Quality of Service for Wireless Sensor Network · S. Vivek Saravanan Final year B.E., Department of EEE SRM Easwasri Engineering College, Chennai, Tamilnadu, India Email: [email protected]

Copyright © 2013 IJECCE, All right reserved1620

International Journal of Electronics Communication and Computer EngineeringVolume 4, Issue 6, ISSN (Online): 2249–071X, ISSN (Print): 2278–4209

Quality of Service for Wireless Sensor NetworkS. Vivek Saravanan

Final year B.E., Department of EEESRM Easwasri Engineering College, Chennai, Tamilnadu, India

Email: [email protected]

A. SolairajuAssociate Professor of Mathematics,

Jamal Mohamed College, Tiruchirappalli, Tamilnadu, IndiaEmail: [email protected]

Abstract – Wireless Sensor Networks (WSNs) are graduallyadopted in the industrial world due to their advantages overwired networks. In addition to saving cabling costs, WSNswiden the realm of environments feasible for monitoring.They thus add sensing and acting capabilities to objects inthe physical world and allow for communication among theseobjects or with services in the future Internet. However, theacceptance of WSNs by the industrial automation communityis impeded by open issues, such as security guarantees andprovision of Quality of Service (QoS). To examine both ofthese perspectives, we select and survey relevant WSNtechnologies dedicated to industrial automation. Wedetermine QoS requirements and carry out a threat analysis,which act as basis of our evaluate ion of the cur-rent state-of-the-art. According to the results of this evaluation, weidentify and discuss open research issues.

Keywords – Wireless Sensor Networks, IndustrialAutomation, Quality of Service, State-of-the-art,WirelessHART, ISA100.11a-2009.

I. INTRODUCTION

Industrial automation has been successfully introducedin a countless amount of industries ranging from food toenergy industries. Even if the products differ from oneindustry to another, the automated processes can beclassified according to three main layers as proposed by[1]: the plant-floor automation layer, the manufacturingexecution system layer and the enterprise resourceplanning layer. Internet technology can be considered asthe link that interconnects all these layers and allows forinformation exchange. For example, it serves as abackbone to interconnect different production locationswithin one enterprise, to transfer production control data innear real-time to the headquarters or to integrate suppliersinto a production workflow.

Fig.1. From the sensor to the customers

Within the scope of this survey, we focus on the plant-floor automation layer including sensors, switches,programmable controllers and motor starters (Fig. 1) thatensure the correct operation of ma-chines and execution ofprocesses, while the remaining layers are dedicated to theoptimization of the production by managing resourceallocation and operation scheduling for example. Inaddition to productivity gain and precision improvement,the automation of processes at the plant-floor automationlayer allows the replacement of workers in harsh andhazardous environments or assigned to tedious tasks [2].The WSNs are part of this layer and can be used formultiple purposes, such as monitoring synchronous orasynchronous events that require periodic data collectionor detecting exceptional events, respectively [3]. Forexample, vibration, heat or thermal sensors can bedeployed in proximity of machines to monitor their health.The analysis of the measured parameters can allow thedetection of abnormal operating conditions and aidstherefore in preventing potential machine failure. Inaddition to machine monitoring, WSNs can be deployed tomeasure basic physical quantities such as pressure,temperature, flow or more complex events such as processquality or automotive performance in industrialenvironments [4].

Although wired sensor networks can also be deployedfor such monitoring scenarios, WSNs present additionaladvantages. In fact, their wireless capability allowsdeployments in hostile environments, where vibrations ormoving parts may prevent the use of cables that would bedamaged or even broken. In addition to reduce cablingcosts, the WSNs provide network flexibility, as the sensornodes may be relocated quickly without necessitatingtime-consuming cable installation and maintenance.However, the nature of the wireless medium opens upsecurity and QoS issues. For example, potential attackersmay easily eavesdrop or manipulate wirelesscommunication in absence of security mechanisms and thewireless channel has to be efficiently allocated betweenthe different devices to provide the required QoS.1.1 Our contributions are as follows:

1. We first select relevant WSN technologies dedicatedto industrial automation and we provide an exhaustivesurvey of their characteristics.

2. We then analyze industrial automation applications todetermine QoS requirements and evaluate the selectedstandards according to each identified QoS requirement.Open issues are discussed based on the results of thisevaluation.

The paper is structured as follows. It provides a detailedoverview of the following WSN technologies: WirelessInterface for Sensor and Actuators (WISA),WirelessHART, ISA100.11a, ZigBee, ZigBee PRO, and

mailto:[email protected]

mailto:[email protected]



802.15.4e Factory Automation MAC Layer. Furtheritfocuses on QoS and security respectively. Both sectionsare composed of an analysis of the respective requirementsand an evaluation of the selected specifications. Openissues are listed and discussed at the end of both sections.

II: SELECTED WIRELESS SENSOR NETWORKS

STANDARDS: STATE-OF-THE-ART

2.1 Wireless communication in industrial automationIt is mostly based on standardized technologies, such as

the IEEE 802.11 [5] and IEEE 802.15 standard families[6], also designated as Wireless Local Area Networks(WLAN) and Wireless Personal Area Networks (WPAN).Both of these standard families were conceived forapplication purposes different than industrial automation.In fact, the IEEE 802.11-based standards offer high datarates in the order of tens of Mbit/s and ranges up totens/hundreds of meters, while the IEEE 802.15-basedstandards only supports data rates of hundreds of kbit/s toseveral Mbits/s with ranges from a few meters up tohundreds of meters. However, to provide greater data rateand range, IEEE 802.11 technology consumes a greaterenergy budget that can limit the benefits obtained bywireless communications.

Indeed, the sensor nodes are either powered by cables orbatteries. In the former case, the advantages provided bywireless communication are partially negated, whereas inthe latter case, the scarce energy resource has to beparsimoniously consumed in order to avoid frequenthuman interventions to recharge the batteries. Energy isthus a major concern in both previous cases and wetherefore focus on the IEEE 802.15-based standards, andparticularly on the IEEE 802.15.1 and IEEE 802.15.4standard, within the scope of this work.2.2 IEEE 802.15.1-based Standards:

The IEEE 802.15.1 standard, also known as Bluetooth®,can be classified to fall between the IEEE 802.11 andIEEE 802.15.4 standards in terms of energy consumptionand data rates. With medium data rates and lower energyconsumption than the IEEE 802.11 standard, IEEE802.15.1 offers an interesting compromise between energyconsumption and data rate, and is therefore particularlysuited for high-end applications requiring high data ratesas well as applications with strong real-time requirementssuch as factory automation. The Wireless Interface forSensor and Actuators (WISA) has been selected as arepresentative 802.15.1-based specification for furtherdiscussion.2.3 Wireless Interface for Sensor and Actuators(WISA):

Released by ABB and presented in [7], the proprietaryWireless Interface for Sensors and Actuators (WISA)specification is based on the IEEE 802.15.1 physical layerand targets factory automation WSNs with packet errorrate less than 1−9 and cycle time of 2ms.2.4 Network Elements and Architecture

WISA networks [8] can be deployed in cellular topologywith up to three cells (Fig. 2).

Fig.2. WISA network elements

Each celluses a different transmission frequency, and iscomposed of a base station and up to 120 end devicesincluding sensors and/or actuators organized in a startopology. The WISA architecture is limited to thephysical and MAC layers, as sensors and actuatorscommunicate exclusively with a central base station in astar topology within each cell. The four uplink channelsare divided into superframes of 2048 µs, which arecomposed of 30 timeslots able to support packets up to 64bit length (Fig. 3).

Fig.3. WISA superframe structure

2.5 IEEE 802.15.4 based StandardsThe IEEE 802.15.4 standard presents lower data rates,

but requires also lower energy budgets. According to [6,10], the standard is appropriate for infrequent exchangesof small packets, when power consumption is an importantissue. The standard is therefore suited for processautomation applications, where continuous productionstreams are monitored.



The IEEE 802.15.4 physical layer is common to all thefollowing standards and operates in the 2.4 GHz frequencyband as well as the 868 MHz and 915 MHz bands inEurope and North America, respectively. The 2.4 GHzfrequency band is divided into 16 channels with a maximaldata rate of 250 Kbits/s per channel and separated by a 5MHz gap, while the 915 MHz band is divided into 10channels with a maximal data rate of 40 Kbit/s each. Thesingle channel in the 868 MHz frequency band presents adata rate of 20 Kbit/s.

Table 1: Scope of the selected technologies

2.6 WirelessHART:The HART Communication Foundation is an

independent and not-for-profit organization that ensuresthe development of the HART Protocol. As technologyowner and central authority, the foundation released theopen WirelessHARTTM standard in 2007, considered by[9] as the only released open wireless standard suitable forprocess measurement and control applications.WirelessHART networks are composed of differentdevices as illustrated in (Fig. 4), including field devices,gateways, network and security managers.

Fig.4. WirelessHART network elements

2.7 Network Elements and ArchitectureThe gateway is a bridge between the field device

network and the host application. The gateway isconfiguredby the network manager using HARTcommands and allows buffering large sensor data, eventnotifications, diagnostics, and command responses. Inaddition to the gateway configuration, the networkmanager configures the remaining devices and maintainsthe whole network.

At the data link layer, the WirelessHART standardcoordinates and manages each device’s transmission timeby using TDMA with timeslots of 10 ms. Each time slot

may be allocated to one source or may be shared betweenseveral sources using the Carrier Sense Multiple Accesswith Collision Avoidance (CSMA/CA) mechanism. Thetimeslot allocation is then communicated by the networkmanager in the superframe to each device. At least onesuperframe (Fig. 5) is continuously repeated at fixed rateand further superframes can be added to support additionaltraffic.

Fig.5. WirelessHART Superframe

At the transport layer, the WirelessHART standardsupports connection-oriented as well as connectionlesscommunication. The connection-oriented communicationsare set up for applications requiring a reliable transfer ofdata between the host application and the field device forexample. The connection set up starts by opening adedicated port on the targeted field device with a specificHART command and the transmission data rate is thennegotiated with the network manager, before the datatransmission between the both entities can begin.2.8 ISA100.11a: Network elements, and Architecture:

The ISA100.11a-2009 standard has been developed bythe ISA100 standards committee, part of the non-profitInternational Society of Automation (ISA) organization,and approved by the ISA Standards and Practices Board inSeptember 2009.

Fig.6. ISA100.11a network elements

In parallel to the main ISA100.11a working group,different working groups address complementary issuesincluding the compatibility of the ISA100.11a standardwith existing wired and wireless standards. ISA100.11a



WSNs can be organized according to different topologyschemes and are composed of field devices, gateway(s),and handheld device(s) as depicted in (Fig. 6).

The ISA100.11a data link layer is responsible for themanagement of the employed TDMA schemes byconfiguring the timeslot durations and managing thesuperframes. The configuration of timeslots can be doneaccording to two different patterns: slotted channelhopping and slow channel hopping. Each link is associatedto one or multiple timeslots of a superframe and its typecan be transmit and/or receive (Fig. 7). Information aboutthe neighbors, the channel offset from the superframehopping scheme as well as possible alternatives for thetransmission and reception are also included. TheISA100.11a data link layer supports the previouslydescribed graph routing as well as source routing.

Fig.7. ISA100.11a superframe

The network layer provides schemes for routing andQoS. It allows energy and bandwidth savings by mappingand translating the 128-bit addresses of the applicationendpoints to 16-bit short addresses used within subnetsand vice versa. These savings can be increased byadapting the packet formats in function of the desiredaddressing, routing or QoS.

Depending on the level of reliability required by theapplication, the ISA100.11a transport layer can supportend-to-end acknowledgements as well as unacknowledgedcommunication. Additionally, flow control, segmentationand reassembly as well as security are supported attransport layer, whereas the application layer ensuresstandard interoperability by using tunneling and nativeprotocols at the gateways. The former carry protocolsused in existing standards such as HART orFOUNDATION Fieldbus, while the latter provide efficientbandwidth utilization and therefore increase the batterylifetime.

III. QUALITY OF SERVICE

Quality of Service (QoS) refers to “the collective effectof service performance which determine the degree ofsatisfaction of a user of the service”. More technically,QoS can be defined as the “well-defined and controllablebehavior of a system with respect to quantitativeparameters”. Interesting parameters to monitor forcomputer networks can be delay, jitter, throughput,fairness, and packet losses for example. However, therequirements of industrial automation networks aredifferent from usual computer networks. The harsh

industrial operation environment may degrade the wirelesscommunication performance due to path loss andshadowing, multi-path propagation, and interference.Moreover, the traffic is mainly composed of short packetscontaining sensor measurements or actuator commandsthat need to be delivered timely, instead of largemultimedia streams or interactive traffic that prevails incomputer networks. We next identify the specific QoSrequirements and evaluate the aforementioned selectedspecifications with respect to their compliance to theserequirements. Open issues are identified and discussedbased on the results of this evaluation.3.1 Requirements

Industrial automation applications can be divided intotwo main categories of systems: close loop and open-loopsystems2 as defined. The applications based on eitherclose or open-loop systems have different QoSrequirements, as close-loop systems monitor discreteoperations, such as actuator control, and open-loopsystems monitor continuous processes, such as cookingraw material.

In close-loop applications, the sensors generate trafficthat needs to be transmitted timely, reliably and accuratelyto the control system, whereas in open-loop applications,real-time processing is typically not required, but theenergy consumption is a crucial point. An additionalclassification of the industrial applications refines thesecategories and is based on the degree of the required real-time guarantees. The classification is composed of fourclasses presenting increasing real-time requirements thatare summarized in Table 2. We next focus on the supportof real-time traffic and the reliability offered by theselected specifications.

Table 2: Industrial automation applications classifiedaccording to their real-time-requirements

3.2 Evaluation of the Selected SpecificationsTo evaluate the support of real-time transmission, the

medium access control mechanisms including prioritymanagement schemes of the selected specifications arecompared. Different diversity parameters includingfrequency and space diversity as well as protocol featuressuch as acknowledgements are additionally considered inorder to estimate the reliability provided by the differentspecifications.3.3 Real-time Support

The selected specifications are mainly based on theIEEE 802.15.4 data link layer, except for the WISAspecification relying on the IEEE 802.15.1 standard. As802.15.4e FA MAC Layer and WISA only address the



specifications of the physical and data link layers, thecomparisons of the selected technologies are mainly basedon the features of their two first layers. To complete thiscomparison, the 802.15.4 FA MAC needs to be evaluatedin the context of a whole system. However, this aspect isout-side the scope of this survey.3.4 Medium Access Control Mechanisms

Each standard uses TDMA as medium access controlmechanism (Table 3). Even if all specifications make useof TDMA access mechanisms, some differences betweenthem can be observed. First of all, the timeslot length canbe configured in the ISA100.11a, Without suchoptimization, time can be lost between two successiveslots if the slot is longer than the data to send. Configuringthe length of the timeslots is thus an important feature toallow for optimizing the real-time support. However, alltimeslots inside each superframe have the same length thatlimits a real adaptation to the requirements of individualapplications, as only the overall traffic mix can beoptimized.

Table 3: Features relevant for real-time operation

Most of the selected specifications offer both kinds oftimeslots. Only the WISA specification uses exclusivelydedicated timeslots. The choice between dedicated andshared timeslots is made difficult by the trade-off betweenreal-time support and optimized utilization of the medium.Indeed, the dedicated timeslots are assigned to oneparticular device only. If this device wants to sendcommands or measurements, the data are transmittedimmediately within the reserved timeslots. Otherwise thetimeslot is lost and other devices wanting to transmitinformation may be constrained to wait longer. In case ofshared slots, the devices have to complete to access themedium. The medium utilization is thus optimizedbecause slots can be utilized by the stations in need ofbandwidth, but the data transmission cannot always beimmediate due to random back-off mechanisms forexample.3.5 Superframe Management

Furthermore, the WirelessHART, ISA100.11a and theZigBee standards allows optimizing the transmission of

superframes to fit the real-time communicationconstraints. The ISA100.11a standard supportsinsertion,removal and activation of superframes during theoperating process, whereas additional super-frames can betransmitted in parallel to the mandatory superframe in theWirelessHART standard. In the without beacon mode ofthe ZigBee standard, no superframe is transmitted and thedevices do not have to wait for the next timeslot totransmit their data. However, this solution has alsodrawbacks. If one device has a great amount of data tosend, the channel would be occupied for a long period oftime and data requiring real-time transmission could notbe send during this period. Moreover, the time to getaccess to the channel may be longer than in case ofdedicated slots, if all the devices have data to be sent. Themanagement of dedicated and shared timeslots hastherefore to be tailored to application characteristics andthe traffic patterns.3.6 Priority Management

In addition to medium access control, flow control withassignment of priority to the packets can be introduced tosupport real-time at higher level. A priority flag indicatesto each device on the path if the arriving packet has to betransmitted without delay or can be buffered. TheWirelessHART standard defines four main priority levelsaccording to classes of data [11]. Control andconfiguration information as well as network diagnosticmessages have the highest priority, followed by a secondcategory composed of process data measurements andnetwork statistic messages. The lowest priority is assignedto the fourth category, which includes packets reportinginformation about events and alarms. Additional classes ofdata are classified into the third category. In comparison,the ISA100.11a standard allows to prioritize the QoScontracts established between the devices and the systemmanager in addition to message priorities. The devicewanting to transmit data first sends the required prioritiesin a contract proposal, as well as other additionalparameters including reliability, periodicity andnegotiability. The system manager replies to this request ina contract response and indicates the provided QoS for therequested communication. Depending on the trafficconditions, the system manager might not be able toprovide the requested QoS. If the device indicates that thecontract is negotiable, the system manager can proposeanother QoS level or postpone the contract to provide therequested QoS level.

The contract priority is set during the contractestablishment and concerns all messages exchangedduring the contract duration. Depending on thesepriorities, the system manager manages the routing and theload balancing to provide the guarantees defined in thecontract. Four levels of contract priority are available atnetwork layer: network control, real time buffer, real timesequential and best effort queued. The first priority levelcan be used by the system manager to communicatecritical information about the network management;whereas the second category is used for periodic dataexchanges with buffer overwrite op-erations in case offresher messages.



The third class, real time sequential, is appropriate forapplications requiring sequential and real-time datadelivery like video or voice-based applications. The lastlevel is adapted for client-server communications. Withinthe contract, message priority can also be assigned bysetting one bit to either low or high. However, the contractpriority takes precedence over the message priority. Eachstandard supports real-time communication in a differentway. While the WISA specification and the 802.15.4e FAMAC layer target applications with strong real-timerequirements, the ZigBeeand the WirelessHART standardare more adapted to applications with softer requirements.The recent ISA100.11a standard supports currently softreal-time requirements3.7: The evaluation of the selected technologies hasshown that: Dedicated timeslots are supported by all standards.Their utilization is a key feature to fulfil strong real-timerequirements and is particularly appropriate, if thecomplete set of the sensors have a continuous stream ofinformation send. To optimize the transmission, dedicatedas well as shared timeslots can be combined to supportfluctuations of the traffic load. Such hybrid modes can beenvisaged in most of the considered technologies exceptfor the WISA standard, which only support dedicatedtimeslots. Tuning the timeslot length is a second key feature thatallows to support application-specific traffic as well asoptimize the real-time support. Except for the WISA andthe WirelessHART standards, all standards offer thisoption. However, the timeslot length is common to alltimeslots within a superframe, which limits the adaptationto variations of traffic as well as real-time requirements.Although this property may be useful for applications withvarious length of data to send, it would introduceadditional overhead and complexity and may thereforeunnecessary. Superframe management is an additional means tomaintain real-time communication in case of very hightraffic load. However, the superframe optimizationproposed by WirelessHART and ISA100.11a remains aminor contribution to the real-time support in comparisonto the aforementioned dedicated timeslots and adaptabletimeslot lengths. Priority mechanisms are provided by theWirelessHART and the ISA100.11a standards. Bothstandards support message-based priority, while theISA100.11a standard offers contract-based priorityadditionally. Although the proposed contracts arepromising in terms of QoS guarantees, the overhead andthe complexity introduced by their management as well asthe additional resulting traffic may limit their practicalfeasibility. Even if message-based priorities may be lessefficient than contracts, their utilization allows to reach abalance between overhead and efficiency and provides acomplementary method to support real-time traffic.3.8 Reliability Support

Industrial WSNs are located in spaces, where equipmentmoves, conditions change and interference perturbs thecommunication. Mechanisms, such as space and frequency

diversity as well as acknowledgements (Table 4), are thusrequired to protect the wireless networks from thedisturbances.3.9 Space Diversity

Space diversity allows bypassing obstacles andinterference by modifying the routing within networksorganized in mesh topology. Such ability is howeverimpossible with star topologies, as each devicecommunicates exclusively and directly with a centralcoordinator. In case of obstacles that block the wirelesscommunication, no alternative path is available and thecommunication cannot be established. Within the selectedstandards, the WISA specification and the 802.15.4e FAMAC standard are foreseen to be deployed in startopology only. Their protection against wireless channelobstructions and fluctuations is therefore reduced incomparison with the other standards. However, themaximal distance between the central coordinator and thesensors is shorter than between devices deployed in meshtopology. The probability of an obstacle breaking thecommunication is therefore low, if careful networkplanning is performed.3.10 Frequency Diversity

In addition to space diversity, frequency diversityreduces the effects of the environment on the wirelesscommunication by limiting the interference. Twofrequency diversity schemes are possible: channelhopping and channel blacklisting. The channel hoppingallows avoiding interference by changing the transmissionfrequency, while the devices maintain lists of frequenciesto avoid due to their significant interference with thechannel blacklisting scheme. The WirelessHART and theISA100.11a standards use both of them and possesstherefore an efficient response against interference.3.11 Acknowledgement Management

Interferences or obstacles can also lead to packet loss.To ensure transmission reliability, acknowledgements(ACK) at Data Link Layer (DLL) are supported by all theselected standards, as well as end-to-endacknowledgements at Transport Layer (TL) for theWirelessHART and ISA100.11a standards. Each standardhowever manages its ACK mechanisms according to itstransmission scheme. For example, the WISAspecification transmits each ACK in the downlink channel,whereas each ACK is transmitted during the same timeslotas the received data in the WirelessHART standard. Oncean ACK is missing, all standards7 support the automaticretransmission of the data.3.12 Open Issues:

The previous evaluation has highlighted that the selectedspecifications differ with respect tothe supported QoSrequirements for industrial applications, particularlyconcerning real-time support and reliability.3.13 Open Standard for Factory Automation

The first conclusion that can be drawn is that nostandard provides currently an open solution to factoryautomation with strict real-time requirements. The WISAspecification is dedicated to this kind of deployments, butit is based on the IEEE 802.15.1 standard that consumesmore energy than the IEEE 802.15.4-based standards.



Moreover, the WISA specification is proprietary, thuslocking the user into a single vendor as well as the designand maintenance of a proprietary set of interfaces. Even ifWISA successfully fulfills the requirements of factoryautomation, it does not support openness andinteroperability.

The 802.15.4 Factory Automation MAC provides apromising perspective, but is still under development.Furthermore, the recently released ISA100.11a standard isexpected to be enhanced by future addenda to make it fitto applications with stronger real-time requirements. As aconsequence, additional research and development isnecessary in order to obtain open and IEEE 802.15.4-based standards, which suit the needs of factoryautomation.3.14 QoS Support in Heterogeneous Networks:

In the process automation domain, the standardsWirelessHART and ISA100.11a occupy the first positionsand a potential convergence is analyzed by the ISA100.12Working Group. WirelessHARTdevices are foreseen to bedeployed within ISA100.11a-based WSNs, as illustrated in(Fig. 8).

Fig.8. Interoperability of WirelessHART and ISA100.11.anetworks

In parallel to the ISA100.12 Working Group, theISA100 Wireless Backhaul Backbone Network WorkingGroup addresses potential interoperability between theISA100.11a.. Additionally, the ISA100.11a standard iscompatible with existing wired standards includingHART, FOUNDATION Fieldbus, Modbus, and Profibus.

The ISA100.11a standard offers therefore a wide panelof compatible standards, which allows manufacturers toreuse existing devices. However, questions regarding theprovided QoS capabilities among heterogeneous WSNshave to be raised. Indeed, such compatibility requirestranslation between the standards at the gateways, whichmay increase the end-to-end transmission delay of endsystems belonging to different kinds of networks. Thus,the efforts to support the QoS requirements would bewasted. Therefore, the support of interoperability betweenthe investigated standards remains an open issue.Although addressing the heterogeneity has a number ofinteresting research challenges, its practical feasibility may

be limited by the technical complexity and financial costsincurred to adaptdifferent technologies.3.15 QoS Support in Multi-hop Networks:

The surveyed standards dedicated to process automationcan also be deployed in mesh topology, thus raising thequestion of QoS support over multi-hop routes. TheWirelessHART and ISA100.11a standards proposesolutions based on priority mechanisms, which arecentrally managed by the network manager. Furtherresearch is therefore mandatory to provide an exhaustivedescription of the required mechanismsto support QoS inmulti-hop networks.

REFERENCES

[1] Geng,H. Manufacturing Engineering Handbook, McGraw-HillProfessional: New York, USA, 2004.

[2] Shell, R.L.; Hall, E.L. Handbook of Industrial Automation;Marcel Dekker: New York, USA, 2000.

[3] Low, K.S.; Win, W.N.N.; Er, M.J. Wireless Sensor Networksfor Industrial Environments,In Proceedings of the InternationalConference on Computational Intelligence for Modelling,Control andAutomation and International Conference onIntelligent Agents, Web Technologies and Internet Commerce(CIMCA-IAWTIC), Vienna, Austria, November 2005.

[4] Mukhopadhyay, S.C.; Huang, Y.M. Sensors: Advancements inModeling, Design Issues, Fabrication and Practical Applications;Springer-Verlag: eidelberg, Germany, 2008.

[5] IEEE Computer Society. IEEE Standard for InformationTechnology, Telecommunications and Information Exchangebetween Systems, Local and Metropolitan Area Networks,Specific Requirements, Part 11: Wireless LAN Medium AccessControl (MAC) and Physical Layer (PHY) Specifications, 2007.

[6] Willig, A.; Matheus, K.; Wolisz, A. Wireless Technology inIndustrial Networks. Proceedings of the IEEE 2005, 93, 1130-1151.

[7] Scheible, G.; Dzung, D.; Endresen, J.; Frey, J.E. Unplugged butConnected - Design and Imple mentation of a Truly WirelessReal-time Sensor/Actuator Interface. IEEE Industrial ElectronicsMagazine 2007, 1, 25-34.

[8] Steigmann, R.; Endresen, J. Introduction to WISA andWPS.http://www.abb.com (accessed on Nov. 2009).

[9] De Biasi, M.; Snickars, C.; Landerns, K.; Isaksson, A.J.Simulation of Process Control with WirelessHART NetworksSubject to Packet Losses. In Proceedings of the Conference onAutomation Science and Engineering (CASE), Washington, DC,USA, August 2008.

[10] Baronti, P.; Pillai, P.; Chook, V.W.; Chessa, S.; Gotta, A.; Hu,Y.F. Wireless Sensor Networks: A Survey on the State of theArt and the 802.15.4 and ZigBee Standards. ComputerCommunications 2007, 30, 1655-1695.

[11] Kim, A.N.; Hekland, F.; Petersen, S.; Doyle, P. When HARTgoes Wireless: Understanding and Implementing theWirelessHART Standard. In Proceedings of the IEEEInternational Conference on Emerging Technologies andFactory Automation (ETFA), Hamburg, Germany, September2008.

AUTHOR’S PROFILE

S.Vivek Saravananis doing Final B.E. (Electrical and Electronics Emgineering), SRMEawasri Engineering College, Ramapuram, Chennai, Tamilnadu, India.

A. Solairajuis working as an “Associate Professor of Mathematics”, Jamal MohamedCollege, Tiruchirappalli-620020, Tamilnadu, India. He wrote five bookson Engineering Mathematics. He has published 150 research papers inNational and International Journals. He is an editor in some National andInternational Journals.

http://www.abb.com

Documents

Quality of Service for Wireless Sensor Network · S. Vivek Saravanan Final year B.E., Department of EEE SRM Easwasri Engineering College, Chennai, Tamilnadu, India Email: [email protected]