Research Article Replica Node Detection Using Enhanced ...downloads.hindawi.com/journals/jcnc/2016/1620343.pdfby enhancing the Single Hop Detection (SHD) method using the Clonal Selection

Research ArticleReplica Node Detection Using Enhanced Single HopDetection with Clonal Selection Algorithm in MobileWireless Sensor Networks

L S Sindhuja and G Padmavathi

Department of Computer Science Avinashilingam Institute for Home Science and Higher Education for Women UniversityBharathi Park Road Mettupalayam Road Coimbatore Tamil Nadu 641043 India

Correspondence should be addressed to L S Sindhuja sindhujakarthick2011gmailcom

Received 7 July 2015 Revised 12 November 2015 Accepted 19 November 2015

Academic Editor Eduardo da Silva

Copyright copy 2016 L S Sindhuja and G Padmavathi This is an open access article distributed under the Creative CommonsAttribution License which permits unrestricted use distribution and reproduction in any medium provided the original work isproperly cited

Security of Mobile Wireless Sensor Networks is a vital challenge as the sensor nodes are deployed in unattended environment andthey are prone to various attacks One among them is the node replication attack In this the physically insecure nodes are acquiredby the adversary to clone them by having the same identity of the captured node and the adversary deploys an unpredictablenumber of replicas throughout the network Hence replica node detection is an important challenge in Mobile Wireless SensorNetworks Various replica node detection techniques have been proposed to detect these replica nodesThesemethods incur controloverheads and the detection accuracy is low when the replica is selected as a witness nodeThis paper proposes to solve these issuesby enhancing the Single Hop Detection (SHD) method using the Clonal Selection algorithm to detect the clones by selecting theappropriate witness nodes The advantages of the proposed method include (i) increase in the detection ratio (ii) decrease in thecontrol overhead and (iii) increase in throughput The performance of the proposed work is measured using detection ratio falsedetection ratio packet delivery ratio average delay control overheads and throughput The implementation is done using ns-2 toexhibit the actuality of the proposed work

1 Introduction

Wireless Sensor Network (WSN) comprises a group of wire-less sensor nodes that form a communication networkThesenodes collect the sensitive information from the region andsend these contents as a message to the base station where itchecks the data and ID sent by the sensor nodesThese sensornodes are normally low priced hardware components withconstraints on memory size and computation capabilitiesThe Mobile Wireless Sensor Networks are similar to WSNexcept that the sensor nodes are mobile in nature Thevarious applications of Mobile Wireless Sensor Networksinclude robotics transportation system surveillance andtracking Researchers focus to integrate Mobile WirelessSensor Networks into ldquothe Internet of Things (IoT)rdquo [1]However a huge amount of security issue arises in the formofattacks due to lack of hardware support and insecure sensornodes One such attack is the node replication attack

In sensor networks attackers capture and compromisenodes to inject fake data into the network that affect thenetwork communication and operations Such type of attackis known as replica node attack [2ndash5] The adversary cap-tures secret keys from the compromised nodes and spreadsthem as replicas in the network Replicas are reflected ashonest by its neighbors and normal nodes are not awareof replicas present as their neighbors It also colludesand acts as legitimate node that provides firmness to thenetwork

The replica nodes are controlled by the adversary Theproblem is the replica nodes also contain the key that isrequired for secured communication in the network Inaddition to these problems mobility of nodes the collusionof replica and sideway attacks are the main difficulty whiledetecting and controlling these replica nodes When thereplicas are not detected then the network will be open toattackers and the network becomes more vulnerable

Hindawi Publishing CorporationJournal of Computer Networks and CommunicationsVolume 2016 Article ID 1620343 13 pageshttpdxdoiorg10115520161620343

2 Journal of Computer Networks and Communications

The detection of replica node in the Mobile WirelessSensor Networks is a crucial task So far only few detectionschemes have been proposed [6ndash9] Since the adversarydistributes replica nodes everywhere in the network themobility-assisted detection scheme is required to detect thereplicas in the network

In the earlier works mobility assisted technique SingleHop Detection (SHD) was proposed In SHD each nodebroadcasts its location claim to its single hop neighborsand selects the witness node The selected witness nodedetects replicas by performing the verification process As aresult it reduces communication overhead However whenthe replicas collude with each other they select replica as awitness node Hence the detection accuracy is low

Themain objective of the proposedwork is to improve thedetection accuracy by selecting the appropriate witness nodewith reduced overheads during the detection of replica nodesin themobile wireless sensor networks Tomeet the objectiveSHD is enhanced using the Artificial Immune System

Artificial Immune System (AIS) is a branch of ArtificialIntelligence based on the principles of Human Immune Sys-tem It provides various solutions to the real world problemsdue to its characteristic features The characteristic featuresinclude learning ability to the new conditions adaptabilityand distributed nature to the diverse environment limitedresources survivability even in the harsh environmentsThe enhancement of SHD with AIS improves the detectionaccuracy

The contribution of the paper includes the enhance-ment of the SHD method by applying the Clonal Selectionalgorithm for selecting the witness nodes that is anothercontribution Due to this the detection ratio is increasedby selecting the appropriate witness nodes and therebythe replica node detection process incurs minimum controloverheads

This paper is organized as follows In Section 2 variousrelated works towards this technique implemented by differ-ent authors are discussed In Section 3 the system model isdiscussed In Section 4 proposed CSSHD method is brieflyexplained and in Section 5 the performance of the proposedtechnique is compared with the existing technique based oncertain performance metrics and finally Section 6 concludesthe paper

2 Previous Works

Identifying the replicas in mobile wireless sensor networksis an enforcing task under hostile environments The exist-ing node replication attack detection methods are eithercentralized or distributed [6ndash12] The centralized detectionmethod has distinct points where it can fail and also can becaptured by the adversary As a result distributed detectionmethods are proposed in the literature The distributedreplica detection methods are of different types [13] namely

(i) information exchange based method(ii) node meeting based method(iii) mobility assisted based detection method

Of the detection methods only the distributed detectionmethods are discussed below

Yu et al [6 12] in 2013 present the eXtremely EfficientDetection (XED) method This is a distributed detectionalgorithm for mobile networks in which the detection isbased on the information exchanged between the nodes inthe network It detects the replica based on the randomnumber exchanged between each other of the two nodesThedetection capability is degraded when the replicas exchangethe exact random value

Ho et al [7] in 2009 presents the Sequential ProbabilityRatio Test (SPRT) at each periodic time the sensor nodetravels to the new position it signs the claim and sends thisclaim to its neighbors and to the base stationThe base stationcalculates the speed depending on the probability ratio testand matches it with the observed speed If the observedspeed is higher than the computedmaximum speed then thedetected node is a replica node

Conti et al [8] in 2008 introduced a clone detectionmechanism called Simple and Co-operative DistributedDetection (SDD amp CDD) To duplicate a node it must beisolated from the network and then its information must beextractedThis information earns some periodic time and theSimple Distributed Detection [8] uses that time period fordetecting the replica The detection capability of Simple Dis-tributedDetection (SDD) does not providemuch accuracy todetect the replica nodes and thus improved to develop a Co-operative Distributed Detection (CDD) procedureThe CDD[8] use the nodes cooperation to improve the replica nodedetection rate The nodes interchange the information whennodes are in the same communication radius This methodincreases the detection probability

Zhu et al [9] in 2007 proposed an Efficient and Dis-tributed Detection (EDD) a node meeting based distributeddetection method The EDD method computes the numberof encountering times between nodes in a given time periodldquo119879rdquo with higher probability If a node with a higher thresholdvalue is encountered then it is considered as a replicanode

Lou et al [14] in 2012 proposed Single Hop Detection(SHD) method It is a mobility assisted based distributeddetection method In SHD method when a node appears atdifferent neighborhood community replica is detected Thismethod improves the communication overhead

All the above methods discussed are compared andTable 1 presents the comparison

The parameters used by various detection methods forevaluation are the number of claims overheads detectionrate and false detection rates From the literature it isobserved that SHD method is efficient when compared toother detection methods in terms of control overheads butthe detection accuracy is low when the replica is selected as awitness node Hence an attempt is made to enhance the SHDmethod usingArtificial Immune System in terms of detectionaccuracy with minimum number of control overheads Thesystem model of the proposed work is explained in the nextsection

Journal of Computer Networks and Communications 3

Table 1 Comparison of node replication attack detection techniques in Mobile Wireless Sensor Networks

Year Author(s) Technique(s) Replica detectionmechanism Parameters used

2007 Zhu et al [9]Efficient and DistributedDetection(EDD)

When a node exceeds thepredefined threshold valuethen it is detected as replica

Detection accuracydetection timestorage overheadcomputation overhead andcommunication overhead

2008 Yu et al [12] Extremely EfficientDetection (XED)

Based on the random valueexchanged between the twonodes replicas are detected


2008 Conti et al [8]Simple and Co-operativeDistributed Detection(SDD amp CDD)

Replicas are detected basedon the periodic time of thenodes in the network

Detection probabilityfalse positive andfalse negative

2009 Ho et al [5 7 11] Sequential ProbabilityRatio Test (SPRT)

Detects the replica whenthe node appears at distinctlocations exceeding thepredefined velocity

Number of claimsfalse positive andfalse negative

2012 Lou et al [14] Single Hop Detection

When a node appears atdifferent neighborhoodcommunity replica isdetected

Detection time andcommunication overhead

3 System Model

Before the application of the enhanced SHD method fordetecting node replication attack the system model used bythe study must be discussed The system model is furtherdiscussed in terms of the network model and the threatmodel

31 Network Model Mobile Wireless Sensor Network is builtup of several sensor nodes which have distinct identity IDfrom 1 to 119899 It uses symmetric routing that refers to theidentical path of the data transfer between the source and thedestination and vice versa Every node sends beaconmessageperiodically in the format (ID neighbor ID) and supportsequal time interval period Each node in the network isconsidered to be mobile It uses random way point (RWP)mobility model [15ndash17] for the mobility The other mobilitymodels available are RandomMobility model RandomWay-point Mobility model Random Direction Mobility modela boundless simulation area mobility model Gauss-MarkovMobility model probabilistic random walk mobility modeland city sectionmobilitymodel [18] out of which the randomway point mobility model is very important for the problemarea due to the flexibility and the reasonable patterns createdto appear in real-life

Each node knows its own location and aims to reach thedestination point with its velocity in a predefined intervalrandomly in the sensing field When a node reaches thedestination point in a network it remains static for a randomamount of time Each time the node uses the same mobilityrule again For minimizing the complexity in the mobilityeach node has ldquokrdquo average neighbors per each move in the

network The value of ldquokrdquo varies within [1 2 119901] where119901 lt 119873 ldquo119873rdquo is the number of nodes in the Mobile WirelessSensor NetworksThe network uses identity-based public keysystem for data protection during packet transmission [15]Each node stores its own private key and a master public keyThe private key is used to sign claim messages

In consolidation the following are the assumptions madein the creation of a network model

(i) Movement of nodes according to Random WaypointMobility model

(ii) Identity-based public key system that is used toprotect the data

32 Threat Model Nodes in the Mobile Wireless SensorNetworks are not resistant proof An active adversary maycompromise sensor nodes and may use those nodes to createattacks in Mobile Wireless Sensor Networks An adversarycan launch Denial of Service (DOS) attacks by jamming thepath from benign nodes [2] To maximize the effectiveness ofan adversary in the network an adversary can also launch anode attack called a replication attack in which the replicanodes can be launched by accessing the nodersquos legitimatesecurity credentials by compromising the nodes immediatelyat every sensor deployment It is to be noted that an adversarycan use these replica nodes in different patterns in an attemptto frustrate the Single Hop Detection (SHD) The replicanodes may collude with each other that lead to corruptionof the detection protocol at zero delay While running nodereplication attack detection protocol an adversary can causethe routing protocol to selectively jam the path or damage fewnodes in the network


Due to this threat in mobile wireless sensor networkmodel the proposed method Enhanced SHD using ClonalSelection algorithm of AIS (CSSHD) is developed

4 Proposed Methodology

The proposed work is based on the mobility assisted dis-tributed detection of replica node inmobileWSNThe replicadetection method is done by enhancing SHD using ClonalSelection algorithm The selection of witness nodes is basedon the Clonal selection algorithmThe existing SHDmethodis explained below

41 SHD The SHD method is a mobility assisted detectionmethod The SHD method [14] is based on the fact that atany occasion the node ID and private key of a node mustnot occur at other neighborhood communities If it occursthen there must be replicas in the network The nodes thatare present in the neighborhood community are defined bythe list of one-hop neighbor nodes Sensor nodes know theirneighbors for communication

The SHD method is composed of two main phases

(i) Fingerprint claim

(ii) Fingerprint verification

411 Fingerprint Claim Every node in the network signs itslist of neighbor nodesThe signedneighbor node list is used asa fingerprint of its current neighborhood community This iscalled fingerprint claimThe fingerprint claim is sent to all itsone-hopneighborhoods After receiving the fingerprint claimfrom a neighboring claim node the receiving node will makethe decision to become a witness node of the claim made ornot When the node decides to become a witness node itverifies the fingerprint claim

412 Fingerprint Verification The fingerprint verificationconsists of two steps namely the local verification processand the global verification process The local verificationprocess is performed when the receiving node accepts thefingerprint claim and decides to be the witness node In thelocal verification process the public key of the neighbor nodeis derived from the ID of the neighbor and the master publickey is verified The fingerprint claims are stored when thepublic key of the witnessed node and its fingerprint are thesame In the global verification phase the nodes exchange thewitness node when the relationship between the neighborsis built When the private key of the nodersquos ID shows twodifferent locations then they are detected as replica

SHD has low detection rate when the witness nodebecomes a replica Hence an attempt is made to enhancethe SHD method in terms of detection accuracy and controloverheads thereby the replicas are not selected as witnessnodes The enhancement of SHD using Clonal SelectionAlgorithmof Artificial Immune System is the proposedwork

The proposed method provides a solution to the problemrelated to the selection of witness node in replica detection

Simulate the mobile WSN for different parameters

Inject node replication attack

Collect the details of the later communication between nodes

Apply SHD method for the simulated mobile WSN

Apply Clonal Selection algorithm to select the witness node

Detect the replicated node in the mobileWSN

Figure 1 Flow diagram of the proposed methodology

The flow diagram of the proposed methodology is given inFigure 1

42 Clonal Selection Algorithm The Clonal Selection algo-rithm is one of the Artificial Immune System algorithmsThe main consideration to propose the Clonal Selectionalgorithm for immunity includes safeguarding the specificmemory set selection and reproducing the most stimulatedantibodies affinity maturation and reselection of the repro-duce

The Clonal Selection algorithm [19 20] is explained asfollows

When the antigens affect an animal the antibodies areproduced by B-lymphocytes Each cell produces an anti-body related to the specific variety of antigen The antigensstimulate the B-cells to divide and mature into plasma cells(terminal antibody secreting cells) As a result of the celldivision process clones are generated The lymphocytes inaddition to cell division also discriminate into long lived B-memory cells These memory cells are circulated throughoutthe body When the cells are stimulated by an antigenlarge lymphocytes are producedThe large lymphocyte whichgenerates high affinity antibodies is selected for particularantigen that triggers the primary response

43 Enhanced Single Hop Detection Using Clonal SelectionAlgorithm (CSSHD) The proposed enhanced SHD methodmakes use of the Clonal selection algorithm for the enhance-ment The enhanced SHD is similar to SHD except that theselection of witness nodes is done by the Clonal Selectionalgorithm


Randomly select an antigen An119895 and assign it to all antibodies in the AbDetermine the vector V119895 which contain affinity of antibodies119873Select the 119899 highest affinity antibodies from AbClone the selected antibodies sum119899119894=1 119904119899Produce the number of clones 119862119895 for each of the 119904119899 selected antibodiesAssign 119862119895 to an affinity maturation processProduce a population 119888119895

lowast

of matured clonesDetermine the affinity Vlowast119895 of the matured clones 119888119895

lowast

Among mature clonesElect witness node by choosing best highest affinity one (Ablowast119895 ) from the mature clones in the groupReplace the 119889 lowest affinity antibodies from Ab119903 with relation to Ag119895 by new individuals

Algorithm 1 Selection of witness node

The proposed CSSHD similar to SHD consists of finger-print claim and fingerprint verification phases In the finger-print claim phase the fingerprint of the nodersquos neighbors isexchanged between the one-hop neighborhoodsThemeetingtime 119872119905 of the two nodes ldquo119909rdquo and ldquo119910rdquo is computed [17] asfollows

119872119905 (119909 119910) = min 119905 10038161003816100381610038161003816119896119909 (119905) minus 119896119910 (119905)10038161003816100381610038161003816le 119877 (1)

where 119877 is transmission range of sensor nodes The mathe-matical formula of expectation119872119905 is computed as E119872119905 E119872119905is the time taken to deliver message from node 119909 to anothernode 119910

Upon receiving the fingerprint claim the receiving nodewill decide to become a witness node based on the Clonalselection algorithm

431 Selection of Witness Node Using Clonal Selection Algo-rithm The selection of witness node is based on the selectionof large lymphocytes in the Clonal Selection algorithm Thenode that has the maximum capability to forward data isselected as witness node The maximum capability of thewitness node is determined by its forwarding capability Theforwarding capability is determined by the trust value ofthe node The trust value [18] is calculated based on thedata packet forwarding ratio (DFR) and the control packetforwarding ratio (PFR) At time ldquo119905rdquo the trust value 119879(119905) ofthe node ldquo119910rdquo is measured by using node ldquo119909rdquo as follows

119879 (119905) = 1205751 lowast CPR119909119910 (119905) + 1205752 lowast PFR119909119910 (119905) (2)

where 1205751 and 1205752 are respective weights and the sum of theweights is equal to 1

When the trust value 119879(119905) is maximum for the node ldquo119910rdquothen the node ldquo119910rdquo is chosen as the witness node The replicacannot claim them as witness node because the trust value iscalculated by the node ldquo119909rdquo and not by the node ldquo119910rdquo

The proposed Clonal Selection for witnessed node isgiven in Algorithm 1

The enhanced SHD with Clonal Selection algorithm isspecified by allowing antigens once from An that performsAlgorithm 1 The 119899 highest affinity antibodies were arrangedin ascending order after selecting them from Ab so that

the number of clones for all these 119899 selected antibodies iscalculated [19] as

119873cl =119899

sum

119896=1

round(120573 sdot119873

119896) (3)

where119873cl is the sum of number of clones produced for eachof the antigens 120573 is a multiple constant 119873 is the sum ofnumber of antibodies and round(sdot) is operator used to roundits values Each term of this sum corresponds to the clone sizeof each selected antibody

When a node becomes a witness node it verifies thefingerprint claim After successful verification process fin-gerprint claims of the witnessed nodes are registered locallyusing Algorithm 2

Upon successful verification and registration locally thewitness nodes undergo global verification process

432 Global Fingerprint Verification In the global finger-print verification process when two nodes ldquo119909rdquo and ldquo119910rdquo meeteach other they communicate and exchange their witnessednode lists by piggybacking the Hello message This is thefirst beacon message exchanged at the time of establishmentof neighbor relationship Soon after the establishment ofneighbors these nodes exchange the fingerprint claims ofnodes with each other Further these nodes are verified forfeasible fingerprint claim conflict with received claims Thegroup meeting time 119866119905 [17] of the two node groups is definedas119866119905 (119866119860 119866119861)

= min 119905 10038161003816100381610038161003816119896119909 (119905) minus 119896119910 (119905)10038161003816100381610038161003816le 119877 119909 isin 119866119860 119910 isin 119866119861

(4)

where 119866119860 and 119866119861 are the two groups of nodes The mathe-matical formula of expectation 119866119905 is calculated as E119866119905

In a fingerprint claim conflict process if two fingerprintclaims have same ID but private key claims have differentneighborhood communities then the replicas are detectedusing Algorithm 3

The flow diagram of the proposed methodology is givenin Figure 2

The pseudocode of the proposed algorithm is given inAlgorithm 4


If (local verification)

If (check (public key of neighbor node fingerprint claim of neighbor node) = 1)Set signature = false

Elseclaimed neighborsfl extract neighbor list from fingerprint claim of a neighbor nodeIf (Node id not in claimed neighbors)

Set signature = falseElse

Set signature = true

Algorithm 2 Local verification

If (globalverification)

Replica list = 0For each ID in the list of witnessed nodes of this node 1198711cap list of witnessed nodes of the meeting node 1198712

Local claim = fetch local claim of id from local stored claimsCompare claim = fetch claim of id from meeting nodeIf (neighbor list of local claim = neighbor list in compare claim)

Set replica = trueAdd id to replica list

Algorithm 3 Global verification

The proposed CSSHD method helps in selecting theappropriate witness node Hence the detection accuracy canbe improved by detecting the replicas with minimum controloverheads The above section clearly explained the proposedmethod The next section explains the experimental setupand results

5 Experimental Setup and Results

During experimentation the characteristics of each nodein the network and its performance are analyzed using theproposed CSSHD method The proposed methodology istested using NS-2 simulator which is common and wellknown network simulator tool The version of NS-2 is ns-234 This tool is mainly used in the simulation area ofMANET wireless sensor network VANET and so forthFigure 3 shows the simulation methodology

The simulation parameters are initialized while develop-ing this concept and are shown in Table 2 These parametersare used for the construction of the network

The simulation time varies from 500 seconds to 1000 sec-onds During the simulation time the statistics are collectedThe statistics includes data packets received control packets

generated sent packets sumof all packets delay total numberof received packets total number of replica nodes correctlyfound total bytes received per second and total number ofkilobytes Using the above statistics the followingmetrics aredefined

(i) Packet delivery ratio(ii) Control overhead(iii) Average delay(iv) Message drop(v) Throughput(vi) Detection ratio(vii) False alarm rate

The performance of the proposed method is evaluated interms of the above parameters

Packet Delivery Ratio (PDR) PDR is defined as percentage ofpackets successfully received at the destinations and the totalnumber of packets sent by the sources defined as follows

PDR =Received PacketsSent Packets

lowast 100 (5)


Each packet Ptransfer

For each node Select randomly and

assign all antibodies in Abselect N highest affinity

Produce clones for selected antibodies

Assign affinity for clones and produce population for

clones

Determine affinity for matured clones

Witness node

local stored fingerprint claims

Local verification

Global verification

Verify the public key and the fingerprint claim of the

neighbor node

Not equal to 1

Set signature as 0

Extract the neighbor listfrom the fingerprint claim

of the neighbour node

Check whether the node id is present in the claimed

neighbors

Present to 1

Present to 1

Set signature as false

Set signature as true

Fetch the local claim ID and compare it with the

meeting node

Check whether the neighbor node is in the

neighbour list

Set replica as true and add ID

to the replica list

Set node as a benign node

End

End

End

Set flag as 0

Yes

YesYes

Yes

Yes

Yes

No

No

End

No

No

No

No

120573 middotN

k)Ncl =

n

sumk=1

round(

Gj

D =L

sumi=1

120575 where 120575120575

Determine vector j and

L1 cap L2

Add node to witness list L1 and claim to

=1 if

0 otherwise

Abi ne Agi

Ab = Abr cup Abm

Figure 2 Flow diagram of the proposed CSSHD method

Overhead It is defined as the percentage of total numbers ofcontrol packets generated to the total number of data packetsreceived during the simulation time given as follows

Overhead =Control Packet generatedData Packets Received

(6)

Average DelayThe average delay is computed by sumof everydata packet delay to the total number of received packets asdefined below in (7) The parameter is measured only whenthe data transmission has been successful

Average Delay =Sum of All Packets Delay

Total Number of Received Packets (7)


Set 1198751 119875119899 are neighbours of 119902For each node 119875119894

Send sign and add to neighbour list 1198751 119875119899Beacon = 119902id neighbour listClaim = beacon sign(beacon)Choose an antigen An119895 (An119895 isin 119860119899)

Add An119895 to Ab = Ab119903 cup Ab119898 where (119903 + 119898 = 119873)

Vector 119881119895 affinity and119873 antibodies in AbChoose119873 highest affinity antibodies from AbAb=Ab119894119873 where Ab

119894

119873 subset of Ag119895Chosen119873 antibodies will be clones119873119888 = sum

119873

119894=1 round(120573 sdot 119873) where119873119888 is the number of clones119873 is the number of antibodies119862 = 1198621 119862119895 where 119862 represents set of clones and 1 le 119895 le 119899

Assign clones 119862119895 to affinity maturation process119862119895 = 119862

lowast119895 of matured clones

Calculate 119881lowast119895 of the matured clone 119862lowast119895For 119895 = 1 to 119899

Select within node 119862lowast119895 = highest affinity (ABlowast119895 )NextIf (node is witness node)

Set flag = 1119882 = 119862

lowast119895 + 119882 where119882 is the witness node list

Add claim of 119862lowast119895 to local stored fingerprint claimElse

Set flag = 0End ifIf (local verification)

Neighbour claim = fingerprint of 1198751 119875119899Neighbour PK = neighbour public key from (119872PK (neighbour Id))Check public key of neighbour nodeIf (neighbour claim = true)

Set signature = 0Else

Signature = 1End ifClaimed neighbours = extract neighbourlist from neighbour claim listIf (id = claimed neighbour)


Set signature = 1End if

End ifIf (global verification)

119861119902

= 0

1198711 = 1198821 119882119899

1198712 = 1198721 119872119899

For each ID in 1198711 cap 1198712

Claim = ID of location from local stored claimsCompare = ID from 1198721 119872119899

If (neighbour list (Claim) = neighbourlist (compare))Add ID to 119861(119902)

End ifNext

End ifNext

Algorithm 4 Proposed enhanced Single Hop Detection using Clonal Selection Algorithm (CSSHD)


for different parameters

Inject replicated nodes


Apply SHD and enhanced SHD using Clonal Selection algorithms (CSSHD)

Evaluate the performance using the collected details

Tabulate and represent graphically the results and compare the performance

Parameters such as(i) Channel type

(ii) Radio propagation model(iii) Network interface type(iv) MAC type(v) Traffic

(vi) Mobility(vii) Routing protocol

(viii) Simulation time(ix) Number of nodes and mobile

nodes(x) Link layer and interface

queue type(xi) Antenna model

(xii) Maximum packet

Simulate the mobile WSN of 1000 lowast 1000

Figure 3 Simulation methodology

Table 2 Simulation parameters

Parameters ValuesChannel type WirelessChannelradio-Propagationmodel TwoRayGroundnetwork interface type WirelessPhyMAC type 802 11interface queue type DropTailPriQueuetime of simulation end 200 slink layer type LLantenna model OmniAntenna

max packet 50 (Minimum 512 bytes Maximum10000 bytes)

number of mobile nodes 50

Simulation time 200 s (Minimum 200 s Maximum10000 s)

routing protocol AODV119883 dimension of topography 1000119884 dimension of topography 1000Traffic tclfilescbrMobility tclfilesspeed5number of replica node 5 (Minimum 5 Maximum 25)Truelink 1

Message Drop Message drop is defined as rate of numberof message received in a packet at the destination by total

number of message sent from source It is represented bypercentage ()

Message Drop

=Number of message received in a packet

Total Number of Message Sentlowast 100

(8)

Throughput Throughput is defined as total file size transmit-ted in a given range It is represented by kbps

Throughput = File sizeTransmission range

(9)

where

Transmission time = File SizeBandwidth

(10)

Detection Ratio Detection ratio is defined as percentage ofreplica node correctly found by total number of replica nodeIt is represented by percentage ()

Detection Rate

=Number of Replica Node correctly found

Total Number of Replica node

lowast 100

(11)

False Alarm Rate False alarm rate is defined as number ofreplica nodes correctly found by total number of replica nodeIt is represented by percentage ()


Table 3 Performance of proposed method and the existing method based on the number of nodes

Number ofnodes

Control overhead (kbps) Message drops () Packet delivery ratio () Average delay (s) lowast 10minus3 Throughput (kbps)SHD CSSHD SHD CSSHD SHD CSSHD SHD CSSHD SHD CSSHD

50 47500 25000 90000 57500 920000 945000 37500 17500 130000 165000100 95000 85000 135000 110000 862500 877500 95000 75000 177500 195000150 145000 125000 195000 165000 802500 830000 195000 137500 245000 270000200 175000 142500 237500 215000 762500 782500 250000 185000 270000 330000Average 115625 94375 164400 136900 836900 858800 144375 103750 205625 240000

Table 4 Performance of proposed and the existing based on the number of replicas

Number ofreplica nodes

Detection ratio () False alarm rate () lowast 10minus3

SHD CSSHD SHD CSSHD5 965000 980000 35000 3000010 947500 961500 70000 4250015 915000 935000 85000 5150020 885000 917500 112500 8150025 865000 900000 137500 95000Average 915500 938800 88000 60100

False Alarm Rate =Total Number of Replica Node minusNumber of replica node correctly found

Total Number of Replica Nodelowast 100 (12)

Figure 4 shows the graph for packet delivery ratio inwhich the proposed method has higher packet delivery ratiocompared to the existing methodThe packet delivery ratio isrepresented by percentage ()

Figure 5 shows the graph for control overhead where theproposed method has lower routing overhead compared tothe existing method

Figure 6 shows the graph for average packet delaywherein the proposedmethod has lowpacket delay comparedto the existing method The average delay is represented bymilliseconds (s)

Figure 7 shows the graph for detection ratio where theproposed method has higher detection rate compared to theexisting method It is represented by percentage ()

Figure 8 shows the graph for false detection ratio inwhich the proposed method has low false detection ratiocompared to the existing method It is represented by per-centage ()

Figure 9 shows the graph for message drops whereinthe proposed method has lower message drop rate than theexisting method It is represented by percentage ()

Figure 10 shows the graph for throughput where theproposed method has higher throughput when compared tothe existing method It is represented by kbps

The overall comparison results of proposed work areshown in Tables 3 and 4 In both Tables 3 and 4 proposedmethod is specified as CSSHD in short

From Table 3 it is observed that the proposed methodshows a better result when compared to the existing methodin terms of control overhead message drops PDR averagedelay and throughput

The performance of the proposedmethodwith that of theexisting method by varying the number of replica nodes isgiven in Table 4

From Table 4 it is observed that the detection ratio of theproposed method is improved whereas the false alarm ratehas reduced to a greater extent when compared to the existingmethod

The overall comparison results of proposed work areshown in Table 5

Table 5 clearly shows the percentage of improvementachieved for various performance metrics of the proposedmethod The proposed work improves its performance inall the metrics particularly the detection ratio is improvedmuch better than the existing method

6 Conclusion

In mobile WSN node replication attack is a crucial one Thevarious replica detectionmethods are information exchangedbased detection node meeting based detection and themobility based detection Out of these three replica detectionmethods the proposed work concentrates on the mobil-ity assisted based detection method The proposed work


Table 5 Comparison result

Metrics Existing SHD method (kbps) Proposed CSSHD method (kbps) Improvement ()Packet delivery ratio 920000 950000 3Control overhead 175000 150000 25Average delay (ms) 250000 190000 6Message drops () 240000 210000 3Throughput (kps) 270000 330000 6Detection ratio () 960000 980000 20False detection ratio () 135000 95000 40

Pack

et d

eliv

ery

ratio

()

50 100 150 200Number of nodes

SHDCSSHD

0102030405060708090

100

Figure 4 Graph for packet delivery ratio


SHDCSSHD

Con

trol o

verh

ead

(kpb

s)

020000400006000080000

100000120000140000160000180000200000

Figure 5 Graph for control overheads

enhances the SHD method using Clonal Selection algorithmof AIS to improve the detection ratio by selecting thebest witness node The proposed CSSHD method is usedin a fully distributed environment where communicationoccurs among single hop neighbors highly strong againstnode collusion and efficient in protecting against multiple


SHDCSSHD

0

50000

100000

150000

200000

250000

300000

Aver

age d

elay

(s)lowast

10minus3

Figure 6 Graph for average delay

replica nodes The experiment is conducted using the ns-2 simulator The proposed method has high throughputless overhead and low false alarm rate The results of theproposed approach are compared with existing methodwhich shows that the average delay control overhead andmessage drops are minimized with higher packet deliveryratio value and higher detection ratio This proves that the


SHDCSSHD

5 10 15 20 25D

etec

tion

ratio

()

Number of replica nodes

80828486889092949698

100

Figure 7 Graph for detection ratio

SHDCSSHD

5 10 15 20 25Number of replica nodes

False

alar

m ra

te (

)lowast10

minus3

0

2

4

6

8

10

12

14

16

Figure 8 Graph for false alarm rate

SHDCSSHD

50 100 150 200

Mes

sage

dro

ps (

)

Number of nodes

0

5

10

15

20

25

Figure 9 Graph for message drops

proposedmethod is efficient towards detecting clones that arenot resilient against collusive replicas with minimum controloverheads

SHDCSSHD

50 100 150 200

Thro

ughp

ut (k

bps)

Number of nodes

0

50000

100000

150000

200000

250000

300000

350000

Figure 10 Graph for throughput

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

References

[1] C P Mayer ldquoSecurity and privacy challenges in the internet ofthingsrdquo Electronic Communications of the EASST vol 17 pp 1ndash12 2009

[2] B Parno A Perrig and V D Gligor ldquoDistributed detection ofnode replication attacks in sensor networksrdquo in Proceedings ofthe IEEE Symposium on Security and Privacy pp 49ndash63 IEEEMay 2005

[3] Y Zeng J Cao S Zhang S Guo and L Xie ldquoRandom-walk based approach to detect clone attacks in wireless sensornetworksrdquo IEEE Journal on Selected Areas in Communicationsvol 28 no 5 pp 677ndash691 2010

[4] B Zhu S Setia S Jajodia S Roy and LWang ldquoLocalized mul-ticast efficient and distributed replica detection in large-scalesensor networksrdquo IEEE Transactions on Mobile Computing vol9 no 7 pp 913ndash926 2010


[5] J-W Ho D Liu M Wright and S K Das ldquoDistributed detec-tion of replica node attacks with group deployment knowledgein wireless sensor networksrdquo Ad Hoc Networks vol 7 no 8 pp1476ndash1488 2009

[6] C-M Yu Y-T Tsou C-S Lu and S-Y Kuo ldquoLocalizedalgorithms for detection of node replication attacks in mobilesensor networksrdquo IEEE Transactions on Information Forensicsand Security vol 8 no 5 pp 754ndash768 2013

[7] J-W Ho M K Wright and S K Das ldquoFast detection of replicanode attacks in mobile sensor networks using sequential analy-sisrdquo in Proceedings of the 28th Conference on Computer Commu-nications (INFOCOM rsquo09) pp 1773ndash1781 Rio de Janeiro BrazilApril 2009

[8] M Conti R Di Pietro L V Mancini and A Mei ldquoEmergentproperties detection of the node-capture attack inmobile wire-less sensor networksrdquo in Proceedings of the 1st ACM Conferenceon Wireless Network Security (WiSec rsquo08) pp 214ndash219 April2008

[9] B Zhu V G K Addada S Setia S Jajodia and S RoyldquoEfficient distributed detection of node replication attacks insensor networksrdquo in Proceedings of the 23rd Annual ComputerSecurity Applications Conference (ACSAC rsquo07) pp 257ndash266IEEE Miami Beach Fla USA December 2007

[10] M Conti R Dipietro and A Spognardi ldquoClone wars dis-tributed detection of clone attacks in mobile WSNsrdquo Journal ofComputer and System Sciences vol 80 no 3 pp 654ndash669 2014

[11] J-W Ho M Wright and S K Das ldquoFast detection of replicanode attacks in mobile sensor networks using sequentialanalysisrdquo in Proceedings of the 28th Conference on ComputerCommunications (IEEE INFOCOM rsquo09) pp 1773ndash1781 IEEERio de Janeiro Brazil April 2009

[12] C-M Yu C-S Lu and S-Y Kuo ldquoMobile sensor networkresilient against node replication attacksrdquo in Proceedings of the5th Annual IEEE Communications Society Conference on SensorMesh and Ad Hoc Communications and Networks (SECON rsquo08)pp 597ndash599 IEEE San Francisco Calif USA June 2008

[13] W Z Khan M Y Aalsalem M N B M Saad and YXiang ldquoDetection and mitigation of node replication attacksin wireless sensor networks a surveyrdquo International Journal ofDistributed Sensor Networks vol 2013 Article ID 149023 22pages 2013

[14] Y Lou Y Zhang and S Liu ldquoSingle hop detection of nodeclone attacks inmobilewireless sensor networksrdquo inProceedingsof the International Workshop on Information and ElectronicsEngineering (IWIEE rsquo12) pp 2798ndash2803 Harbin China March2012

[15] T Spyropoulos K Psounis and C S Raghavendra ldquoEffi-cient routing in intermittently connected mobile networks thesingle-copy caserdquo IEEEACM Transactions on Networking vol16 no 1 pp 63ndash76 2008

[16] T Spyropoulos K Psounis and C S Raghavendra ldquoPerfor-mance analysis of mobility-assisted routingrdquo in Proceedingsof the 7th ACM International Symposium on Mobile Ad hocNetworking and Computing (MobiHoc rsquo06) pp 49ndash60 FlorenceItaly 2006

[17] T Spyropoulos A Jindal and K Psounis ldquoAn analyticalstudy of fundamental mobility properties for encounter-basedprotocolsrdquo International Journal of Autonomous and AdaptiveCommunications Systems vol 1 no 1 pp 4ndash40 2008

[18] T Camp J Boleng and V Davies ldquoA survey of mobility modelsfor ad hoc network researchrdquo Wireless Communications andMobile Computing vol 2 no 5 pp 483ndash502 2002

[19] LNDeCastro andF J VonZuben ldquoLearning andoptimizationusing the clonal selection principlerdquo IEEE Transactions onEvolutionary Computation vol 6 no 3 pp 239ndash251 2002

[20] E Ulker and S Ulker ldquoComparison study for clonal selectionalgorithm and genetic algorithmrdquo International Journal ofComputer Science amp Information Technology vol 4 no 4 pp107ndash118 2012

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Active and Passive Electronic Components

Control Scienceand Engineering

Journal of



RotatingMachinery


Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design



Shock and Vibration


Civil EngineeringAdvances in

Acoustics and VibrationAdvances in



Electrical and Computer Engineering

Journal of

Advances inOptoElectronics


Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of


Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014


Chemical EngineeringInternational Journal of Antennas and

Propagation




Navigation and Observation



DistributedSensor Networks



The detection of replica node in the Mobile WirelessSensor Networks is a crucial task So far only few detectionschemes have been proposed [6ndash9] Since the adversarydistributes replica nodes everywhere in the network themobility-assisted detection scheme is required to detect thereplicas in the network

In the earlier works mobility assisted technique SingleHop Detection (SHD) was proposed In SHD each nodebroadcasts its location claim to its single hop neighborsand selects the witness node The selected witness nodedetects replicas by performing the verification process As aresult it reduces communication overhead However whenthe replicas collude with each other they select replica as awitness node Hence the detection accuracy is low

Themain objective of the proposedwork is to improve thedetection accuracy by selecting the appropriate witness nodewith reduced overheads during the detection of replica nodesin themobile wireless sensor networks Tomeet the objectiveSHD is enhanced using the Artificial Immune System

Artificial Immune System (AIS) is a branch of ArtificialIntelligence based on the principles of Human Immune Sys-tem It provides various solutions to the real world problemsdue to its characteristic features The characteristic featuresinclude learning ability to the new conditions adaptabilityand distributed nature to the diverse environment limitedresources survivability even in the harsh environmentsThe enhancement of SHD with AIS improves the detectionaccuracy

The contribution of the paper includes the enhance-ment of the SHD method by applying the Clonal Selectionalgorithm for selecting the witness nodes that is anothercontribution Due to this the detection ratio is increasedby selecting the appropriate witness nodes and therebythe replica node detection process incurs minimum controloverheads

This paper is organized as follows In Section 2 variousrelated works towards this technique implemented by differ-ent authors are discussed In Section 3 the system model isdiscussed In Section 4 proposed CSSHD method is brieflyexplained and in Section 5 the performance of the proposedtechnique is compared with the existing technique based oncertain performance metrics and finally Section 6 concludesthe paper

2 Previous Works

Identifying the replicas in mobile wireless sensor networksis an enforcing task under hostile environments The exist-ing node replication attack detection methods are eithercentralized or distributed [6ndash12] The centralized detectionmethod has distinct points where it can fail and also can becaptured by the adversary As a result distributed detectionmethods are proposed in the literature The distributedreplica detection methods are of different types [13] namely

(i) information exchange based method(ii) node meeting based method(iii) mobility assisted based detection method

Of the detection methods only the distributed detectionmethods are discussed below

Yu et al [6 12] in 2013 present the eXtremely EfficientDetection (XED) method This is a distributed detectionalgorithm for mobile networks in which the detection isbased on the information exchanged between the nodes inthe network It detects the replica based on the randomnumber exchanged between each other of the two nodesThedetection capability is degraded when the replicas exchangethe exact random value

Ho et al [7] in 2009 presents the Sequential ProbabilityRatio Test (SPRT) at each periodic time the sensor nodetravels to the new position it signs the claim and sends thisclaim to its neighbors and to the base stationThe base stationcalculates the speed depending on the probability ratio testand matches it with the observed speed If the observedspeed is higher than the computedmaximum speed then thedetected node is a replica node

Conti et al [8] in 2008 introduced a clone detectionmechanism called Simple and Co-operative DistributedDetection (SDD amp CDD) To duplicate a node it must beisolated from the network and then its information must beextractedThis information earns some periodic time and theSimple Distributed Detection [8] uses that time period fordetecting the replica The detection capability of Simple Dis-tributedDetection (SDD) does not providemuch accuracy todetect the replica nodes and thus improved to develop a Co-operative Distributed Detection (CDD) procedureThe CDD[8] use the nodes cooperation to improve the replica nodedetection rate The nodes interchange the information whennodes are in the same communication radius This methodincreases the detection probability

Zhu et al [9] in 2007 proposed an Efficient and Dis-tributed Detection (EDD) a node meeting based distributeddetection method The EDD method computes the numberof encountering times between nodes in a given time periodldquo119879rdquo with higher probability If a node with a higher thresholdvalue is encountered then it is considered as a replicanode

Lou et al [14] in 2012 proposed Single Hop Detection(SHD) method It is a mobility assisted based distributeddetection method In SHD method when a node appears atdifferent neighborhood community replica is detected Thismethod improves the communication overhead

All the above methods discussed are compared andTable 1 presents the comparison

The parameters used by various detection methods forevaluation are the number of claims overheads detectionrate and false detection rates From the literature it isobserved that SHD method is efficient when compared toother detection methods in terms of control overheads butthe detection accuracy is low when the replica is selected as awitness node Hence an attempt is made to enhance the SHDmethod usingArtificial Immune System in terms of detectionaccuracy with minimum number of control overheads Thesystem model of the proposed work is explained in the nextsection



















3 System Model



































lowast


lowast




119872119905 (119909 119910) = min 119905 10038161003816100381610038161003816119896119909 (119905) minus 119896119910 (119905)10038161003816100381610038161003816le 119877 (1)










119873cl =119899

sum

119896=1

round(120573 sdot119873

119896) (3)






(4)



























lowast 100 (5)







clones


Witness node


Local verification

Global verification


neighbor node

Not equal to 1

Set signature as 0




neighbors

Present to 1

Present to 1




meeting node


neighbour list


to the replica list


End

End

End

Set flag as 0

Yes

YesYes

Yes

Yes

Yes

No

No

End

No

No

No

No

120573 middotN

k)Ncl =

n

sumk=1

round(

Gj

D =L

sumi=1

120575 where 120575120575


L1 cap L2


=1 if

0 otherwise

Abi ne Agi

Ab = Abr cup Abm




(6)









119894


119873






Set flag = 1119882 = 119862










119861119902

= 0

1198711 = 1198821 119882119899

1198712 = 1198721 119872119899




End ifNext

End ifNext


























Message Drop



(8)



(9)

where


(10)


Detection Rate



lowast 100

(11)




Number ofnodes


50 47500 25000 90000 57500 920000 945000 37500 17500 130000 165000100 95000 85000 135000 110000 862500 877500 95000 75000 177500 195000150 145000 125000 195000 165000 802500 830000 195000 137500 245000 270000200 175000 142500 237500 215000 762500 782500 250000 185000 270000 330000Average 115625 94375 164400 136900 836900 858800 144375 103750 205625 240000




















6 Conclusion





Pack

et d

eliv

ery

ratio

()


SHDCSSHD

0102030405060708090

100



SHDCSSHD

Con

trol o

verh

ead

(kpb

s)

020000400006000080000

100000120000140000160000180000200000




SHDCSSHD

0

50000

100000

150000

200000

250000

300000

Aver

age d

elay

(s)lowast

10minus3




SHDCSSHD

5 10 15 20 25D

etec

tion

ratio

()


80828486889092949698

100


SHDCSSHD


False

alar

m ra

te (

)lowast10

minus3

0

2

4

6

8

10

12

14

16


SHDCSSHD

50 100 150 200

Mes

sage

dro

ps (

)

Number of nodes

0

5

10

15

20

25



SHDCSSHD

50 100 150 200

Thro

ughp

ut (k

bps)

Number of nodes

0

50000

100000

150000

200000

250000

300000

350000




References
























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation



























3 System Model



































lowast


lowast




119872119905 (119909 119910) = min 119905 10038161003816100381610038161003816119896119909 (119905) minus 119896119910 (119905)10038161003816100381610038161003816le 119877 (1)










119873cl =119899

sum

119896=1

round(120573 sdot119873

119896) (3)






(4)



























lowast 100 (5)







clones


Witness node


Local verification

Global verification


neighbor node

Not equal to 1

Set signature as 0




neighbors

Present to 1

Present to 1




meeting node


neighbour list


to the replica list


End

End

End

Set flag as 0

Yes

YesYes

Yes

Yes

Yes

No

No

End

No

No

No

No

120573 middotN

k)Ncl =

n

sumk=1

round(

Gj

D =L

sumi=1

120575 where 120575120575


L1 cap L2


=1 if

0 otherwise

Abi ne Agi

Ab = Abr cup Abm




(6)









119894


119873






Set flag = 1119882 = 119862










119861119902

= 0

1198711 = 1198821 119882119899

1198712 = 1198721 119872119899




End ifNext

End ifNext


























Message Drop



(8)



(9)

where


(10)


Detection Rate



lowast 100

(11)




Number ofnodes


50 47500 25000 90000 57500 920000 945000 37500 17500 130000 165000100 95000 85000 135000 110000 862500 877500 95000 75000 177500 195000150 145000 125000 195000 165000 802500 830000 195000 137500 245000 270000200 175000 142500 237500 215000 762500 782500 250000 185000 270000 330000Average 115625 94375 164400 136900 836900 858800 144375 103750 205625 240000




















6 Conclusion





Pack

et d

eliv

ery

ratio

()


SHDCSSHD

0102030405060708090

100



SHDCSSHD

Con

trol o

verh

ead

(kpb

s)

020000400006000080000

100000120000140000160000180000200000




SHDCSSHD

0

50000

100000

150000

200000

250000

300000

Aver

age d

elay

(s)lowast

10minus3




SHDCSSHD

5 10 15 20 25D

etec

tion

ratio

()


80828486889092949698

100


SHDCSSHD


False

alar

m ra

te (

)lowast10

minus3

0

2

4

6

8

10

12

14

16


SHDCSSHD

50 100 150 200

Mes

sage

dro

ps (

)

Number of nodes

0

5

10

15

20

25



SHDCSSHD

50 100 150 200

Thro

ughp

ut (k

bps)

Number of nodes

0

50000

100000

150000

200000

250000

300000

350000




References
























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation



































lowast


lowast




119872119905 (119909 119910) = min 119905 10038161003816100381610038161003816119896119909 (119905) minus 119896119910 (119905)10038161003816100381610038161003816le 119877 (1)










119873cl =119899

sum

119896=1

round(120573 sdot119873

119896) (3)






(4)



























lowast 100 (5)







clones


Witness node


Local verification

Global verification


neighbor node

Not equal to 1

Set signature as 0




neighbors

Present to 1

Present to 1




meeting node


neighbour list


to the replica list


End

End

End

Set flag as 0

Yes

YesYes

Yes

Yes

Yes

No

No

End

No

No

No

No

120573 middotN

k)Ncl =

n

sumk=1

round(

Gj

D =L

sumi=1

120575 where 120575120575


L1 cap L2


=1 if

0 otherwise

Abi ne Agi

Ab = Abr cup Abm




(6)









119894


119873






Set flag = 1119882 = 119862










119861119902

= 0

1198711 = 1198821 119882119899

1198712 = 1198721 119872119899




End ifNext

End ifNext


























Message Drop



(8)



(9)

where


(10)


Detection Rate



lowast 100

(11)




Number ofnodes


50 47500 25000 90000 57500 920000 945000 37500 17500 130000 165000100 95000 85000 135000 110000 862500 877500 95000 75000 177500 195000150 145000 125000 195000 165000 802500 830000 195000 137500 245000 270000200 175000 142500 237500 215000 762500 782500 250000 185000 270000 330000Average 115625 94375 164400 136900 836900 858800 144375 103750 205625 240000




















6 Conclusion





Pack

et d

eliv

ery

ratio

()


SHDCSSHD

0102030405060708090

100



SHDCSSHD

Con

trol o

verh

ead

(kpb

s)

020000400006000080000

100000120000140000160000180000200000




SHDCSSHD

0

50000

100000

150000

200000

250000

300000

Aver

age d

elay

(s)lowast

10minus3




SHDCSSHD

5 10 15 20 25D

etec

tion

ratio

()


80828486889092949698

100


SHDCSSHD


False

alar

m ra

te (

)lowast10

minus3

0

2

4

6

8

10

12

14

16


SHDCSSHD

50 100 150 200

Mes

sage

dro

ps (

)

Number of nodes

0

5

10

15

20

25



SHDCSSHD

50 100 150 200

Thro

ughp

ut (k

bps)

Number of nodes

0

50000

100000

150000

200000

250000

300000

350000




References
























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation











lowast


lowast




119872119905 (119909 119910) = min 119905 10038161003816100381610038161003816119896119909 (119905) minus 119896119910 (119905)10038161003816100381610038161003816le 119877 (1)










119873cl =119899

sum

119896=1

round(120573 sdot119873

119896) (3)






(4)



























lowast 100 (5)







clones


Witness node


Local verification

Global verification


neighbor node

Not equal to 1

Set signature as 0




neighbors

Present to 1

Present to 1




meeting node


neighbour list


to the replica list


End

End

End

Set flag as 0

Yes

YesYes

Yes

Yes

Yes

No

No

End

No

No

No

No

120573 middotN

k)Ncl =

n

sumk=1

round(

Gj

D =L

sumi=1

120575 where 120575120575


L1 cap L2


=1 if

0 otherwise

Abi ne Agi

Ab = Abr cup Abm




(6)









119894


119873






Set flag = 1119882 = 119862










119861119902

= 0

1198711 = 1198821 119882119899

1198712 = 1198721 119872119899




End ifNext

End ifNext


























Message Drop



(8)



(9)

where


(10)


Detection Rate



lowast 100

(11)




Number ofnodes


50 47500 25000 90000 57500 920000 945000 37500 17500 130000 165000100 95000 85000 135000 110000 862500 877500 95000 75000 177500 195000150 145000 125000 195000 165000 802500 830000 195000 137500 245000 270000200 175000 142500 237500 215000 762500 782500 250000 185000 270000 330000Average 115625 94375 164400 136900 836900 858800 144375 103750 205625 240000




















6 Conclusion





Pack

et d

eliv

ery

ratio

()


SHDCSSHD

0102030405060708090

100



SHDCSSHD

Con

trol o

verh

ead

(kpb

s)

020000400006000080000

100000120000140000160000180000200000




SHDCSSHD

0

50000

100000

150000

200000

250000

300000

Aver

age d

elay

(s)lowast

10minus3




SHDCSSHD

5 10 15 20 25D

etec

tion

ratio

()


80828486889092949698

100


SHDCSSHD


False

alar

m ra

te (

)lowast10

minus3

0

2

4

6

8

10

12

14

16


SHDCSSHD

50 100 150 200

Mes

sage

dro

ps (

)

Number of nodes

0

5

10

15

20

25



SHDCSSHD

50 100 150 200

Thro

ughp

ut (k

bps)

Number of nodes

0

50000

100000

150000

200000

250000

300000

350000




References
























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation































lowast 100 (5)







clones


Witness node


Local verification

Global verification


neighbor node

Not equal to 1

Set signature as 0




neighbors

Present to 1

Present to 1




meeting node


neighbour list


to the replica list


End

End

End

Set flag as 0

Yes

YesYes

Yes

Yes

Yes

No

No

End

No

No

No

No

120573 middotN

k)Ncl =

n

sumk=1

round(

Gj

D =L

sumi=1

120575 where 120575120575


L1 cap L2


=1 if

0 otherwise

Abi ne Agi

Ab = Abr cup Abm




(6)









119894


119873






Set flag = 1119882 = 119862










119861119902

= 0

1198711 = 1198821 119882119899

1198712 = 1198721 119872119899




End ifNext

End ifNext


























Message Drop



(8)



(9)

where


(10)


Detection Rate



lowast 100

(11)




Number ofnodes


50 47500 25000 90000 57500 920000 945000 37500 17500 130000 165000100 95000 85000 135000 110000 862500 877500 95000 75000 177500 195000150 145000 125000 195000 165000 802500 830000 195000 137500 245000 270000200 175000 142500 237500 215000 762500 782500 250000 185000 270000 330000Average 115625 94375 164400 136900 836900 858800 144375 103750 205625 240000




















6 Conclusion





Pack

et d

eliv

ery

ratio

()


SHDCSSHD

0102030405060708090

100



SHDCSSHD

Con

trol o

verh

ead

(kpb

s)

020000400006000080000

100000120000140000160000180000200000




SHDCSSHD

0

50000

100000

150000

200000

250000

300000

Aver

age d

elay

(s)lowast

10minus3




SHDCSSHD

5 10 15 20 25D

etec

tion

ratio

()


80828486889092949698

100


SHDCSSHD


False

alar

m ra

te (

)lowast10

minus3

0

2

4

6

8

10

12

14

16


SHDCSSHD

50 100 150 200

Mes

sage

dro

ps (

)

Number of nodes

0

5

10

15

20

25



SHDCSSHD

50 100 150 200

Thro

ughp

ut (k

bps)

Number of nodes

0

50000

100000

150000

200000

250000

300000

350000




References
























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation















clones


Witness node


Local verification

Global verification


neighbor node

Not equal to 1

Set signature as 0




neighbors

Present to 1

Present to 1




meeting node


neighbour list


to the replica list


End

End

End

Set flag as 0

Yes

YesYes

Yes

Yes

Yes

No

No

End

No

No

No

No

120573 middotN

k)Ncl =

n

sumk=1

round(

Gj

D =L

sumi=1

120575 where 120575120575


L1 cap L2


=1 if

0 otherwise

Abi ne Agi

Ab = Abr cup Abm




(6)









119894


119873






Set flag = 1119882 = 119862










119861119902

= 0

1198711 = 1198821 119882119899

1198712 = 1198721 119872119899




End ifNext

End ifNext


























Message Drop



(8)



(9)

where


(10)


Detection Rate



lowast 100

(11)




Number ofnodes


50 47500 25000 90000 57500 920000 945000 37500 17500 130000 165000100 95000 85000 135000 110000 862500 877500 95000 75000 177500 195000150 145000 125000 195000 165000 802500 830000 195000 137500 245000 270000200 175000 142500 237500 215000 762500 782500 250000 185000 270000 330000Average 115625 94375 164400 136900 836900 858800 144375 103750 205625 240000




















6 Conclusion





Pack

et d

eliv

ery

ratio

()


SHDCSSHD

0102030405060708090

100



SHDCSSHD

Con

trol o

verh

ead

(kpb

s)

020000400006000080000

100000120000140000160000180000200000




SHDCSSHD

0

50000

100000

150000

200000

250000

300000

Aver

age d

elay

(s)lowast

10minus3




SHDCSSHD

5 10 15 20 25D

etec

tion

ratio

()


80828486889092949698

100


SHDCSSHD


False

alar

m ra

te (

)lowast10

minus3

0

2

4

6

8

10

12

14

16


SHDCSSHD

50 100 150 200

Mes

sage

dro

ps (

)

Number of nodes

0

5

10

15

20

25



SHDCSSHD

50 100 150 200

Thro

ughp

ut (k

bps)

Number of nodes

0

50000

100000

150000

200000

250000

300000

350000




References
























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation














119894


119873






Set flag = 1119882 = 119862










119861119902

= 0

1198711 = 1198821 119882119899

1198712 = 1198721 119872119899




End ifNext

End ifNext


























Message Drop



(8)



(9)

where


(10)


Detection Rate



lowast 100

(11)




Number ofnodes


50 47500 25000 90000 57500 920000 945000 37500 17500 130000 165000100 95000 85000 135000 110000 862500 877500 95000 75000 177500 195000150 145000 125000 195000 165000 802500 830000 195000 137500 245000 270000200 175000 142500 237500 215000 762500 782500 250000 185000 270000 330000Average 115625 94375 164400 136900 836900 858800 144375 103750 205625 240000




















6 Conclusion





Pack

et d

eliv

ery

ratio

()


SHDCSSHD

0102030405060708090

100



SHDCSSHD

Con

trol o

verh

ead

(kpb

s)

020000400006000080000

100000120000140000160000180000200000




SHDCSSHD

0

50000

100000

150000

200000

250000

300000

Aver

age d

elay

(s)lowast

10minus3




SHDCSSHD

5 10 15 20 25D

etec

tion

ratio

()


80828486889092949698

100


SHDCSSHD


False

alar

m ra

te (

)lowast10

minus3

0

2

4

6

8

10

12

14

16


SHDCSSHD

50 100 150 200

Mes

sage

dro

ps (

)

Number of nodes

0

5

10

15

20

25



SHDCSSHD

50 100 150 200

Thro

ughp

ut (k

bps)

Number of nodes

0

50000

100000

150000

200000

250000

300000

350000




References
























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation

































Message Drop



(8)



(9)

where


(10)


Detection Rate



lowast 100

(11)




Number ofnodes


50 47500 25000 90000 57500 920000 945000 37500 17500 130000 165000100 95000 85000 135000 110000 862500 877500 95000 75000 177500 195000150 145000 125000 195000 165000 802500 830000 195000 137500 245000 270000200 175000 142500 237500 215000 762500 782500 250000 185000 270000 330000Average 115625 94375 164400 136900 836900 858800 144375 103750 205625 240000




















6 Conclusion





Pack

et d

eliv

ery

ratio

()


SHDCSSHD

0102030405060708090

100



SHDCSSHD

Con

trol o

verh

ead

(kpb

s)

020000400006000080000

100000120000140000160000180000200000




SHDCSSHD

0

50000

100000

150000

200000

250000

300000

Aver

age d

elay

(s)lowast

10minus3




SHDCSSHD

5 10 15 20 25D

etec

tion

ratio

()


80828486889092949698

100


SHDCSSHD


False

alar

m ra

te (

)lowast10

minus3

0

2

4

6

8

10

12

14

16


SHDCSSHD

50 100 150 200

Mes

sage

dro

ps (

)

Number of nodes

0

5

10

15

20

25



SHDCSSHD

50 100 150 200

Thro

ughp

ut (k

bps)

Number of nodes

0

50000

100000

150000

200000

250000

300000

350000




References
























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation











Number ofnodes


50 47500 25000 90000 57500 920000 945000 37500 17500 130000 165000100 95000 85000 135000 110000 862500 877500 95000 75000 177500 195000150 145000 125000 195000 165000 802500 830000 195000 137500 245000 270000200 175000 142500 237500 215000 762500 782500 250000 185000 270000 330000Average 115625 94375 164400 136900 836900 858800 144375 103750 205625 240000




















6 Conclusion





Pack

et d

eliv

ery

ratio

()


SHDCSSHD

0102030405060708090

100



SHDCSSHD

Con

trol o

verh

ead

(kpb

s)

020000400006000080000

100000120000140000160000180000200000




SHDCSSHD

0

50000

100000

150000

200000

250000

300000

Aver

age d

elay

(s)lowast

10minus3




SHDCSSHD

5 10 15 20 25D

etec

tion

ratio

()


80828486889092949698

100


SHDCSSHD


False

alar

m ra

te (

)lowast10

minus3

0

2

4

6

8

10

12

14

16


SHDCSSHD

50 100 150 200

Mes

sage

dro

ps (

)

Number of nodes

0

5

10

15

20

25



SHDCSSHD

50 100 150 200

Thro

ughp

ut (k

bps)

Number of nodes

0

50000

100000

150000

200000

250000

300000

350000




References
























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation












Pack

et d

eliv

ery

ratio

()


SHDCSSHD

0102030405060708090

100



SHDCSSHD

Con

trol o

verh

ead

(kpb

s)

020000400006000080000

100000120000140000160000180000200000




SHDCSSHD

0

50000

100000

150000

200000

250000

300000

Aver

age d

elay

(s)lowast

10minus3




SHDCSSHD

5 10 15 20 25D

etec

tion

ratio

()


80828486889092949698

100


SHDCSSHD


False

alar

m ra

te (

)lowast10

minus3

0

2

4

6

8

10

12

14

16


SHDCSSHD

50 100 150 200

Mes

sage

dro

ps (

)

Number of nodes

0

5

10

15

20

25



SHDCSSHD

50 100 150 200

Thro

ughp

ut (k

bps)

Number of nodes

0

50000

100000

150000

200000

250000

300000

350000




References
























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










SHDCSSHD

5 10 15 20 25D

etec

tion

ratio

()


80828486889092949698

100


SHDCSSHD


False

alar

m ra

te (

)lowast10

minus3

0

2

4

6

8

10

12

14

16


SHDCSSHD

50 100 150 200

Mes

sage

dro

ps (

)

Number of nodes

0

5

10

15

20

25



SHDCSSHD

50 100 150 200

Thro

ughp

ut (k

bps)

Number of nodes

0

50000

100000

150000

200000

250000

300000

350000




References
























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation




























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation











RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation









Documents

Research Article Replica Node Detection Using Enhanced ...downloads.hindawi.com/journals/jcnc/2016/1620343.pdfby enhancing the Single Hop Detection (SHD) method using the Clonal Selection