksareunpacked,

inspectedandmovedtothefiringlocations.Thefiringcircuits

aretheninstalledandtested.SinceNDP2011,fireworkshave

been launched from a barge inMarina Bay and around the

floating platform. There were also fireworks launched from

performers’personalequipmentlikemotorcyclesandtorches.

Figure5illustratesthesettingupofaerialshellsatthebarge.

Thesurroundingsarecheckedaftereachdisplayforfireworks

thathavefailedtolaunchordroppedprematurely.Thecause

of each defect is investigated and the faulty fireworks are

disposedof.



SPECTATORS SAFETY

Fireworks produce debris that could hurt people within the

vicinity(seeFigure6).

There could also be fallout hazards from malfunctioning

fireworkssuchas theburstingofaerialshellson theground

aftertheyfailtoigniteintheairandtheimpropermountingof

fireworksleadingtothewrongorientationwhenfired.

Figure5.Workflowofaerialshellsatbarge

Figure6.Examplesofdebrisfromfireworks

CardboardBaseofShell PlasticWhistlingComponent CardboardDebrisofComet


of fireworks shells (see Figure 8). The wind conditions are

monitoredinrealtimeandthecommitteecaninhibitselected

fireworks if thewindspeedanddirectionthreatentoexceed

the safe limits. This ensures that fireworks displays remain

safe for the performers and spectators around the barge at

alltimes.

In addition, the team ensures that all materials used in the

constructionof theaerial shell arecombustible.Mostof the

materialwillbeburnedtoashesduringtheburstingoftheaerial

shell,minimising the amount of debris falling to theground.

Theinstallationandpositioningoffireworksmortartubesare

alsoscrutinisedtomitigatethehazardoflaunchingtheshells

intothespectators.

Figure7.Possiblefalloutdistanceatvariouswindspeeds

The team reviewed internationalbestpractices todetermine

the optimal safety distance and adopted the uK approach

whichcalculatesthesafetydistancebasedonvariousfactors

at each fireworks display scenario (Smith, 2011). The aerial

shellsafetydistanceconsidersthesizeoffireworks,firingangle

andwindconditions.Thesizeofanaerialshelldeterminesthe

launchvelocityandair-burstdiameter.Thefiringangle,wind

directionandwindspeedwoulddeterminethetrajectoryofthe

aerialshellduringtheinitialflightandthedispersionofdebris

aftertheair-burst.

Theteamalsodevelopedawindchart tohelptheFireworks

Committee address the impact of different wind conditions

on thesafetydistance (seeFigure7).Thewindchartshows

themaximumwindspeedineachdirectionfordifferentsizes



PERFORMERS SAFETY

Fireworksareusedbyperformersinseveralsegmentsofthe

NDPaspyrotechnicprops.In2012,thesesegmentsincluded

the Singapore Soka Association performance, SAF Military

Police precision drills and motorbikes ride-past. To ensure

performers’ safety, the team participated actively in the

choreographies so that the fundamental safety principle of

maintainingsafetydistanceisintegratedintotheperformances.

Forperformancesthatrequirehand-heldpyrotechnictorches,

fireworksthatdonotigniteclothingareused.Theperformers

are equipped with protective eyewear, headgear and fire-

resistant clothing to further mitigate the risk of burns. The

team also worked with the show choreographer to ensure

that the amount of fireworks required to meet performance

requirementswaskepttoaminimum.

Thepyrotechnicstorchwasdesignedtoensurethatitsparts

donothurtperformers.Thefiringcircuitforthepyrotechnics

torch incorporatedmaster and safety switches. This design

required theperformer tomake twosimultaneousactions to

ignitethefireworkseffectsoastoavoidaccidentalignition.

SYSTEM SAFETY APPROACH

The team took a system safety approach to ensure and

enhance safety in dealingwith the fireworks effectively. The

priority was to use safe configurations (through appropriate

safetydesignprotocols)beforeprovidingprotectivedevices,

warning devices and relying on procedures. Some of the

considerationstakenduringtheNDPillustratethisapproach.

a) Display Design and Configuration – During the grand

finalewhichtakesplaceattheplatformandspectatorstand,

mines and comets are fired away from spectators and

performers.onlyfireworkswithouthazardousdebrisareused

onstage.Performersareallowedtouseonlyfireworkswhich

produceharmlesssparks.

b) Protection-Performersputongogglesandfire-resistant

costumes to furthermitigate the risksofburns.Partsof the

stagewhichareaffectedbysmoulderingdebrisareconstructed

withfire-resistantmaterials.

Figure8.WindchartaroundhighlevelbargeinMarinaBay,2014


c) Warning - A monitoring device is used to measure the

wind speed and direction throughout the show.Whenwind

conditionsexceedsafelimits,thefiringofselectaerialshells

wouldbecurbed.

d) Procedures - Fireworks safety distances are integral to

the choreography of the show. The control tower monitors

performers’ complianceandcanceasefiring in theeventof

deviation.

The four levels of risk mitigation are standard observations

ofthesystemsafetyapproach.

RISK ENDORSEMENT AND ACCEPTANCE

AMinistryofDefenceSafetyBoardassessestherisksposed

by NDP fireworks in accordance with the risk management

frameworkestablishedforNDPfireworksdisplay.Theresidual

risks,whicharereducedtoaslowasreasonablypracticable,

are then accepted by the NDP Executive Committee.

Subsequently, the teammonitors the fireworks performance

duringeachrehearsalandontheactualparadetoensurethat

thecontrolmeasuresareeffective.

CONCLUSION

The team recognises the benefits of the systematic hazard

identification and risk management approach in managing

the complex demands of the NDP fireworks display. It has

adaptedsafetyknowledgeonmilitaryexplosivesforfireworks

displaysuccessfully.Thesafetycontrols imposedonmilitary

explosives are usedwhen applicable to enhance the safety

ofstorage, transportation,preparation,firinganddisposalof

commercialfireworks.

The system safety approach enables the team to identify

possible hazards that are beyond product safety and

regulations. It allows the team to prescribe measures to

minimise risks to the public, performers and fireworks

technicians. This ensures safe fireworks displays during all

NDPs,fromrehearsalstothegrandfinaleonNationalDay.

ACKNOWLEDGEMENTS

TheauthorswouldliketothankMryenChongLian(retiredin

2014),ME6oliverLanChiWai,ME6AdrianLimBengBoon

and ME6 Cheong Heng Wan for contributing to the safety

practicesforNDP2012,NDP2013andNDP2014.

REFERENCES

American Fireworks Standards Laboratory. (2011). AFSL

standardsfordisplayfireworks.Bethesda,MD:Author.

Russell, M. S. (2009). The chemistry of fireworks (2nd ed.).

Cambridge,uK:RoyalSocietyofChemistry.

Smith, T. (2011). Firework displays: explosive entertainment.

uS:ChemicalPublishingCompany.

ENDNOTES

1 Systemsafetyperspectivemeansthesafetyreviewofthe

interfacesamongthefireworksproducts,displaysites,display

operators, equipment, installation, weather, NDP performers

andthegeneralpublic.

2 Highexplosivesaresubstancesormixtureof substances

whichcandetonateundernormalconditions.

3 Sympatheticexplosion is thesimultaneous initiationofan

explosivechargebyanearbyexplosion.



ME5 ONG Woei Leng is one of the

Commanding officers of SAF Ammunition

Command. He is responsible for the safe

storage and maintenance of ammunitions

kept in the depot as well as the daily

operationsofthedepot.Hewaspreviously

astaffofficerintheExplosiveSafetyBranch

in the SAFAmmunitionCommand. As the

DeputyChiefSafetyofficerandChiefSafetyofficerforNDP2011

and 2012 respectively,Woei Leng ensured the safe conduct of

thefireworksdisplayduringtheshows. InNDP2014,heserved

astheDeputyChairmanoftheFireworksCommittee.WoeiLeng

graduatedwithaBachelorofArts (Psychology)degree fromthe

EdithCowanuniversity,Australia,in2010.

BIOGRAPHY

SIM Gim Young is a System Manager

(Systems Management) managing the

operationsandsupport for explosivesand

pyrotechnicsintheSingaporeArmedForces

(SAF). In 2013and2014, he led theDSTA

team supporting the National Day Parade

(NDP) Fireworks Committee in assessing

the safety of fireworks display. He also

conducts quantitative risk assessments on explosive sites and

advises on the riskmitigationmeasures. gim young graduated

withaBachelorofEngineering (MechanicalEngineering)degree

fromNanyangTechnologicaluniversityin2008.

LEEChungKiatisHeadExplosivesSafety

(Systems Management). He is a licensed

authorityonmilitaryexplosivefacilitiesand

advisestheSAFonexplosivesstorageand

transport safety. He is also the Chairman

of the Explosives Safety Technical Sub-

CommitteeofExplosivesFireandChemical

Safety Committee. Chung Kiat graduated

withaMasterofScience(ExplosiveordnanceEngineering)degree

fromCranfielduniversity,uKin2005.

OEISuCheokisaSeniorPrincipalEngineer

(Systems Management). He develops and

implementsthesafetymanagementsystem

for DSTA and the Ministry of Defence

(MINDEF).Healso supports systemsafety

analysis efforts to enhance safety through

riskmitigation.SuCheokextendshissafety

management competence beyond DSTA

inhis rolesas secretariat ofMINDEF’sWeaponSystemsSafety

Advisory Board as well as an executive committee member of

the InternationalSystemSafetySociety (SingaporeChapter).He

graduatedwithaBachelorofEngineering(ChemicalEngineering)

degreefromtheNationaluniversityofSingaporein1985.




PRoTECTIoNANDRESILIENCyFoRSINgAPoRE’SCRITICALINFRASTRuCTuRES

INTRODUCTION

AfterSingaporegainedindependencein1965,itwasnecessary

tobuilduplocalprotectivedesigncapabilitiesquicklyforthe

development of key installations, defence infrastructure and

facilities. Early protective designmethodologieswere based

onprotectionagainstwell-prescribedthreats.Thesebuilding

designs were standardised and often replicated for greater

developmental efficiency. The building design philosophy

was tofirstspecify thedesign-basis threat–comprising the

weaponandstand-off,andthentodesignthebuildingbased

onspecificprotectioncriteria.Thesecriteriawereoftenrelated

to specific building responses to weapons threats, with the

assumptionthatallcriticalcontentswithinthebuildingwould

havesimilardamagethresholds.

Whilethisdesignphilosophywasadequateinthepast,society

hasseenrapidchanges,especiallyoverthelastthreedecades.

ONGKweeSiangSteve,CHONGOiYinKaren,SEEThongHwee

ABSTRACT

Theapproach todesigningcritical infrastructureso that theyareprotectedagainstdiscreteandwell-defined threats isgenerallywell understood.However, in the faceof asymmetrical threats and vague terrorist intentions, suchprotectivedesignapproachesareoflimiteduse.Theincreasingconnectivityandcomplexitiesofmodernsocietycancompoundtheproblem,thuscausingunintendedconsequences.Itisthereforenecessarytorethinkhowcriticalinfrastructuresshouldbesecured.

ThisarticledrawsfromDSTA’sexperienceindesigningcriticalinfrastructuresfortheMinistryofDefenceandtheSingaporeArmedForces.Itillustrateshowprotectionandresiliencycanbebalancedtoimprovethesurvivabilityofcriticalinfrastructures,taking into account system connectivity, vulnerabilities and themeans to enhance recovery. Diagrams are provided toillustrate ways to integrate R&D and findings from international collaborations in Protective Technology into designingforresiliency.Examplesofnumericalsimulationsandexplosivetestsarealsousedtodemonstratethenecessaryvigourneededintestingassumptions,validatingconceptsanddevelopinganimplementablesolution.Thisarticleemphasisesthatthoroughandresponsibleprotectivetechnologyworkmustbenestedwithinrealistictestsandrelevantexperience.

Keywords:protectionandresiliency,criticalinfrastructures,connectivity,survivability,protectivedesign

Advances in technology have resulted in globalisation and

increasedconnectivitythathavealsochangedthethreatspace.

These new realities call for a reviewof thewaySingapore’s

criticalinfrastructuresareprotected.

THE RISE IN MODERN SYSTEMS COMPLEXITY

The advent of the computer sparked rapid advances in

technology, enabling product research, development and

prototyping within a virtual environment. This reduced

developmentaltimeandcostsgreatly.Coupledwiththegrowth

of the Internet, the development ofwireless and broadband

technologiescatalysedthegrowthofinformationtechnology,

expanding network access. This has resulted in higher

demandsforinformationexchangeanddataconnectivity.


Theneedtocompetegloballyfurtherdrovethedevelopment

ofinterconnectedinfrastructuresystemstomeetproductivity

goals.Mostsystemstodayaredesignedto integrateasone

networktodelivercapabilitiesandhavebecomemorecomplex

as a result. This interconnectedness alsomeans that failure

in one component can result in cascading failures in other

systems clusters. New threats, such as cyber attacks, have

alsosproutedandgrown.Againstthisbackdrop,newrealities

thatchallengethewaycritical infrastructuresaretraditionally

protected have emerged. Where people, critical equipment

andfunctionswereoncehoused indiscretecritical facilities,

theyarenowhousedinconnectednetworksoffacilities.

NEW REALITIES - CHANGING THREATS AND EVOLVING NETWORKS

WhatWasDesignedinthePastMayNotBeRelevantToday…ThreatsHaveChangedandOften,WhatCountsareNetworksofThingsRatherThanStandaloneFacilities

The threats that protective buildings and infrastructurewere

designedagainstinthepasthavechanged.Theearlytypesof

weaponscompriseddifferentcategoriesofartilleryand‘dumb’

weaponsthatwereairdropped.Thismethodwasinaccurate

and had limited penetration capabilities and range. Modern

weaponrynowrangesfromguidedweaponsthatcanbefired

fromlongerdistances,tomasssaturationthreatsfromrockets,

artilleryandmissiles.Warheadtechnologyhasadvancedwith

morepowerfulexplosivesaswellasdifferentkillmechanisms

suchasshapedcharges,runwaydenialrounds,fragmentation

rounds and thermobaric charges. Fuse technology has also

progressed to facilitate the development of penetrating

warheads. These weapons of enhanced capabilities can be

developedfaster,makingitharderforprotectiveinfrastructures

to keep up with commensurate protection levels without

overwhelming costs and disruptions to operations. Adding

to this ever-evolving andwide spectrumofmodernweapon

threatsistheneedforcriticalfunctionstooperateinnetworks

ofbuildingsandinfrastructure.Thisgivesrisetothequestion

ofwhether the traditional approach to protective designwill

becomeobsoleteinthefuture.

Increasingly,ItIsWorldwideConnectivityThatEmboldensAdversaries

Beyondspurringmilitaryweaponstechnologydevelopments,

worldwideconnectivityhasincreasinglyemboldenedterrorist

activities,spinningoffemergentthreats.Terrorismhasevolved

over the years, from onewhere therewas little connectivity

andwhereknowledgeinbombmakingwasconfinedtoafew,

toahighlyconnectedenvironmentwheredecentralised,non-

hierarchal leadershipscollaborate, tapand share knowledge

onlineeasily.Furthermore,suchdecentralisedbutconnected

terrorist networks have become harder to detect. Terrorist

organisationshavethusturnedthethreatofincreasedexposure

due to the use of the Internet and telecommunications into

opportunitiestobetterthemselvesandtheiroperations.

UnintendedConsequencesCanAriseFromEver-EvolvingThreats

Threats and their effects have becomemore unpredictable.

Theresultingcomplexconsequencesmaynotbeanticipated

during the design phase. The September 11 attacks on the

World Trade Centre (WTC) and the Pentagon in 2001 used

civilianaircraftasaweapon.WhiletheWTCtwintowerswere

designedtowithstandaircraft impactanddidso initially, the

eventualcollapseofthetowersarosefromthelargemagnitude

of aviation fuel fires that weakened the building structure.

Thus,thedesignofinfrastructurehastoconsiderawiderange

ofpotentialthreats.

TheRangeofPotentialTargetsCanSpikeDramatically

Traditionally, the focus has been on the protection of key

installations and not on soft targets. The Bali bombings in

2002, aswell as the JWMarriot andRitzCarltonbombings

inJakarta in2009,allshowedthatsoft targetsareattractive

to terrorists. The mode of operation in the Bali bombings

comprised multiple attacks with the first bomb occurring

inside a night club, and subsequent car bombs outside to

cause maximum death and injury to fleeing victims. The

JemaahIslamiyaharrestsinSingaporefurtherdrovehomethe

pointofpotentialterroristattacksonthehomefront,withsoft

targetlistsextendingtoincludeselectedtrainstations.Asone

considers these recent terrorist incidents, the listof facilities

toprotectcanspikedramatically,drainingatunprecedented

ratesthealreadylimitedresourcesfordefenceandsecurity.


ModernBuilt-upEnvironmentsarePronetoAdverseCollateralDamageFarBeyondtheImmediateVicinityofaBlast

Beyond connectivity enabled via computer networks,

connectivity arising from the need to build and operate

in dense clusters can make it difficult to anticipate where

threatscanemergefrom,andwhichthreatsshouldprotective

design be applied to. Attacks against targets can result in

collateraldamagewithwidespreadimpact.Thebombattack

inSeptember2004ontheAustralianEmbassyinJakartawas

oneagainstarelativelyhardtarget.Itresultedin11fatalitiesin

theimmediatevicinity.Whiletheembassystructureremained

intact, windows in adjacent buildings up to 500m away

shattered,injuringmorepeople.

Thedirectandindirecteffectsofablastdetonatinginatypical

urbanstreetare illustrated inFigure1.Theextentof injuries

due to primary and secondary effects in this hypothetical

scenariowasassessedusingDSTA’s in-houseconsequence

Figure1.Hazardareaunderprimaryandsecondaryblasteffects


analysistool.Fromthisillustration,itisevidentthatsecondary

effectsofblastlikeglazingdamageordebrishazardsinabuilt-

up environment resulted in human casualties in zones that

extendedfarbeyondtheimmediatevicinityofablast.

TheDiverseSpectrumofOperatingNetworksandConstituentsRequireVaryingTailoredProtection

In the past, systems were less automated with low

interconnectivity to operating networks beyond protected

facilities.Today,relativelysmallincidentscanhavewidespread

impactintermsofconnectivityandfunction.Thefireincidentat

theBukitPanjangExchangeon9october2013resultedinthe

breakdownoftelecommunicationservicesinthenorthernand

westernpartsofSingapore.Thisaffectedtelecommunications

and broadcast services to 270,000 subscribers, including

residential users, a few government agencies, financial

institutions and businesses. A similar shutdown in critical

communicationssystemssuchasairtrafficcontrolcanresult

inpotentiallywiderconsequences.

ImpactOftenExtendsBeyondInitialDesignBoundaries

Key installations have traditionally been given standalone

protection.However, thepeopleoperatingthese installations

liveaspartof thewidercommunity.TheoutbreakofSevere

AcuteRespiratorySyndrome(SARS) in2003hasshownthat

threatssuchaspandemicscandisruptallsectors,rangingfrom

airtraveltohealthservices.Likewise,whendesigningagainst

threats like those from bombs, one has to look beyond the

projectboundariestoensurethattheplacementofprotected

developmentsinanareadoesnot‘passon’threateffectsto

neighbouringareas.

Protective engineering is evolving from a unique auxiliary

capability initially meant only for specialised facilities into a

commonfeature foran infrastructurethat takes intoaccount

the protection of the community that it is a part of. This

demandsnotjustachangeintechnologyoranalyses,butalso

achangeinmindsettolookbeyondone’sowntaskarea.

Toavoidbeingunder-designedinprotectionagainstpotential

threats,radicallydifferentapproachestocriticalinfrastructure

protectionarerequired.

NEW APPROACHES TO CRITICAL INFRASTRUCTURE PROTECTION

Infrastructuresshouldnotonlybeable towithstandattacks,

butalsorecoverafteranattackandresumefunction.Assuch,

itisnecessarytobuildresiliencyintocriticalinfrastructures.

Resiliency is the ability to resume normal operations and

function after an attack. Developing infrastructure resiliency

does not only mean improving the physical protection

of infrastructures to withstand attacks. It also allows the

infrastructuresystemtosustainlimitedextentofdamage,with

recoverysystemsthathavebeenputinplacetoensurereturn

tonormalcywithinashorttime.Abalanceneedstobestruck

between providing full physical hardening and designing to

allowpartialdamagewithswiftsystemrecovery.

Designing a system with resiliency is a prerequisite for the

continued survival of communities after attacks. Systems

designedwithresiliencyhaveparticularattributeswhichenable

communitiestorecoverquicklyfromdisasters.Theseattributes

include the ability to resist, absorb, recover fromand adapt

quickly to disruptions, and to resume system performance.

Somelevelofsystemdamagemaybeacceptable.

Resilientsystemdesignbeginswithanintimateunderstanding

of how a system works as well as how it degrades and

recovers.This,togetherwiththeabilitytodeterminetheexact

levelsofsystemdamagesustained,allowscomponentstobe

enhancedwheretherepairandsystemrecoverycanbedone

withinrequiredtimeframes.

Design for resiliency can be achieved through a right

combination of protective engineering design, system

redundancy,design robustnessandcontingencyplanning to

counterasymmetricalthreatsordisruptions.

In theArt ofWar, ancientmilitary strategistSunTzu,wrote:

“知彼知己，百戰不殆”.Thisistranslatedas“Knowyourenemy

andknowyourself,andahundredbattlescanbefoughtwithout

losing a single one”. In the context of designing protective

infrastructures, knowing your enemy involves understanding

the threat. Knowing yourself involves understanding the

operationalneedsandpotentialweaknesses.

The boundaries of this paradigm can be extended. Apart

from knowing yourself and the enemy, understanding the

interconnectionswithsurroundingelementsisjustasimportant

as it raises awareness of what could potentially go wrong.



Theseformthebackboneofdesigningforresiliencywiththe

preservationofaninfrastructure’scorecapabilityinmind.

WhatEmergesFromtheExtendedParadigm

Protection Concept Development Without Definition of Threat

Itispossibletodesignfacilitiesforprotectionwithoutdefining

a precise threat. This is done by expanding the area of

coveragebeyondtheimmediatefacility,consideringsystems

vulnerabilitiesanddesigningtoincorporatemitigationsystems.

For example, a new annex building next to a cluster of key

buildingsmaybeconstructedandthequestionofhowresilient

or protected should the annex be arises. The base level of

protection and protective detailing can be determined by

examiningtheannexbuildinginrelationtotheexistingcluster,

and the effect of the cluster’s surroundings in introducing

threatstotheannexbuilding.Afterthis,adetailedexamination

of the annex building’s constituents can be carried out and

specificareasfurtherreinforcedifnecessary.

Beyond the design of buildings, the concept of developing

protectionoptionswithoutaprecisethreatcanalsobeapplied

to infrastructurenetworks, includingnetworks forpowerand

fuel.When an engineer designs a power or fuel distribution

network system infrastructure, factors to consider include

thetypeofcriticalfunctionthatthenetworksupportsandthe

environmentthenetworkisoperatingin,asopposedtowaiting

forthedefinitionofthethreat.Aspartofthedesigniterations,

the network design can be scrutinised and vulnerabilities

identified.

Vulnerabilities can include a single-point-of-failure, common

modes failure and areas where even rudimentary forms of

protectiondonotexist.Strategiestoovercomesingle-points-

of-failureinthesystemcanincludetheincorporationofalternate

distributionpathstocriticalnodes,orthephysicalseparation

ofcriticaldistributionnodes.Strategiestoovercomecommon

modefailurescanincludetheuseofindependentbackup.For

fueldistributionnetworks,backupcancomeasalternatefuel

supplyfromfuelbowsers.Inthecontextofpowerdistribution

networks,thiscanbestandalonebackupgenerators.

Customised Protection of Critical System and Equipment

Considering the varying damage tolerance of different

systems,physicalhardeningneedstobecustomisedtomatch

what the buildings contains. In the event that the threat is

biggerordifferentfromwhatwasanticipatedinthehardened

design,orifthecostofhardeningisprohibitive,othermeansto

ensuresystemavailabilityandquickrecoveryareneeded.For

example,thereareseveraloptionstoprotectasatelliteantenna

dishagainstweaponeffects.Tominimisethreatexposure,the

missioncriticalsystemcanbesitedawayfromareasproneto

attacks.Toreducetimeneededforrecovery,mobileantennas

canbeutilised.

Future protected facilities need to move away from mass

produced one-size-fits-all approaches to customised ones

designed not just for the individual facility, but for a larger

networkorcommunitywhichthefacilityisapartof.

Including Time Domain and Usage Pattern Considerations When Designing Critical Infrastructure Protection

Improving resiliency through system design in space alone

may not suffice. operational characteristics such as time

and usage patterns need to be considered. How people

respondplaysanimportantroleinachievingmissionsuccess.

understanding howpeople respond to crises over time and

how usage of infrastructures varies as stages of a crisis

unfoldwill beessential.Buildinghardenedshelters inpublic

undergroundtrainstationsmayprovideprotectiontomasses

of travelling commuters in times of crisis. However, people

in high-rise residential buildingsmay not able to get to the

publicshelterintime.Forthem,individualhouseholdshelters

meet their protection needs better because they can get to

thesheltersquicklyandcancarryonwithotheractivities in

betweenalerts.Thisallowsagreaterlevelofnormalcyevenin

timesoftension,withbenefitsforthepopulationtobeableto

weatherprolongedperiodsof tension incrises.Furthermore,

a shelter servesmultiple uses, including community use for

public shelters and family use for household in peace time.

However,onemustnotoveroptimisedesignsortrytosqueeze

toomanyusagepatternsintoaprotecteddesign.Inthecase

ofhouseholdshelters,iftheshelterisusedonlyasastoreroom

itmightnot fulfil itsoriginal intent toshelter familieswithout

preparationstoemptyoutmassivestores.


6

6

6

6

0 10 20 30 40 50 60 70 80 90 100

1

2

3

4

Extent of Injury

Floo

r Level

Human Injury Profile at 7am

Minor Non-‐Life threatening Life threatening Fatality

53

57

59

60

8

8

8

7

3

3

3

2

15

10

4

2

0 10 20 30 40 50 60 70 80 90 100

1

2

3

4

Extent of Injury

Floo

r Level

Human Injury Profile at 10am


Threat Scenario A large bomb detonates at standoff distance of 10m from the 5-storey office building outline as shown in the diagram. Building Information Height = 30m ; Offset = 25m; Width = 50m ; Span = 50m Total Occupancy = 100pax per floor (total = 400pax in building) Assumed Occupancy = 5%(7am); 100%(10am); 25%(12pm)

14

14

15

15

2

2

2

1

1

1

1

0

3

2

1

0

0 10 20 30 40 50 60 70 80 90 100

1

2

3

4

Extent of Injury

Floo

r Level

Human Injury Profile at 12pm


Figure2.Expectedhumaninjuryprofileoverdifferenttimesoftheday

Checking Complex Usage Interactions in Time and Space for Emergent Vulnerabilities

The probability of threat occurrence and severity of its

consequences can fluctuate over the time-space domain.

Figure 2 illustrates the expected profile of human injury in

anofficebuildingwhena largebombdetonatesat the front

of thebuildingoverdifferent times in theday.Changing the

operational flowwithin the infrastructure facilitycanmitigate

the effect. For example, the consequences of an explosion

atanoperationalfacilitycanbereducedbyvaryingthetimes

whenpeoplemove throughabuilding such that it doesnot

coincide with times when a large bomb may be around. A

thorough understanding of the operational processes over

time is needed. Table-top exercises should be conducted

to simulate operational processes and study how people

andprocessesreact to the introductionofdisruptiveevents.

Thiswouldalsohelpplannersanddesignersappreciatehow

infrastructuresystemsandpeoplerespondtocrises.Realistic

trainingregimeswillfurtherbuildupconfidenceandknow-how

incrisismanagement.

Beyondthefocusonmodellingweaponseffectsonbuildings,

modelling and simulation can be extended to workflow

analysis,andcanenabledesignoptimisationforsurvivability

and resiliency. For facilities where mass congregation of

people or vehicles is expected during operation, modelling

tosimulatehumanandtrafficflowswillprovidecriticalinputs

toplanners,designersandstakeholderson theadequacyof

infrastructure system formission support. ground exercises

are needed to validate planning and design assumptions.

Fromthisunderstanding,anestimateofhowmuchandwhere

protectionandresiliencycanbestbe injected intoabuilding

systemcanbemade.



OPERATIONALISING THE NEW CRITICAL INFRASTRUCTURE PROTECTION DESIGN FRAMEWORK

Asystematicanditerativeapproachtoidentifycrediblethreats

and address the comparative risks and vulnerabilities is


MissionIdentificationandVulnerabilityAssessment

The process of designing critical infrastructures beginswith

knowing the mission and identifying mission critical assets

thatneedtobeprotected.Athreat independentvulnerability

assessment is then conducted to identify single points of

failure, common modes failure, and areas of inadequate

protection.

ConsequenceAssessment

Threatsarethenintroducedandconsequencestofunctionality

and collateral effects are analysed. For a car bomb threat,

parameters of interest include the location of the bomb

in relation to the building. Consequence assessments are

performed using physics-based computational models

Define Mission Statement

Iden0fy Key

Mission Assets

Consequence Assessment

Mi0ga0on Strategy

Implementa0on

Threat-‐Independent Vulnerability Assessment

Review Design Op0ons to improve cost effec0veness of design

Measure of Effec0veness of Design Op0ons

Cost Analysis

Threat/Hazard Assessment

Iden0fy credible threats/hazards through Opera0onal Analysis, modelling and simula0on

Risk-‐based Assessment

Probability of Survivability of System

Impact to opera0on, social, security etc.

Figure3.Vulnerabilityandconsequencesassessmentframework

to derive blast loads on the building. These blast loads are

thenusedtoassesshowthetargetedbuildingrespondsand

the collateral effects on surrounding buildings. DSTA has

conductedalargebodyofresearchworkonexplosioneffects,

structuralresponseandprogressivecollapseincollaboration

withlocalresearchinstitutesandoverseascollaborators.DSTA

hasalsobuiltupcomputationalknow-howtomodelexplosion

effects.Explosivetestsareconductedtoensurethevalidityof

the research outcomes andmodels against realistic threats.

Figure 4 illustrates a collaborative explosive testing effort to

derive blast pressure data for validation against numerical

blastpredictionmodels.

Research outcomes are codified into analysis software

anddesignguides thatcanbeaccessedbyawiderpoolof

engineers.Figure5showsdatafromexplosivetestsconducted

onlocalwindowtypestoensurethatthewindowperformance

isconsistentwithpredictionsfromfastrunningtools.

Mitigation

oncepotentialconsequenceshavebeenassessed,systems

tomitigate vulnerabilities and consequences are developed.

operational analysis tools can be used to quantify the

effectiveness of different design options, thereby facilitating

designoptimisation.


Application

Physical Modelling

Mathematical Modelling

1

Conceptualization

Numerical Simulation Explosive Testing Preliminary Calculations

Model Validation

Acceptable Agreement

Blast Effect Prediction for Consequence Assessment Reality of Interest

Figure4.ValidationofnumericalmodelsthroughcollaborationwithuSCombatingTerrorismTechnicalSupportofficeonurbancanyonexplosivetests

Figure5.Comparisonbetweenwindowresponseexplosivetestresultswithin-houseengineeringtoolsforexplosionconsequenceassessment

observeddamageinthelowhazardrange

PredictionofBlastLoads WindowResponseModule



OptimisingResourcestoYieldCost-EffectiveSolutions

To ensure resource optimisation, cost-benefit analysis is

carried out on various implementable design options. The

relevant stakeholders must strike a balance between the

financial commitment to improve system resiliency and the

acceptanceofresidualriskassociatedwithtimeforattacks.

ClosingTechnologyGaps

DSTA is involved in researchwork toclose technologygaps

in critical infrastructure resiliency. These have evolved from

thetraditionalfocusonhardeninginfrastructurestoresistand

absorb extreme loads, to programmes that facilitate system

resiliencyandrecovery.onesuchresearchprogrammederives

data on the survivability of buried utility networks against

weaponsattack.Thedataisincorporatedintoanoperational

analysistoolthatdeterminestheoverallsystemutilitynetwork

survivabilityusingfaulttreeanalysis.

CONCLUSION

This paper illustrated how achieving a balance between

protectionandresiliencycanimprovetheoverallsurvivability

of critical infrastructures. A critical infrastructure design

framework was also discussed, which involves more effort

spentassessingone’sownvulnerabilitiesandinterconnections

than before. It is also coupled with greater efforts to look

beyondtraditionalprojectboundariesconstantly.

However,protectionofcritical infrastructuresystemscan lag

behind technological advancements and resourcefulness of

adversaries. Therefore, the design of critical infrastructures

needs dogged perseverance and an attention to detail.

Implemented protection concepts should be reviewed

periodically,orruntheriskofobsolescence.Moreoftenthan

not, lapses in protection of a critical infrastructure system

surface only after attack events. The review of protective

conceptsneedstobecarriedoutcollectivelybybothtechnical

and operational communities. Thereafter, the potential for

future upgrades should be incorporated into the protective

designwherepossible.

REFERENCES

AttackbyStratagem.(n.d.).InChineseTextProject.Retrieved

fromhttp://ctext.org/art-of-war/attack-by-stratagem

Corley,g., Hamburger, R., &McAllister, T. (2002). Executive

summary. In T.McAllister (Ed.),World TradeCenter building

performancestudy:datacollection,preliminaryobservations,and

recommendations (pp.1-7).Retrievedfromhttp://www.fema.

gov/media-library-data/20130726-1512-20490-7075/403_

execsum.pdf

Infocomm Development Authority of Singapore. (2014,

May).Fire IncidentatBukitPanjangExchangeon9October

2013. Retrieved from https://www.ida.gov.sg/~/media/Files/

About%20us/Newsroom/Media%20Releases/2014/0506_

CompletesInvestigation/Factsheet_FireIncidentBukitPanjang.

pdf

NationalConsortiumfortheStudyofTerrorismandResponses

to Terrorism. (2004). Incident Summary. InGlobal Terrorism

Database. Retrieved from http://www.start.umd.edu/gtd/

search/IncidentSummary.aspx?gtdid=200409090001


BIOGRAPHY

ONG Kwee Siang Steve is a Manager

(Building and Infrastructure) involved

in the design development and project

managementof building infrastructures for

theMinistryofDefenceand theSingapore

ArmedForces.Heisalsopartofthemulti-

disciplinary DSTA team that carries out

vulnerability assessments and mitigation

studies. Steve graduated with a Bachelor of Engineering (Civil

Engineering) degree with First Class Honours from Nanyang

Technologicaluniversityin2006.HefurtherobtainedaMasterof

Science(ProjectManagement)degreefromtheNationaluniversity

ofSingapore(NuS)in2013.

CHONGOiYinKarenisHeadEngineering

(Building and Infrastructure)who is driving

R&Defforts inProtectiveEngineering.She

has extensive experience in explosive

testing, protective systems design and

blast modelling and analysis. She won

the Defence Technology Prize Team

(Engineering) Award in 1999, 2006, 2007

and2011.KarengraduatedwithaBachelorofScience (Nuclear

Engineering) degreewith FirstClassHonours fromQueenMary

College,universityofLondon,uK,in1986.Shefurtherobtained

aDoctorofPhilosophy(NuclearEngineering)degreefromQueen

MaryandWestfieldCollege,universityofLondon,uK,in1991.

SEE Thong Hwee is Head Capability

Development (Building and Infrastructure)

who oversees building infrastructure

development of joint facilities. He also

providesprotectiveengineeringconsultancy

for critical infrastructures. Thong Hwee

hasplayeda key role in extendingDSTA’s

protective technology capabilities to

Singapore’s homeland security. He was involved in numerous

projects that improved thephysical resiliencyof critical national

infrastructures, such as the national power grid, mass rapid

transportnetwork,variousgovernmentfacilitiesandseveraliconic

buildingdevelopments.ThongHweegraduatedwithaBachelor

ofEngineering(CivilEngineering)degreewithHonoursfromNuS

in 1997. He further obtained a Master of Science (Engineering

Mechanics) with Specialisation in Explosives Engineering, from

theNewMexicoInstituteofMiningandTechnology,uSA,in2002.



NOTES


NOTES


NOTES

Documents

HORIZONS - DSTA