Abstrato

An Efficient Genetic Approach Based Geocasting Protocol for Video Streaming in MANET

Saranya.S, P.D.R. Vijayakumar

Today’s need and trend of communication is supported by the primary services called group oriented services. In recent years, such services are provided with the help of MANETs. There are constraint while developing multicast routing protocols which starts with the residual energy of the battery of a mobile node is finite (before recharging is done). Genetic Algorithm (GA) presents an improved solution for the multi-constrained multicast routing problem. GA allows for self-configuration systems and maintains state information about the neighbouring network. And we obtain the alternative path or backup path to avoid reroute discovery in the case of link failure or node failure. By choosing proper fitness function and values for metrics such as initial population size, crossover and mutation that closely relates to the chosen scenario, the genetic algorithm optimizes the routes in terms of selected metrics. To implement this group communication, we propose an Efficient Geographic Multicast Routing protocol (EGMP) with the help of virtual zone based structure along with the GA for better efficiency in communication. This EGMP protocol deals with the position information which is used to construct zone structure, multicast tree and multicast packet forwarding. The performance metrics such as Packet Delivery Ratio (PDR), End to End delay and Control Overhead of EGMP are also evaluated through simulations and quantitative analysis by varying number of nodes, zone size and group size. In order to overcome problems such as performance degradation or network partition, data can be replicated in mobile nodes. The replicas ensure that the performance is not degraded. However, chances are that some nodes may not cooperate and rather behave selfishly. The selfishness of some of the nodes may lead to the performance degradation in terms of accessing data. In this paper, we examine the impact of selfish nodes in a mobile ad hoc network from the perspective of replica allocation. We term this selfish replica allocation. In particular, we develop a selfish node detection algorithm that considers partial selfishness and novel replica allocation techniques to properly cope with selfish replica allocation. The conducted simulations demonstrate the proposed approach outperforms traditional cooperative multicast routing protocol and replica allocation techniques in terms of packet delivery ratio, throughput, loss, data accessibility, communication cost, and delay.