The Performance Evaluation of Genetic Zone Routing Protocol for MANETs

1. Introduction.
A mobile ad hoc network (MANET) is a collection of wireless mobile nodes that form a temporary network on the fly that operates without the support of any fixed network infrastructure. MANETs are created dynamically and they provide special challenges beyond those in standard data networks [1]. Some examples of the possible uses of ad hoc networking [3], [4] include students using laptop computers to participate in an interactive lecture, business associates sharing information during a meeting, soldiers relaying information for situational awareness on the battlefield, and emergency disaster relief personnel coordinating efforts after a hurricane or earthquake. In such networks, each mobile node operates not only as a host but also as a router and cooperates dynamically to establish routing among them to discover "multi-hop" paths through the network to any other node. There are various issues related to ad hoc networks [4]. Several protocols have been proposed for routing in such an environment. These protocols can broadly be classified into two types: proactive and reactive routing protocols. Proactive or table-driven protocols try to maintain routes to all the nodes in the network at all times by broadcasting routing updates in the network. Examples are Destination Sequenced Distance Vector Routing (DSDV) and Fisheye State Routing (FSR). On the other hand, reactive or on-demand protocols attempt to find a route to the destination, only when the source has a packet to send to the destination. Examples are Dynamic Source Routing (DSR), Ad hoc On demand Distance Vector (AODV), and Temporarily Ordered Routing Algorithm ( TORA). In between the above two extremes, there are the hybrid protocols. The Zone Routing Protocol (ZRP) is a hybrid protocol. ZRP is a routing framework composed of the proactive IntrAzone Routing Protocol (IARP), reactive IntErzone Routing Protocol (IERP), and the Bordercast Resolution Protocol (BRP).

Dept. of ISE, EWIT


Page 1

The Performance Evaluation of Genetic Zone Routing Protocol for MANETs

2. Conventional Routing Protocols.
Several protocols have been proposed for routing in a wireless environment. These protocols can broadly be classified into two types: Proactive routing protocols and Reactive routing protocols. Proactive or table-driven protocols try to maintain routes to all the nodes in the network at all times by broadcasting routing updates in the network. Reactive or on-demand protocols attempt to find a route to the destination, only when the source has a packet to send to the destination. Let us see how these protocols function in some detail.

2.1) Proactive Routing Protocol.

Motivation [5] to both improve protocol convergence and reduce traffic has led to the development of proactive path finding algorithms that combine features of the distance vector and link state approaches. Each node constructs a minimum spanning tree based on knowledge of its neighbors¶ minimum spanning trees and the link costs to each neighbor.

Realizations of the path finding algorithms, like the wireless routing protocol (WRP), are able to eliminate the counting-to-infinity problem and reduce the occurrence of temporary loops, often with less control traffic than traditional distance vector schemes.

The main disadvantage of WRP is in the fact that routing nodes constantly maintain full routing information in each network node, which was obtained at relatively high cost in wireless resources. Routing protocols that are based on a source initiated query/reply process have also been introduced. Such techniques typically rely on the flooding of queries to discover a destination. In the temporally ordered routing algorithm (TORA), the resulting route replies are also flooded in a controlled manner to distribute routes in the form of directed acyclic graphs (DAG¶s) rooted at the destination. In contrast, protocols such as dynamic source routing (DSR) and AODV unicast the route reply back to the querying source along a path specified by a sequence of node addresses accumulated during the route query phase. In the case of DSR, the node addresses are accumulated in the query packet and are returned to the source to be used for
Dept. of ISE, EWIT 2010-11 Page 2

if a physical subnet becomes partitioned).The Performance Evaluation of Genetic Zone Routing Protocol for MANETs source routing. as it is based on hopping within virtual networks. on the other hand. the reliance on packet flooding may still lead to considerable control traffic in the highly versatile wireless network environment.2) Reactive Routing Protocol. only when the source has a packet to send to the destination. which are determined by the sources and the destination addresses and not by the nodes¶ geographical locations. the dynamic assignment of unique node addresses can be quite challenging in an ad hoc network. EWIT 2010-11 Page 3 . 2. Furthermore. The Reactive or on-demand protocols attempt to find a route to the destination. The virtual network routing is an interesting idea. even if the address assignment is centrally controlled within each physical subnet (i. in which the nodes that belong to the source¶s physical and virtual networks are polled about the address of the destination. Routing is based on temporary addresses. of ISE. In particular. a query phase is initiated by the source. especially when innovative route maintenance schemes are employed.. duplicate addresses may arise. The on-demand discovery of routes can result in much less traffic than standard distance vector or link state schemes. A temporary address is a concatenation of the node¶s address on each one of the two networks. In order to communicate with a node. Dept.e. distributes the discovered routes in the form of next hop information stored at each node in the route. a physical and a virtual network. a node is required to acquire a new temporary address. However. There are routing algorithms for ad hoc networks where each node belongs to two networks. Upon physical migration. However. AODV. the routing can be far from optimal.

IERP is a family of reactive routing protocols that offer enhanced route discovery and route maintenance services based on local connectivity monitored by IARP.The Performance Evaluation of Genetic Zone Routing Protocol for MANETs 3. The conventional routing protocols have significant disadvantages such as waste of network capacity since in proactive scheme most of the routing information is never used. ZRP Overview. the search throughout the network. reactive IntErzone Routing Protocol (IERP) and the Bordercast Resolution Protocol (BRP). reducing organizational overhead. On one hand. So what is needed is a protocol that initiates the route determination procedure on-demand. The local routing information is referred to quite often in the operation of the ZRP. of ISE. On the other hand. increasing reliability and performance. as opposed to querying all the network nodes. Correspondingly. EWIT 2010-11 Page 4 . These protocols are discussed in some detail in following sections. The zone routing protocol (ZRP) provides efficient and fast discovery of routes by integrating the two radically different classes of traditional routing protocols. ZRP is a routing framework composed of the proactive IntrAzone Routing Protocol (IARP). The ZRP [5] is an example of a hybrid reactive/ proactive routing protocol. allowing optimal routes to be discovered.1) The ZRP. Dept. is performed by efficiently querying selected nodes in the network. and in reactive schemes a global search is carried out every time a route request is encountered at a particular node in the network. Routing is flat rather than hierarchical. 3. it limits the scope of the proactive procedure only to the node¶s local neighborhood. although it is global. However. The protocol identifies multiple loop-free routes to the destination. but at limited search cost. minimizing the waste associated with the purely proactive schemes. the most appealing feature of the protocol is that its behavior is adaptive. and reducing the threat of network congestion. based on the current configuration of the network and the behavior of the users. IARP maintains routing information for nodes that are within the routing zone of the node.

The Performance Evaluation of Genetic Zone Routing Protocol for MANETs 3. Thus. expressed in hops. but rather nodal connectivity (hops).2) The Notion of a Routing Zone and Intrazone Routing. The remaining nodes are categorized as interior nodes. Dept. EWIT 2010-11 Page 5 . we depict zones as circles around the node in question. K L C B J G D E F H S A I Figure 1: A routing zone of radius two hops. (Authors: Pearlman. 1 as follows. Node L is outside S¶s routing zone. nodes A±F are interior nodes while G±K are peripheral nodes. 1. Haas. Note that in this example. However. Peripheral Nodes are nodes whose minimum distance to the node in question is exactly equal to the zone radius. The nodes of a zone are divided into Peripheral Nodes and Interior Nodes. of ISE. Marc R. Zygmunt J. An example of a routing zone (for node S) of radius two hops is shown in Fig. The routing zone has a radius r. in Fig. nodes A±K are within the routing zone of the central node S. [5] ) For the purpose of illustration. Each node is defined with a separate routing zone and zones of neighboring nodes overlap. the zone is not a description of physical distance.

the peripheral node forwards the query to its peripheral nodes. the path to the destination is known. and a route to all the nodes in its zone. EWIT 2010-11 Page 6 . If so. We assume that a node learns its zone through some sort of a proactive scheme. and no further route discovery processing is required.The Performance Evaluation of Genetic Zone Routing Protocol for MANETs Because each node maintains its own routing zone. The IERP operates as follows: the source node first checks whether the destination is within its zone. More on this in section 3. 3. If the destination is not within the source¶s routing zone. distance to. the study on the topic suggests that the tradeoffs are not strongly affected by the particular choice of the proactive scheme used.4.3) Interzone Routing and the ZRP For Reconfigurable Wireless Networks¶ the coverage of a routing zone is relatively small compared to the size of the network. This is achieved by a packet delivery service called Bordercasting. the amount of update traffic required to maintain a routing zone does not depend on the total number of network nodes (which can be quite large). Thus. the source bordercasts a route request to all its peripheral nodes. (Note that a node has the information about the identity. Now. If not. The IERP is responsible for reactively discovering routes to destinations located beyond a node¶s routing zone. and the desired routing information cannot be immediately provided by the IARP.) If so. of ISE. most destinations lie outside of a node¶s routing zone. Because the updates are only propagated locally. While the performance of the ZRP depends on the choice of IARP implementation. The knowledge of the routing zone topology allows a node to efficiently continue the propagation of a query in the more likely case that destination can be found. the zones of neighboring nodes can heavily overlap. a route reply is sent back to the source indicating the route to the destination. in turn. which we refer to here as the IARP. all the peripheral nodes execute the same algorithm: they check whether the destination is within their zone. Dept. Each node is assumed to maintain routing information only to those nodes that are within its routing zone.

overlapping queried routing zones).The Performance Evaluation of Genetic Zone Routing Protocol for MANETs An example of this route discovery procedure is demonstrated in Fig. H sends the query to B. 2. In particular. the design objective of query control mechanisms should be to reduce the amount of route query traffic by steering threads outward from the source¶s routing Dept. which recognizes D as being in its routing zone and responds to the query.e.. F C D E B H S G A Figure 2: An example of IERP operation (Authors: Pearlman. Zygmunt J. each one of these nodes forwards the query by bordercasting the query to its peripheral nodes. S first checks whether D is within its routing zone. and H. excess route query traffic is a result of overlapping query threads (i. Now. EWIT 2010-11 Page 7 . This depicts that. [5] ) On the downside IERP may result into heavy traffic due to the fact that the zones overlap heavily. in turn. A key feature of the routing zone is that a node¶s response to a route query contains information about that node¶s entire routing zone. Otherwise S bordercasts a query to its peripheral nodes that is S sends a query to nodes C. S knows the route to D. This is called as control traffic problem. If so. Haas. G. Thus. of ISE. To find a route within the network. Marc R. after verifying that D is not in its routing zone. The source node S needs to send a packet to the destination D. indicating the forwarding path: S±H±B±D.

This problem is addressed primarily through appropriate mechanisms of query detection and query termination. 3.4) Bordercasting Routing Protocol (BRP) The construction of a routing zone requires a node to first know who its neighbors are. of ISE. A neighbor is defined as a node that can communicate directly with the node which is one hop away. Instead of broadcasting packets. Bordercasting could be implemented through network layer unicasting or multicasting of messages to the peripheral nodes which prevents non-peripheral nodes from accessing the bordercasted messages. bordercast packet delivery service is provided by the Bordercast Resolution Protocol (BRP). IERP forwards queries with BRP. Bordercasting utilizes the topology information provided by IARP to direct query request to the border of the zone. EWIT 2010-11 Page 8 . ZRP uses a concept called bordercasting. Route updates are triggered by NDP. Dept. neighbor discovery may be implemented through a MAC-level Neighbor Discovery Protocol (NDP). In other cases.The Performance Evaluation of Genetic Zone Routing Protocol for MANETs zone and away from each other. IERP uses the routing table of IARP to respond to route queries. BRP uses the routing table of IARP to guide route queries away from the query source. Identification of a node¶s neighbors may be provided directly by the Media Access Control (MAC) protocols. which notifies IARP when the neighbor table is updated. The fact that the topology of the local zone of each node is known can be used to reduce traffic when global route discovery is needed. In this case.

1 Genetic algorithm Approach. iii. 4. multicast routing problem [7-9]. Genetic Zone Routing Protocol. of ISE. Dept. Initial Population: Generate random population of n chromosomes (suitable solutions for the problem. New population: Create a new population by repeating following steps until the new population is complete. Fitness: Evaluate the fitness f(x) of each chromosome x in the population. Genetic Algorithms perform much better with rugged landscapes because of their population based approach spreading ³probes´ throughout the search space. in this case µroutes¶) ii. A simple genetic algorithm consists of the following steps: i. Note that one of the big advantages of genetic algorithms is the ability to parallelize them on a large scale by spreading the evaluations across different machines. Investigators have applied GAs to the shortest path routing problem [6]. EWIT 2010-11 Page 9 . In the study we preferred genetic algorithm as the optimization algorithm because of the confidence that it would work due to its robustness.The Performance Evaluation of Genetic Zone Routing Protocol for MANETs 4. A large amount of work has been done on the application of genetic algorithms or evolutionary algorithms to communications networks.

Any on-demand routing protocol uses route discovery and route maintenance procedures for finding the path to the destination which in turn return a single shortest path to the destination. Loop: Go to step (ii). and return the best solution in current population. EWIT 2010-11 Page 10 . If no crossover was performed. In this case. stop. In this study. Test: If the end condition is satisfied. the bigger chance to be selected) b. ZRP is a hybrid protocol which has the hands on routes if the destination is within the routing zone. The idea of using GA leads to finding multiple shortest (near shortest some times) paths to provide load balancing and tolerance in the networks.The Performance Evaluation of Genetic Zone Routing Protocol for MANETs a. vi. GA has been applied at the border nodes in order to find the alternative routes. d. 4. In any case. Mutation: With a mutation probability mutate new offspring at each locus (position in chromosome). Replace: Use new generated population for a further run of algorithm. This causes unnecessary wastage of network resources and also wastage of time. it makes use of route discovery with IERP and BRP. Selection: Select two parent chromosomes from a population according to their fitness (the better fitness. Crossover: With a crossover probability cross over the parents to form a new offspring (children). the actual problem comes when the destination is outside the zone. iv. then rediscovery of the new shortest path is required. offspring is an exact copy of parents. the border node will apply GA by Dept. if that route fails and/or congestion occurs and that route leads to the delays or packet losses. Instead of rediscovering the path to the destination every time on failure of the existing path due to link/node failure and/or congestion occurrence. This route discovery goes with border nodes between the zones. c. Accepting: Place new offspring in a new population. v. of ISE. Genetic Zone Routing Protocol (GZRP) is an extension of ZRP adopting the concept of Genetic Algorithm (GA).2 Application of Genetic Algorithm to ZRP. However.

route. of ISE. EWIT 2010-11 Page 11 . frequency. The first route to the destination in the list is considered the default route. The destination entry indicates the destination node of packets.The Performance Evaluation of Genetic Zone Routing Protocol for MANETs making use of topological database available with that node to find alternative routes that may be shortest or near shortest paths.1) Routing Table at border node Table 1 is a routing table generated at border node with BRP. This reduces the load on a single route by equally distributing the packet delivery through the available alternative routes.2. there are set of alternative routes. 4.2. and metric. For each destination. the routing table is empty. « « 4. Sateesh Kumar. S.. When a packet is generated at a node. The routing table consists of the entries including destination. This field will be useful in order to load balance the network. The default metric used throughout the work is hop count. Ramachandram [11] ) Destination Route 7-3-5 Frequency 7000 2000 1000 20000 18000 12000 « Metric 5 7-2-6-5 7-3-2-6-5 4-9-8 Hop Count 8 4-5-8 4-5-9-8 Hop Count .. A route entry is a list of node IDs along the route.2) Encoding Method Dept. Table 1: Routing table at border node with BRP ( Authors: P. The frequency entry specifies the number of packets sent to the destination by the route. a default route is generated by the IARP routing framework and will be inserted in to the routing table. In initial state.

Each locus of the chromosome represents an order of a node (indicated by the gene of the locus) in a routing path.e. Population consists of the number of chromosomes. of ISE. B1 B2 -------- Bk-1 Bk D S is the Source node D is the Destination node Bi is the ith Border node in the routing path k is the route size ( excluding D) 4.2. An example of chromosome encoding is given below: S Where. The population size is needed to increase exponentially with the complexity of the problem (i. A larger population is quite useful but it demands excessive costs in terms of both memory and time. It never needs more than N number of nodes in a network to form a routing path. Dept. EWIT 2010-11 Page 12 .The Performance Evaluation of Genetic Zone Routing Protocol for MANETs The first step in GA is to encode the elements of chromosomes. Recent studies have shown. One of the issues in selecting the initial population of chromosomes is its size. The gene (node) of the first locus is always reserved for the source node. Further. however. Hence. the maximum size of a chromosome length can be N.3) Initial Population. that satisfactory results can be obtained with a much smaller population size. the literature suggested that random initialization method can be adopted in order to generate the new population. A chromosome encodes the problem by listing up node IDs from its source node to destination node based on topological information (routing table of IARP part of that node) of the network. length of the chromosome) in order to generate good solution. A chromosome (routing path) of the GA consists of sequences of positive integers that represent the IDs of nodes through which a route path passes.

j = { 1. which plays the role of a chromosome map (masking) providing information on whether the link from node i to node j is included in a routing path or not. of ISE. The costs are specified by the cost matrix C=[Cij]. The fitness function has a higher value when the fitness characteristic of the chromosome is better than others. respectively.  Ii. if the link from node i to node j exists 0. EWIT 2010-11 Page 13 . Source and destination nodes are denoted by S and D.4) Fitness Function.j). otherwise The objective function can be formulated as.2. There is a cost Cij associated with each link (i.j). where Cij denotes a cost of transmitting a packet on link (i. The fitness function of GAs is generally the objective function that requires to be optimized. Ci.A). The fitness function in the shortest path routing problem is obvious because the shortest path computation amounts to finding the minimal cost path.j  = { 1.The Performance Evaluation of Genetic Zone Routing Protocol for MANETs 4.j · Ii. The underlying topology of multi-hop networks can be specified by the directed graph G=(N.j Subject to the condition that the path between source and destination can have no loops only when. -1 0. if i=S if i=D otherwise Dept. It can be defined as follows: Ii. Fitness calculation is most crucial in the GA operation. Each link has the link connection indicator denoted by Iij. where N is a set of nodes (vertices) and A is a set of its links (arcs and edges).j Ii.

One partial route connects the source node to an intermediate node and the other partial route connects the intermediate node to the destination node.5) Selection. of ISE. the proposed GA technique employs the roulette wheel selection which is most widely used. Physically.The Performance Evaluation of Genetic Zone Routing Protocol for MANETs And. However. the same chromosome should not be picked twice as a parent. There are many techniques like roulette wheel selection. EWIT 2010-11 Page 14 . The selection (reproduction) operator is intended to improve the average quality of the population by giving the high-quality chromosomes a better chance to get copied into the next generation.6) Crossover. However. 0 if iD if i=S 4. but there is no requirement that they be located at the same locus. In the proposed scheme. tournament selection and truncation selection for selecting the chromosomes for next generation. In this technique.2. But the mechanism of the crossover is not the same as that of the conventional one-point crossover. Dept. Crossover examines the current solution in order to find the better ones.2. 4.  Ii. crossover in the shortest path routing problem plays the role of exchanging each partial route of two chosen chromosomes in such a manner that the offspring produced by the crossover represents only one route. the crossover does not depend on the position of nodes in routing paths. That is. This dictates selection of one-point crossover as a good candidate scheme for the proposed GA. two chromosomes are selected based on the probability. two chromosomes chosen for crossover should have at least one common gene except for the source and destination nodes. stochastic remainder selection.j = { ” 1.

2. S B1 B2 B3 B5 B6 B7 D S B2 B4 B5 D It is possible that loops are formed during crossover. Physically. the bias can be ignored. S After mutation. However. thereby keeping away from local optima.The Performance Evaluation of Genetic Zone Routing Protocol for MANETs Before crossover. The population undergoes mutation by an actual change or flipping of one of the genes of the candidate chromosomes. Repair and penalty functions are the usual counter measures. A topological information database is utilized for the purpose. S B2 B1 B3 D B2 B4 B5 D Dept. it generates an alternative partial-route from the mutation node to the destination node in the proposed GA. Mutation node. mutation may induce a subtle bias. S S B2 B1 B3 B2 B5 B4 B6 B5 B7 D D After crossover. Before mutation.7) Mutation. Of course. 4. of ISE. EWIT 2010-11 Page 15 . A simple counter measure must be taken in this regard.

and its efficiency in terms of consuming node battery power. Packet delivery ratio is important as it describes the loss rate that will be seen by the transport protocols. This metric characterizes both the completeness and correctness of the routing protocol.2. ZRP and GZRP. 5. The metrics used for evaluation are (i) Packet Delivery Ratio (PDR). The parameters used in the simulation model are summarized in following tables. As mentioned earlier. Control Overhead is an important metric for measuring scalability of a protocol. Average end-to-end delay is important as it describes quality of link establishment and the ability of forwarding of the packets to the intended nodes. Table2 summarizes the parameters used for GA and Table-3 summarizes the parameters used for Dept. it is excluded from the candidate nodes forming the rest of the path. of ISE.The Performance Evaluation of Genetic Zone Routing Protocol for MANETs Depending on the mutation point. It must be noted that none of the chromosomes of the initial population or after the mutation is infeasible because when once a node is chosen. EWIT 2010-11 Page 16 . Our aim of simulation study is to investigate the impact of node mobility on the performance of both the protocols.8) Repair Function. One of the nodes. connected directly to the node at mutation point. crossover may generate infeasible chromosomes that violate the constraints of generating loops in the routing paths. The simulator used for evaluation of the protocols is GloMoSim (Global Mobile Information System Simulator) [10]. a gene from the chosen chromosome is selected (node B2).1 Evaluation Methodology. the degree to which it will function in congested or low-bandwidth environments. 5. 4. Experimental procedure. (ii) Total Overhead (iii) Average end-to-end Delay. which in turn affects the maximum throughput that the network can support. is chosen randomly as the first node of the alternative partial route.

The network interface is modeled after IEEE 802. of ISE.The Performance Evaluation of Genetic Zone Routing Protocol for MANETs modeling the simulation to evaluate the protocol. EWIT 2010-11 Page 17 . Ramachandram [11] ) Parameter Crossover Rate Crossover Point Mutation Rate Mutation Point Population Size Selection Technique Size of alternative routes in the routing table 60% Value Randomly selected 2 point crossover.11 product. Sateesh Kumar. Roulette wheel method 3 Dept. S.5 % Chromosome size / 2 2 times the number of nodes. No data was collected for the first 10 seconds to avoid measurements before intrazone route discovery process stabilized. Table 2: Parameters used for GA ( Authors: P. 0.

Mobility model Minimum Speed: 1 m/s. Total Overhead and Average end-to-end delay at different mobility conditions of the nodes. Still. . S.The Performance Evaluation of Genetic Zone Routing Protocol for MANETs Table 3: Parameters of simulation. there is an improvement nearly 5% in all the mobility conditions. Source: CBR Traffic Number of sources: 10 Rate: 1 packet/sec. 200 m. .2 Results Analysis. As shown in Figure 3.900s. . Sateesh Kumar. the change in PDR for GZRP is almost similar to ZRP for different mobility conditions.11 2 Mbps. ( Authors: P. This section provides the comparative study of both ZRP and GZRP for different evaluation strategies like PDR. 5. MAC Bandwidth Transmission range 802. Maximum Speed: 20 m/s. This improvement is seen due to availability of alternative routes even in the link/node failures. Ramachandram [11] ) Parameter Simulation time Number of experimented trials Network coverage area Number of nodes Routing zone radius 900 Sec. . of ISE. 100s. EWIT 2010-11 Page 18 . Dept. 5 1500 X 300 m2 50 2 Value Model: Random Way Point Pause Time: 0s. Node Placement: Random. . Packet size: 512 bytes.

GZRP gives almost 20% to 30% better results compared to ZRP. Ramachandram [11] ) Dept. This may be due to less or no usage of the alternative routes in the lower mobility conditions. The delay incurred was almost same. There is an improvement in the reduction of control overhead in the network due to GZRP as high as 35% and as low as 15%. At a very high mobility condition. Sateesh Kumar S. This is a very good result compared to ZRP. EWIT 2010-11 Page 19 . both the protocols behave like the same. Figure 3: Packet Delivery Ratio for ZRP and GZRP ( Authors: P.The Performance Evaluation of Genetic Zone Routing Protocol for MANETs The effect of average end-to-end delay on the mobility of the nodes is shown in Figure 4. The reason behind the results may be due to reduction is congestion due to load balancing of the network. This reduced the control overhead due to rediscovery of the new routes whenever there is a need. of ISE. The results show that delay factor is very much reduced as the pause time is less (high mobility). At lower mobility conditions or when the nodes are not moving. as the mobility of the nodes comes to a stable condition. the results are proved to be good and there is a considerable reduction in control overhead of around 35%. But. Even at the high mobility conditions. This is a good improvement. GZRP has utilized the advantage of availability of alternative routes. there is improvement of 15% of reduction in control overhead. The Figure 5 shows the effect of control overhead on the mobility of the nodes.

Sateesh Kumar S. of ISE.The Performance Evaluation of Genetic Zone Routing Protocol for MANETs Figure 4:End-to-End Delay for ZRP and GZRP ( Authors: P. Sateesh Kumar S. Ramachandram [11] ) Dept. EWIT 2010-11 Page 20 . Ramachandram [11] ) Figure 5:Control Overhead for ZRP and GZRP ( Authors: P.

In this paper. The results are proved to be interesting. These alternative routes are more efficiently used by GZRP to work in congestion situations and also at the node/link failure times. EWIT 2010-11 Page 21 . This protoco1 uses Genetic Algorithmic approach to find a limited set of alternative routes in order to load balance the network. we have proposed an extension to the Zone Routing Protocol with the use of genetic algorithm. Dept. GZRP is more efficient compared to ZRP as it reduces considerably the average end-toend delay and control overhead.The Performance Evaluation of Genetic Zone Routing Protocol for MANETs 6. We have called this new routing protocol as Genetic Routing Protocol (GZRP). of ISE. The results show that there is a reduction at the maximum in delay by 30% and in control overhead by 35%. Conclusion.

edu/projects/glomosim. Prentice Hall. Changjia. and H. G. Dept. of ISE. Int. 1998. 2001 Willium Stalings. Sato. References. Y. Ramanathan and J. 40(2): 20 . [9] Z.Comput. Pearson Education. S Ramchandram. "Determining the Optimal Configuration for the Zone Routing Protocol". Royer. ³A genetic algorithm for the multiple destination routing problems´. IEEE Communications Magazine. Xu. and Z. IEEE Trans. Communication Technology (WCC-ICCT 2000).8.cs. pp. India. Conf. 1998. C. EWIT 2010-11 Page 22 . Man. [7] J. ³A genetic algorithm for determining multiple routes and its applications´. ³The Performance Evaluation of Genetic Zone Routing Protocol for MANETs´. pp2774-2779. IEEE Journal on Selected Areas in Communications. 2000 [4] Elizabeth. J H. IEEE Int. Evol. IEEE Int. No.ucla. Conf. August 1999 [6] M. Nov. ³ A migration scheme for the genetic adaptive routing algorithm´.The Performance Evaluation of Genetic Zone Routing Protocol for MANETs 7. pp. 2. pp.B. 137-140. Mobile Communications. New Jersey. Wireless communications and Networks. Haseyama. [1] R. [5] Pearlman. in Proc. Gang. [8] Y. 1999. Proc. Journal of Theoretical and Applied Information Technology. Kitajima. 2005-2008.2. and Y. Marc R. 1248-1253. roh: "A Review of Current Routing Protocols for Ad hoc Mobile Wireless Networks". 150-161. 2000. Hass. Li.2002 [2] [3] Schiller. [10] GloMoSim: "Global Mobile Information Systems Simulation Library" available at http://pcl. Circuits and Systems. Redi. Xiawei. [11] P. April 1999. Sateesh Kumar. Symp. IEEE Personal Communications.. Systems. Munemoto. Zygmunt J. and Z. Takai. and Cybernetics. in Proc. Vol 17. Leung. Inagaki. "A brief overview of ad hoc networks: challenges and directions". Chai-Keong. ³A genetic algorithm for multicasting routing problem´. vol. M.

Master your semester with Scribd & The New York Times

Special offer for students: Only $4.99/month.

Master your semester with Scribd & The New York Times

Cancel anytime.