Performance Comparison Of Ad Hoc Wireless Network Routing Protocols Hu - PDF document

Performance Comparison Of Ad Hoc Wireless Network Routing Protocols Humaira Ehsan' and Zartash Afzal Uzmi National Univer-si@ of Computer and Emerging Sciences, Islamabad, Pakistan. Lahove University of Management Sciences, Lahore, Pakistan. humaira_79@yahao.com , zartash@Zums. edu.pk i 2 Abstract In recent yean tnuny protocols .fbr ad hoc wireless jiebvwnrks hav2 been developed b14f veq~ Me iybrrnution is available on the free because waste a lot bandwidth and packets inay never reach their destination. route failures and a stable a result of node protocol should handle stale and types the free that few become isolated out of transmission range manner, different formed and such routing protocol and suggest some are normally not symmetric bidirectional, therefore protocol should unidirectional links. There are many existing protocols which their fimctionality routing protocol should destination pair order to reduce Multiple routcs are these networks the a lot reduce delay protocol should not overload congestion near The performance protocol should by increasing decreasing the energy constrained some random period handle such nodes without causing pararneters can routing protocols Path Optiinality Average end order delivery Routing Overhead following parameters varied while routing protocols protocols I. Network size-measured Average number single node. measured in Traffic pattern routing protocols wireless networks which are: Table driven routing demand routing protocols called proactive protocols. These protocols routes between network and routes information periodic route update The updates been developed modifying the distance vector statc algorithms. Protocols store into various routing tables. updates thesc protocols converge and generate routing overhead are not very suitable hoc wireless existing table driven protocols are: Sequenced Distance Vector Clustered Gateway Switch Routing Protocol This protocol RIP routing table. This sequence number field differentiate between fresh routes routes 121. Each node maintains a routing which contains next hop reachable destinations. Each entry routing table consists reach the sequence number received sequence number associated with each determine the node receives new information about particular route with the kept while two updates with same sequence number routing table is updated advertisemmts or whenever Full dumps some route changes the whole routing table the latest updated information is the protocol critically depends periodic update interval value. this value very sinall large routing overhcad because updatcs and very smaIl highly unfavorable for which have high mobility a large in this maintain the valid all the time. Routes these protocols demand routing protocols. demand routing protocols are: Distance Vector Routing Temporally Ordered Routing source routing the determines the entire path through which packet should header. Source dynamic source routing. Each node maintains a route in the route cache contain The route determined either making a route discovery process. When then it is used directly and this recursively until thc complete path packet hcadet along that that In route discovery process the sender generates a route request packet contains in which next hop unique request request packet. The source node broadcasts neighbors receive this them has corresponding entry route reply the initiator coinplcte route will further broadcast the route request. reply packet source using the path the route request packet also use some nodes should exponential backoff while sending a route reply nodc which shortest path that they can learn ncw well as reply promiscuous listening increases greater number to bc periodic updates has routing fact that route in every packet. This overhead bursty traffic. maintains a routing table but other nodes. broadcast route discovery method similar dynamic source routing. dynamically creates entries the routing tables a packet the routing Each neighboring node receives that If there the touting table this node from which received the request packet. routing table a period which tbe request packet can propagatc through the whole network and reply packet. forwarded until reaches some node cntry for the destination its routing table reaches the destination. route rcply The entries routing tables the intermediate nodes form the reverse route reply packet travels the reverse node which reply packet a forward pointer from which that path is source to which data packets travel later protocol assumes that unidirectional links. uses the same to send a reply sequence numbers to determine which and which which This protocol uses a periodic hello messages to determine local connectivity. This mechanism protocol is not as much that for nodes. Thc protocol finds multiple routes destination pair. new toute the user mobility scenarios but high mobility when links minimize reaction topological changes. potentially far-reaching control message propagation from the rate rate The basic hnctionality of the protocol consists creating routes, maintaining routes and erasing routes. models the network undirected. Each directed from to node node maintains a metric “height”. used in assigning directions with each can be Reactive mode route creation requires the destination node. done by constructing a acyclic graph rooted the destination using queryireply process. a route required the source broadcasts a until it more routers have a route to the destination. router that destination sends receives a packet sets height one the height of destination initiates route creation (optimization) packet, then processed null height. Routers with a used for computations. link failure loses its last designed such the number participate in failure reaction the purpose a performance comparison performance simulations are performed hoc routing protocols simulations are done have used three i.e. normalized routing packet delivery average end measure performance. packet. Each routing packet one transmission. scalability of routing overhead increases with the It is the packets received sent. This packets dropped the difference between sending time receiving time includes all possible delays buffering during route discovery latency, queuing sources used continuous bit which it packets based thc network's both the time at which and the node when when For the simulations the sending rate is fixed to second and generate node movement scenarios. field configuration simulation time. rate decreases. simulations nodes are within the specified field random movement other position. movement speed distributed between following parameter values used for generating various Field configuration: The simulation results are presenied the basis for all the which result lesser number mobilitv reaches cases because on-demand routing techniques. table driven routing protocol number of demand routing increases the Normalized routing various number show a comparison between delivery fraction Normalized routing load caching and protocols for on demand converge to almost time reaches the number destination therefore various number The average end delay for for all cases. protocols and increases as the mobility increases. with an increased number there are more link failures therefore there route discoveries. AODV takes discovery process this leads This paper compared four main hoc routing protocols which while DSDV a tablc driven protocol. other protocols scenarios and routing load to end delays while packet delivery mobility scenarios. because it caching aggressively maintains multiple simulation time frequently and cache mostly therefore in better David B. Johnson and Davis A. Maltz, “Dynamic Source Routing in Ad Hoc Network,” Thielinski and H.Korfh. Eds. Mobile Computing, Ch. 5, Kluwer, 1996. [6] J. Broch et al., “A Performance Comparison of Ad-Hoc MultiHop Wireless Networks Routing Protocols,” Protocols,” Vincent D. Park and M. Scott Corson, “A highly adaptive distributed routing algorithm algorithm David B. Johnson, ”Routing in Ad Hoc Networks of Mobile Hosts,” Proc. lEEE Whp. Mobile Cump. ond Apps., Dec. 1994, pp. 158-163. [9] Charles E. Perkins and Elizabeth Elizabeth D. Chakeres and Elizabeth M. Royer, “The Utility Hello Messages Messages 111 S. Corson and J. Macker, “Mobile Routing Protocol Performance Issues Evaluation Considerations,” Considerations,” Charles E. Perkins and Pravin Dynamic Destination-Sequenced Distance-Vector Distance-Vector S. R. Das et al., “Comparative Performance Evaluation of Routing Protocols Protocols 141 P. Johansson et al., “Routing for Mobile Mobile 2. J. Hass, “A New Routing Protocol Reconfigurable Wireless Wireless Shree Murthy and J. J. Garcia-Luna-Aceves, “A Routing Protocol for Packet Radio Networks”.