Published on

  • Be the first to comment

  • Be the first to like this

No Downloads
Total Views
On Slideshare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide


  1. 1. Salman Abdullah Alhamoodi, Krishna Kumar Raman / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 2, Issue 5, September- October 2012, pp.144-148Performance Analysis with AODV Routing Protocol for Wireless Adhoc Network Salman Abdullah Alhamoodi, Krishna Kumar Raman M.Tech (CSE), ManavRachna International University Faridabad (Haryana) Associative Professor CSE Dept ManavRachna International UniversityFaridabad (Haryana)Abstract Ad hoc networks are describe by Routing protocol (DSR) [6], [2] and Ad Hoc On-multihop wireless connectivity, frequently varying Demand Vector protocol (AODV) [10], [11]. DSRnetwork topology and the need for capable and AODV share an interestingdynamic routing protocols. We evaluate the frequentcharacteristic they both commence routingperformance of two important on- demand activities on an ―on demand‖ source. Thisrouting protocols for portable ad hoc networks reactivecharacter of these protocols is animportantDynamic source Routing (DSR) and Ad Hoc On- departure from more traditional proactiveprotocolsDemand Distance Vector Routing (AODV). A [4], that find routes among all source-destinationcomprehensive simulation model with MAC and pairs regardless of the utilize or need of such routes.substantial layer models is used to study The key motivation after the design of on-demandcommunication between layers and their protocols is the decrease of the routing load. Highpresentation implications. We demonstrate that routing pack usually has a significant impact on loweven while DSR and AODV share a similar on- bandwidth wireless links.demand performance, the differences in the While DSR and AODV share the on-protocol mechanics can lead to important demand performance[9] in that they initiate routingperformance differentials. The performance behavior only in the presence of data packets indiscrepancyis analyzed using unreliableload on require of a route, many of their routing mechanicsnetwork, node mobility and size of the network. are very dissimilar. In particular, DSR uses sourceBased on the performance, we can find out the steering, but AODV uses a table-driven routingsuperior protocol among DSR & AODV outline and destination succession numbers. DSR does not rely on any timer-based actions, but AODVKeywords: DSR, AODV, Proactive Routing, does to a certain amount. One of our goals in thisThroughput, PDF study is to take out the relative merits of these apparatus. The motivation is that animprovedINTRODUCTION understanding of the virtual merits will serve as a In an ad hoc network, mobile nodes cornerstone for expansion of more successful routingconversewith each other using multi-hop wireless protocols for wireless ad hoc networks.links. There is no motionless infrastructure such as The rest of the paper is organized as follows. In thebottom stations. Each node in the network also acts following segment, we briefly review the DSR andas a router, forwarding informationpackets for other AODV protocols. In segment III, we present thenodes. A central confront in the design of ad related work cited in literature. Section IV explores ahocnetworks is the growth of dynamic routing simulation model and results to perform detailedprotocols that can competently find routes among evaluation of the two protocols, focusing on thetwo communicating nodes. The routing protocol dissimilarity on their dynamic behaviors that canshould be able to keep up with the elevated degree of direct to performance dissimilarity. This lays downnode mobility that often modify the network many of the circumstance of the performance study.topology drastically and impulsively. Such networks The section concludes the comparison study.have been considered in the past in relation to coverresearch, often under the name of packet radio II. DESCRIPTION OF PROTOCOLSnetworks (see, for example, [7]). Recently there has A. DSRbeen a transformed interest in this field due to the The key characteristic of DSR [2], [6] is thefrequentavailability of low-cost laptops and palmtops use of source routing. That is, the sender knows thewith radio edge. Interest is also partly fueled by the absolute hop-by- hop route to the purpose. Thesegrowing passion in running common network routes are stored in a direction cache. Theprotocols in active wireless environments without the information packets carry the basic route in theobligation of specific infrastructures. packet descriptor. Our goal is to hold out a systematic When a node in the ad hoc network effort toperformance study of two energetic routing send aninformation packet to a destination for whichprotocols for ad hoc networks — Dynamic Source it does not previously know the route, it uses a route 144 | P a g e
  2. 2. Salman Abdullah Alhamoodi, Krishna Kumar Raman / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 2, Issue 5, September- October 2012, pp.144-148discoveryprocedure to dynamicallyresolve such a packets.route. Route detection works by flooding the Asignificant feature of AODV is thenetwork with route demand (RREQ) packets. Each maintenance of timer- based position in each node,node getting a RREQ, rebroadcasts it, unless it is the regarding the consumption of individual routingpurpose or it has a route to the purpose in its route table entries. A routing bench entry is ―expired‖ ifcache. Such a node replies to the RREQ with a not used lately. A set of predecessor nodes isdirection reply (RREP) packet that is routed back to maintained per steering table entry, which denotesthe original source. RREQ and RREP packets are the position of neighboring nodes that use this accessalso source routed. The RREQ builds up the path to route data packets. These nodes are notified withnegotiate so far. The RREP routes back to the source RERR packets if subsequent hop link breaks. Eachby traversing this conduit backwards. The route ancestor node, in turn, forwards the RERR toconceded back of the RREP packet is cached on the itspossesssets of predecessors, thus successfullysource for future use. erasing all routes using the broken link. If any connection on a source route is The recent condition of AODV [11]broken, the source node is notified using a route includes an optimizationprocedure to control thefault (RERR) packet. The source eradicates any RREQ flood in the direction discovery process. Itroute using this connection from its cache. A new uses an expanding ringto exploreinitially to discoverroute finding process must be commenced by the routes to an unidentified destination. In thesource, if this route is motionless needed. expanding ring investigate, increasingly larger DSR makes very destructive use of source neighborhoodsare exploring to find the destination.routing and direction caching. No special apparatus The explore is controlled by the TTL field in the IPto detect routing loops is desirable. Also, any description of the RREQ packets. If the route to aforwarding node accumulation the source route in a formerly known destination is needed, the prior hop-packet it forwards for probable future use. Several wise detachment is used to optimize the search.extra optimizations have been projected and havebeen estimated to be very effective by the authors of III. RELATED WORKthe protocol [9], as illustrated in the following. (i) Two current efforts are the most associatedSalvage: Atransitional node can use an alternate with our work, as they use the same ns-2basedroute foritspossess cache, when aninformation replication environment.Broch, Maltz, Johnson, Hupacket meets a failed link on its source route. (ii) and Jetcheva, the uniqueauthors of the simulationGratuitous route repair: A source node getting a model, assess four ad hoc routing protocols togetherRERR packet piggybacks the RERR in the with AODV and DSR [1]. They used only 50 nodesubsequent RREQ. This helps hygienic up the models with comparable mobility and trafficcaches of other nodes in the network that might have scenarios that we dilapidated. Traffic loads arethe failed link in one of the cached source routes. reserved low (4 packets/Sec, 10-30 sources, 64 byte(iii) Promiscuous listen: When a node overhears a packets). Packet deliverypart, the number of routingpacket not address to itself, it checks if the packet packets and allocation of path lengths were used asmight be routed via itself to gain a shorter direction. presentation metrics. An earlier description ofIf so, the node propels a gratuitous RREP to the AODV was used to lock the query controlsource of the route with this new, improved route. optimizations. DSR demonstrated vastly betterAside from this, promiscuous eavesdrop helps a routing load performance, and somewhat betternode to learn different routes lacking directly packet delivery and route length presentation. This isparticipating in the routing procedure. along the column of our observations for the loads that were measured. Routing load performance andB. AODV packet delivery ratio have enhanced, however, in the AODV [14], [11] split DSR’s on-demand current AODV model for equivalent loads, thoughcharacteristics in that it also determine routes on an DSR leftovers a superior protocol for lower loads―as needed‖ source via a parallel route discovery with a tiny number of nodes.process. However, AODV accept a very different A more modern work, Johansson, Larssson,apparatus to maintain routing information. It uses Hedman and Mielczarek [9] comprehensive theconventional routing tables, one entry per purpose. above work by using originalmobility models. ToThis is a departure from DSR, which can preserve distinguish these models, a new mobility metric ismultiple route cache entries per purpose. Without initiated that measures mobility in terms ofsource routing, AODV relies on routing bench comparative speeds of the nodes rather thanentries to broadcast a RREP back to the source and, unconditional speeds and pause times. Again, simplysubsequently, to route data packets for the purpose. 50 nodes were used. Aninadequate amount of loadAODV uses succession numbers maintained at each test was achieved, but the number of sources haspurpose to determine the freshness of routing been always small (15). Throughput, hold up andinformation and to avoid routing loops [10]. These routing load (both in quantity of packets and bytes)succession numbers are carried by all steering were precise. The AODV model used hello 145 | P a g e
  3. 3. Salman Abdullah Alhamoodi, Krishna Kumar Raman / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 2, Issue 5, September- October 2012, pp.144-148messages for district detection in addition to the link B. Number of nodes Vs Packet Dropcoat feedback. The DSR model did not use A packet is dropped in two cases: the bufferlicentiouslistening thus losing some of its is complete when the packet needs to be buffered andadvantages. In malice of the differences in the the instance that the packet has been bufferedreplica implementations, the overall observation was exceeds the boundary. Packet dropping wascomparable to ours. In low loads DSR was more experimental for more than a few nodes and variedefficient, while AODV was more effective at the nodes each instance and the dropped was countedsuperior loads. The packet wise steering load of at the target node during a complete recreation periodDSR had been almost always considerably lower whose quantity was as in Table 2.than AODV, though, though the byte wise routing Table 2. Packet dropped at different nodesload has been frequently compared. The authorsattributed the relative poor performance of DSR to Protocol Packet drop for number of nodesthe source-routing expenses in data packets. Theyused small information packets (64 bytes) thus 5 10 15 20manufacture things somewhat unfavorable for DSR. AODV 13 12 12 11With 512 byte packets, we didn’t locate source DSR 10 8 9 10routing overheads to be a very importantperformance issue. . There exists other effort that thecomparative presentation of these two on-demandprotocols with our own [4]. The performance of thetwo protocols was instigated similar. However, thereplication environment was rather limited, with noconnection or physical layer models. The steeringprotocol models also did not contain manyfunctional optimizations.IV. PERFORMANCE EVALUATIONA. Number of nodes Vs ThroughputThe count of nodes was varied every time and the Figure 2. Packet drop over number ofnodesthroughput was intended at the destination nodeduring a complete simulation period whose quantity Well organized protocols can intelligentlywas as in fig. 1. discover out routing way thus packets dropping speed Table 1.Number of nodes Vs Throughput reduces for them. The packet dropped for DSR is a smaller amount than that of AODV as it out performs Protoco Throughput for number of nodes with fewer nodes and no intermittent update is l maintained in DSR 5 10 15 20 DSR 10.7 72.38 206.98 368.45 C. Packet Received Vs Propagation Delay 3 Packet getting statistic was performed for AODV 40.9 431.4 1639.7 3759.8 more than a few propagation delays in case of all 2 0 2 0 Adhoc protocols, whose nature of packet difference AODV shows superior throughput than the becomes as in fig 3. DSR executes better when theDSR. The AODV has a large amount more direction- propagation stoppage of nodes increases becausefinding packets than DSR since the AODV avoids nodes turn out to be more stationary will guide tocircle and brightness of routes while DSR uses stale steadier path beginning source to target. DSR isroutes. Its throughput is superior to AODV at higher to AODV particularly when the node’selevated mobility. propagation stoppage begins to increase. Table 3. Packets received at different delays Protocol Packets received at propagation delay 10 20 30 40 50 AODV 0 453 596 596 596 DSR 0 336 504 676 820 For AODV, it shows important dependenceFigure 1.Throughput over number of nodes on route immovability, thus its packet received rate 146 | P a g e
  4. 4. Salman Abdullah Alhamoodi, Krishna Kumar Raman / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 2, Issue 5, September- October 2012, pp.144-148is lower. Although, the quantity of packetconventional is inversely proportional to propagationstoppage, DSR has better presentation than AODV. Figure 4. Throughput over different pause times E. Selecting the best routing protocolsFigure 3: Packet received over propagation delay Arithmetical analysis on the investigationD. Throughput Vs Simulation Time data shown in on top of mentioned tables and Throughput was gained at the target node consequent figures produces Table 5 where from theagainst different dimension of networks and diverse most excellent performing procedure withthe reproduction instance uniformly for each admiration to an exact network stricture can beprotocol whose calculate was as in fig 4. Throughput chosen to optimize MANET presentation.is the common rate of victorious message releaseover a communication channel. This data might be Table 5. Throughput at different simulation delaysdelivered over a physical or reasonable link, or pass PARAMETERS THE BESTduring a certain network node. The throughput is PROTOCOLfrequently measured in bits for each second Number of nodes Vs AODV(byte/Sec), and sometimes in data packets for each Throughput Number of nodes Vs Packet DSRsecond or data packet for each time slot. This is the Drop Packets received Vs DSRcalculate of how quickly a conclusion user is capable Propagation Delayto receive data. It is determined as the percentage of Throughput Vs Simulation AODV, DSRthe sum data received at required propagation time. TimeA superior throughput determination directly impactsthe user’s awareness of the superiority of service V. Conclusion:(QoS). Presentation analysis has been finished onTable 4. Throughput at different simulation delays two glowing known Adhoc networking protocols DSR and AODV. An inclusive simulation revise has Protocol Throughput at delay in ms been presented to evaluate these routing procedures 10 20 30 40 50 60 70 using an unreliable workload such as throughput, packet delivered, packet drop and broadcast delay of AODV 0 53 68 68 68 84 84 ad-hoc system. AODV is better in case of DSR 0 37 56 80 80 80 80 throughput. Based on the fig 4, it is shown that AODV References: [1] J. Broch, D. A. Maltz, D. B. Johnson, Y-and DSR have nearly same execution strategy C.Hu, and J. Jetcheva. A performancebecause of avoiding the configuration of loops and comparison of multi-hop wireless ad hocits uses stale routes in container of broken links. The network routing protocols. In Proceedings ofspeed of packet established for AODV is improved the 4th International Conference on Mobilethan the DSR because this intermittent broadcast also Computing and Networking (ACMadds a large in the clouds into the network. For MOBICOM’98), pages 85–97, OctoberAODV, the routing in the clouds is not likely 1998.pretentious as generated in DSR. For AODV, it [2] Josh Broch, David Johnson, and Davidshows important confidence on route constancy, thus Maltz.The dynamic source routing protocolits throughput is minor when the instance decreased. for mobile ad hoc networks.http://www.ietf.org/internet-drafts/ draft-ietf-manet-dsr-01.txt, Dec 1998. IETF Internet Draft (work in progress). [3] S.R. Das, R. Castaneda, J. Yan, and R. Sengupta. Comparative performance evaluation of routing protocols for mobile, ad hoc networks. In 7th Int. Conf. on 147 | P a g e
  5. 5. Salman Abdullah Alhamoodi, Krishna Kumar Raman / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 2, Issue 5, September- October 2012, pp.144-148 Computer Communications and Networks (IC3N), pages 153–161, October 1998.[4] Z. J. Haas and M. R. Pearlman. The performance of query control schemes for the zone routing protocol. In Proceedings of ACM SIGCOMM’98 Conference, pages 167–177, Vancouver,Sept. 1998.[5] Per Johansson, Tony Larsson, NicklasHedman, and BartoszMielczarek. Routing protocols for mobile ad-hoc networks - a comparative performance analysis. In Proceedings of the 5thInternational Conference on Mobile Computing and Networking(ACM MOBICOM’99), August 1999. to appear.[6] D. Johnson and D.Maltz. Dynamic source routing in ad hoc wireless networks.In T. Imielinski and H. Korth, editors, Mobile computing, chapter 5. Kluwer Academic, 1996.[7] John Jubin and Janet D. Tornow.The DARPA packet radio network protocols. Proceedings of the IEEE, 75(1):21–32, January1987.[8] J. Macker and S. Corson. Mobile ad hoc networks (MANET).http: //www.ietf.org/ html.charters/manet-charter.html, 1997.IETFWorking Group Charter.[9] David Maltz, Josh Broch, JorjetaJetcheva, and David Johnson.The effects of on- demand behavior in routing protocols for multihop wireless ad hoc networks.IEEE Journal on Selected Areas in Communication, 1999.to appear.[10] Charles Perkins and Elizabeth Royer. Ad hoc on-demand distance vector routing. In Proceedings of the 2nd IEEE Workshop on Mobile Computing Systems and Applications, pages 90–100, Feb1999.[11] Charles Perkins, Elizabeth Royer, and Samir Das. Ad hoc on demand distance vector (AODV) routing. http:// www.ietf.org/internet-drafts/ draft-ietf- manet-aodv-03.txt, June 1999. IETF Internet Draft (work in progress).[12] Magnus Frodigh, Per Johansson and Peter Larsson ―Wireless ad hoc networking-The art of networking without a network‖ Ericsson Review No. 4, 2000.[13] D. B. Johnson and D. A. Maltz, ―Dynamic Source Routing,‖ in Ad Hoc Wireless Networks, Mobile Computing, T. Imielinski and H. Korth (Eds.), Chapter 5, pp. 153– 181, Kluwer Academic Publishers, 1996.[14] David B. Johnson and David A. Maltz, ―Dynamic source routing in ad hoc wireless networks,‖ In Mobile Computing, Imielinski and Korth (Eds), chapter 5, pages153–181, Kluwer Academic Publishers, 1996. 148 | P a g e