Dharam Vir, Dr. S.K. Agarwal, Dr. S.A.Imam / International Journal of Engineering Researchand Applications (IJERA) ISSN: 2248-9622 www.ijera.comVol. 3, Issue 3, May-Jun 2013, pp.335-340335 | P a g eTraffic based energy consumption analysis and improve thelifetime and performance of MAC protocols in ad hoc wirelesssensor networksDharam Vir*, Dr. S.K. Agarwal**, Dr. S.A.Imam****(Department of Electronics Engineering, YMCA University of Science & Technology, Faridabad, India -121006)** (Department of Electronics Engineering, YMCA University of Science & Technology, Faridabad, India -121006)*** (Department of Electronics & Comm. Engineering, Jamia Millia Islamia, New Delhi, India - 110019)ABSTRACTWireless microsensor networks lendthemselves to trade-offs in energy and quality.The service lifetime of such sensor nodes dependson the power supply and energy consumption ofnodes, which is usually dominated by thecommunication subsystem. One of the keychallenges in unlocking the potential of such datagathering sensor networks is conserving energy soas to maximize their post deployment activelifetime. This paper described the researchcarried on the continual development of the novelenergy efficient analysis of random placed nodesalgorithm that increases the WSNs lifetime andimproves on the QoS parameters yielding higherthroughput, average end to end delay jitter fornext generation of WSNs. Another key advantageof the novel analysis of random placed nodesalgorithm is that it can be implemented withexisting energy saving protocols like LEACH,SMAC and TMAC to further enhance thenetwork lifetimes and improve on QoSparameters. The main aim of this paper is toimprove energy in nodes and to analyze the mostenergy efficient MAC protocol in order to classifythem and compare their performances. We areimplemented some of WSN MAC protocol underQualNet 5.0 with the purpose to evaluate theirperformances.Keywords - MAC protocol, Energy consumption,Delivery ratio, Throughput, DelayI. INTRODUCTIONAd hoc wireless sensor networks (WSNs)are formed from self-organising configurations ofdistributed, energy-constrained, autonomous sensornodes. The nodes are microelectronics devices isequipped with heavily integrated sensing,processing, and wireless communication capabilitiesand are equipped with an independent power source,such as a small battery . When these nodes arenetworked together in an ad-hoc fashion, they forma sensor network. The nodes gather data via theirsensors, process it locally or coordinate amongstneighbours, and forward the information to the useror, in general, a data sink. Another cause for energyconsumption in the MAC layer is idle channelsensing. If the nodes receive scheduling packets orwant to transmit a message they need to sense thechannel very often and wait until the channel issensed idle, and that consumes energy. In a sharedmedium, the data transmitted by one node isreceived by all nodes within that transmission range;hence a node may waste energy in receiving packetsthat are not destined for it. Nearly all the MACprotocols operate by sending control packets ofdifferent types e.g. synchronization, scheduling,RTS, CTS, ACK etc, this results in more energyconsumption for resource limited wireless sensornodes. From above it can be concluded that not onlyis the energy consumption and network lifetime animportant issue, but network throughput can also benot ignored. However a balance needs to be madebetween network lifetime and throughput based onrequirements of the application. The service life ofsuch nodes depends on the power supply and theenergy consumption, which is typically dominatedby the communication subsystem. One of the keychallenges in unlocking the potential of such datagathering sensor networks is conserving energy soas to maximize their post-deployment active lifetime. Research has been carried out in many differentareas of WSNs from enhancing medium accesscontrol (MAC) protocols to improving topologymanagement schemes where energy can be reduced.Many of the efficient MAC routing protocols try toincrease the network lifetime by transitioning theidle nodes to sleep state. This not only increases theoverhead cost of introducing synchronisationpackets, but also has detrimental effects on thenetwork QoS parameters that introduce end to enddelay, average jitter and results in decreasedthroughput  .The purpose of putting nodes to sleep ismore useful for nodes that are furthest away fromthe base station. In the case of the nodes nearest tothe base station, having much higher traffic toforward at peak rates these nodes cannot actuallyshare the same sleep-awake schedule that is plannedfor the rest of the WSN. In energy efficient routing
Dharam Vir, Dr. S.K. Agarwal, Dr. S.A.Imam / International Journal of Engineering Researchand Applications (IJERA) ISSN: 2248-9622 www.ijera.comVol. 3, Issue 3, May-Jun 2013, pp.335-340336 | P a g eprotocols, lot of emphasis is given on minimisingthe transmission energy consumption, by eitherintroducing multi-hop in the network, or finding theminimum energy route, where the message will besent via a calculated multi-hop route. Theseprotocols fail to scale the size of the networks. Themajority of these protocols which is finds theminimum path and then always chooses that pathregardless of the amount of traffic. Hence nodesrequired to forward messages in that minimumenergy path list will always die first. Theseprotocols fail to address the issue of energyconsumption based on traffic. In many wirelesssensor networks, the nodes nearest the base stationwill always be the busiest as traffic will beapproaching these nodes from all directions of thesensor fields. These nodes provide the last hop for asuccessful transmission. One of the significantdifferences between the new power analysis randomalgorithm introduced in this paper and the existingenergy saving protocols is that the new poweranalysis of random algorithm for balances theenergy on the network traffic and also introducesideal mode that further extends network lifetime bysaving idle energy consumption. This balances theenergy consumption of sensor nodes at all stages ofthe network. At a point furthest away from the basestation, the transmission distance is longer as thetraffic is much lower and the nodes can afford totransmit at a longer distance. For nodes nearest tothe base station, where the traffic is higher, thetransmission range is intelligently calculated to beshorter, so more data can be forwarded to the basestation without consuming all the cluster headenergy  .To evaluate the QoS parameters in terms ofthroughput, average end to end delay and averagejitter for analysis of randomly placed nodes in thenetwork.The rest of the paper is described in section2 MAC protocol leads to energy waste and lifetimeparameters. Section 3 describes the classification ofwireless MAC protocols. Section 4 Energy efficientrouting protocol analytical and simulated models for802.11e and 802.11. Section 5 QualNet simulationand results analysis. Section 6 conclusions generatedby the current work and discuss possible futurework.II. MAC PROTOCOL LEADS TO ENERGYWASTE AND WSN LIFE REDUCTIONThe MAC protocol leads to energy waste and WSNlife reduction, such as:1) Idle listening:A node doesn’t know when will bereceiving a frame so it must maintain permanentlyits radio in the ready to receive mode, as in the DCFmethod of the wireless networks protocol (IEEE802.11). This mode consumes a lot of energy, nearlyequal to the one consumed in receipt mode. Thisenergy is wasted if there isn’t any transmission onthe channel .2) Collisions:They concern the MAC contentionprotocols. A collision can occur when a nodereceives two signals or more simultaneously fromdifferent sources that transmit at the same time.When a collision occurs, the energy provided forframe transmission and reception is lost.  .3) Overhearing:It occurs when a node receives packets thatare not destined to him or redundant broadcast.4) Protocol Overhead:It can have several origins as the energieslost at the time of transmission and reception of thecontrol frames. For example, the RTS/CTS (Requestto send /Clear to send) used by some protocolstransport no information whereas their transmissionconsumes energy .5) Overmitting:It occurs when a sensor node sends data toa recipient who is not ready to receive them. Indeed,the sent messages are considered useless andconsume an additional energy .6) Packets size:The size of the messages has an effect onthe energy consumption of the emitting andreceiving nodes. In the other case, a hightransmission power is necessary for large sizepackets .7) Traffic fluctuation:The fluctuations of the traffic load can leadto the waste a node’s energy reserves. Therefore, theprotocol should be traffic adaptive. In, the authorsintroduce a comprehensive node energy model,which includes energy components for radioswitching, transmission, reception, listening, andsleeping .III. Classification of Wireless MAC Protocols:In a wireless sensor network sensors nodesare a low cost, resource constrained devices and areoften positioned randomly. In many applicationsthey are placed in inaccessible locations, makingbattery replacement unfeasible. As a consequence,power efficiency is an important requirement in amedium access control protocol for most wirelesssensor networks. The simplest way to conserveenergy by the MAC layer protocols is by turning offthe radio whenever the node is not transmittinghence leaving the node in sleep mode. Radio energyconsumption is a major component contributing tothe overall energy consumption at each node. Thewireless MAC protocols can be divided generally in
Dharam Vir, Dr. S.K. Agarwal, Dr. S.A.Imam / International Journal of Engineering Researchand Applications (IJERA) ISSN: 2248-9622 www.ijera.comVol. 3, Issue 3, May-Jun 2013, pp.335-340337 | P a g ethree main categories, Centralised, Distributed &Hybrid MAC Protocols as shown in Figure 1 .1) Centralised MAC Protocols:In centralised MAC protocols, it is theresponsibility of a single controller to allocate thechannel access for all the nodes in the sensor field toprovide a collision free environment. Channelmultiplexing techniques similar to Time DivisionMultiple Access (TDMA), Frequency DivisionMultiple Access (FDMA) and Code DivisionMultiple Access (CDMA) have been mentioned.Due to limited channel bandwidth and very largenumber of nodes, when implementing FDMA, itbecomes impractical for each node to have a uniqueoperating frequency  .Figure 1. Classification of Wireless MAC Protocols.2) Distributed MAC Protocols:These MAC protocols allow multiplechannel access to all the participating nodes basedon some rules. The prime example of this protocol isCSMA/CA, where all the participants regularlysense the medium to see if it is idle. If the channel isfound to be busy, the transmission is deferred untilthe channel becomes idle. The probability ofcollisions are avoided by introducing a time delayprocedure e.g. random back-off procedure asemployed in IEEE 802.11 distributed coordinationfunction (DCF). However the packet collisioncannot be completely avoided in distributed MACprotocols due to the “hidden” and “exposed” nodeproblem which has a very large impact on thenetwork QoS. To overcome this problem the DCFuses RTS and CTS control messages to reserve thetransmission time between two nodes  .3) Hybrid MAC Protocols:The existing centralised and distributedMAC protocols do not provide analyses of energysaving and network scalability for WSNs. An idealprotocol would exhibit the controllability ofcentralised protocols and the flexibility ofdistributed protocols. WSNs have uniquecharacteristics e.g. low energy reserves, compacthardware, small transmission ranges, event or taskdriven and have high redundancy  .IV. ENERGY EFFICIENT ROUTINGPROTOCOLS:This paper is primarily based on findingnew and effective routing algorithms to minimisethe energy consumption of wireless sensornetworks. Ad hoc wireless networks do not have anynetwork infrastructure. The wireless nodes in the adhoc network try to maintain connectivity viawireless communication. The power aware wirelessrouting protocol is two basic classifications;Activity based routing protocols andFigure 2. The classification of energy efficientwireless ad hoc routing.Connectivity based protocols  .Fig. 2 shows the classification of energy efficient adhoc routing protocols. The energy efficient routingprotocols can broadly be divided into fivecategories, active energy saving protocols,maximizing network lifetime protocols, passiveenergy saving protocols, topology control protocolsand broadcasting protocols.A. Maximizing Network Lifetime Protocols:The protocols associated with maximizingthe network lifetime and balance the energyconsumption of all the nodes in the network toovercome the problem of overuse of some nodes.One way to minimize energy consumption is to usethe Minimal Battery Cost Routing (MBCR) . Inthis routing protocol the total energy consumption ofdifferent routes is calculated adding the battery costfor each hop until the packets reaches destination.The aim is again to find the lowest energyconsumption path .1) Components of a Wireless Ad hoc Sensornetwork:In order to fully understand thecharacteristics and behavior of wireless sensornetworks, we need to define all the components thatcan be considered vital in the abstract simulationmodel. The following is the list of the importantcomponents that will constitute our simulationmodel.Power-aware wireless routing protocolsConnectivity BasedActivity BasedUnicasting Broadcasting TopologyControlPassiveEnergy savingActive EnergySavingMaximizingNetwork
Dharam Vir, Dr. S.K. Agarwal, Dr. S.A.Imam / International Journal of Engineering Researchand Applications (IJERA) ISSN: 2248-9622 www.ijera.comVol. 3, Issue 3, May-Jun 2013, pp.335-340338 | P a g e There will be fixed or immobile devices, callednodes that have data processing and datatransmission capabilities. These nodes will bepowered by on-board batteries that have limitedenergy. The nodes also need be equipped withgeographic localization devices . As nodes will have communication capabilities,e.g. either with the base station or with othernodes within their vicinity, they will closelyfollow the rules and algorithms defined by theOSI protocol stack. Hence all the nodes willhave their communications regulated by thecharacteristics associated to the OSI protocollayers: application, presentation, session,transport, network, data link, (MAC), andphysical layer . Node energy power relies on the use of onboard batteries. Every action involving the useof the radio has energy cost. As the nodes willbe communicating with the base station or withtheir neighbours, they will be consumingenergy in the transmission or reception ofpackets. When device is switched on or off,consumes energy, hence affecting the availableenergy and node life. The transmission model also need to bedynamic as the transmission range will need tobe varied depending on the amount of traffic atdifferent parts of the network. Radio propagation is affected by the adverseenvironmental conditions in which the networkis placed. The traffic load varies at different locations inthe network. Different traffic loads can becreated using Constant Bit Rate sources(CBRs). CBR sources needs to be incorporatedwith the abstract simulation model .A. Selection of Network simulators for WSNs:QualNet (Quality Networks) work fromScalable Network Technologies is a commercialproduct that is an improved modeling tool derivedfrom GloMoSim. It has a dedicated and fullyimplemented protocols and modules for both thewired and wireless scenarios including ad hoc,cellular and satellite models. The Simulator, thiscomponent is designed to include high fidelitymodels of networks of tens of thousands of nodeswith heavy traffic and high mobility.B. Analytical and Experimental Model for802.11e and 802.11 using QualNet 5.0 Simulator:The default values for all parameters in theconfiguration file of QualNet were used, i.e., thetransmission rate is set at 11 Mbps, and the powerconsumption is 900 mW for both receiving/idle andtransmitting states. The transmission range for eachnode is 100m (receiver threshold is -75dB). CBRtraffic is generated from node 1 to 30 times with 5second interval; the data size is 512 bytes andsimulation time is 300 seconds .Figure 3. Routing path of data (source to sink)running scenarios of 30 nodes for maximizing thelifetime of node in random algorithmV. RESULT ANALYSISFigure 4. Shows the outcome of average throughputfor randomly placed nodes.Figure 5. shows the outcome of average jitter inrandomly placed nodes.Figure 6. Shows the outcome of average end to enddelay (s) in randomly placed nodes.Figure 7. Shows the outcome of 802.11MACBroadcast Packets Sent to channel in randomlyplaced nodes.
Dharam Vir, Dr. S.K. Agarwal, Dr. S.A.Imam / International Journal of Engineering Researchand Applications (IJERA) ISSN: 2248-9622 www.ijera.comVol. 3, Issue 3, May-Jun 2013, pp.335-340339 | P a g eFigure 8. Shows the outcome of 802.11MACBroadcast Packets received clearly in randomlyplaced nodes.Figure 9. Shows the outcome of energy modeltraffic based energy consumption in transmit moderandomly placed nodes.Figure 10. Shows the outcome of energy modeltraffic based energy consumption in Ideal moderandomly placed nodes.Figure 11. Shows the outcome of energy modeltraffic based energy consumption in received moderandomly placed nodes.VI. CONCLUSIONAd hoc wireless sensor networks (WSNs)are formed from self-organising configurations ofdistributed, energy constrained, autonomous sensornodes. The service lifetime of wireless sensor nodesdepends on the energy consumption of thecommunication subsystem and channel capacitybeing used for data transmission. This improvesmethodology of the 802.11 MAC protocols led toincrease the lifetime of nodes and throughput ofnetworks. On the increase in node density, averageend to end delay decrease for all implementedschemes. This is due to the large no of nodes gettinginvolves in routing, which results in quicker datadelivery to the sink through these nodes.An increase in Transmission rate provides adecrease in Energy consumption for all schemes,since all the schemes use their power conservationmethods effectively in the transmitting the largenumber of data packets transmitted per unit of time.Moreover the startup energy overhead alsodecreases on an increase in the Transmission Rate.Hybrid showed the lowest Energy consumption anddistributed MAC protocol scheme highest energyconsumption. So increase in Transmission Rate canbe opted as a good method to decrease the energyconsumption in wireless sensor networks. On anincrease in Transmission rate, throughput increasefor all schemes because of large amount of datawhich is transmitted gets serviced by thecommunication system. Centralised MAC protocolscheme provides the very large throughputcompared to other two schemes. Thus increase intransmission rate can be effective method toincrease the throughput in a wireless sensornetworks. An increase in transmission rate initiallyprovide a slight increase in average end to end delayfor hybrid and distributed MAC protocol schemes.The distributed MAC protocol provides theminimum delay. Therefore the increase inTransmission rate has very little effects on theaverage end to end delay encountered by the datapackets in a wireless sensor networks. Thebandwidth decreases in a wireless sensor networkhaving a higher error rate. In a wireless sensornetwork, on an increase in the error rate, the averageend to end delay increase considerably for all threeschemes. If proper adoptive schemes are used in awireless network, the average end to end delayencountered by the data packets can be considerablyreduced, which is proved in the superiorperformance exhibited by all schemes.Future work:This work had implemented an energyefficient channel adaptive MAC protocol in awireless sensor network with static nodes. Thisscheme had provided improvement gain in Energyefficiency, throughput, bandwidth, average end toend delay and delivery ratio. But the superior natureof this scheme depends on many environmentalfactors. Such as operation scenarios, specific datatype etc.REFERENCES T. van Dam and K. Langendoen, “Anadaptive energy efficient MAC protocol forwireless sensor networks,” in ACM SenSys03, November 2003. Heinzelman, W. R., A. Chandrakasan, et al.(2000). "Energy-efficient communicationprotocol for wireless micro-sensornetworks." System Sciences, 2000.Proceedings of the 33rd Annual HawaiiInternational Conference. V. Rajendran, K. Obraczka, and J. Garcia-Luna-Aceves, “Energy-efficient, collision-
Dharam Vir, Dr. S.K. Agarwal, Dr. S.A.Imam / International Journal of Engineering Researchand Applications (IJERA) ISSN: 2248-9622 www.ijera.comVol. 3, Issue 3, May-Jun 2013, pp.335-340340 | P a g efree medium access control for wirelesssensor networks,” in ACMSenSys 03,November 2003. Iqbal, Q. R., D. J. Holding, et al. (2007)."Energy efficient radio range allocation inad hoc wireless sensor networks."Proceedings of the 2007 spring simulationmulti-conference-Volume 1: 58-65. Al-Obaisat Y, Braun R “On WirelessSensor Networks: Architectures, Protocols,Applications, and Management”Auswireless Conference 2006. M. Takai, J. Martin, and R. Bagrodia.“Effects of wireless physical layermodeling in mobile ad hoc networks”. In2nd ACM International Symposium onMobile Ad Hoc Networking & Computing(MobiHoc 2001), pp. 87–94. ACM Press,October 2001. I. Rhee et al., “Z-MAC: a Hybrid MAC forWireless Sensor Networks”. ACM SenSys,San Diego, CA, Nov. 2005. M. Zorzi and R. R. Rao, “Energyconstrained error control for wirelesschannels,” IEEE PersonalCommunications, pp. 27-33, December1997. W. Ye, J. Heidemann, and D. Estrin. "AnEnergy-Effcient MAC Protocol forWireless Sensor Networks". INFOCOM2002, volume 3, pp 1567-1576, June 2002 Li, J., D. Cordes,"Power-aware routingprotocols in ad hoc wireless networks."Wireless Communications, IEEE: 69-81.(2005). V. Rajendran, K. Obraczka, and J. Garcia-Luna-Aceves, “Energy-efficient, collision-free medium access control for wirelesssensor networks,” in ACMSenSys 03,November 2003. Giridhar, A. and P. R. Kumar (2005).Maximizing the functional lifetime ofsensor networks, IEEE Press Piscataway,NJ, USA. S. Singh and C. Raghavendra. PAMAS:Power aware multi-access protocol withsignalling for ad hoc networks. ACMSIGCOMM Computer CommunicationReview, vol. 28(3):pp. 5–26, July 1998. Y. Tseng, C. Hsu, and T. Hsieh. “Power-saving protocols for IEEE 802.11-basedmulti-hop ad hoc networks”. In 21stConference of the IEEE Computer andCommunications Societies (INFOCOM2002), vol. 1, pp. 200–209. New York, NY,June 2002. QualNet 5.0 user manual.http://www.scalablenetworks.com