Reliable and Efficient Routing Using Adaptive Genetic Algorithm in Packet Switched Networks

To identify the optimal route is a complex task in packet switched network because optimization depends upon a number of parameters. In this paper Genetic Algorithm is used to locate the optimal route. Genetic Algorithm starts with a number of solutions where each solution is represented in the form of chromosome using the permutation encoding scheme. The success of Genetic Algorithm depends upon the number of operators such as selection, mutation and crossover. Needless to say crossover is most innovative. In this paper crossover operators proposed namely 1-point, 2-point, and adaptive, have been customized according to the need of computer network. The fitness of each solution is evaluated in terms of historical reliability factor, node success/failure and delay. The performance of the proposed approach has been compared with Dijkstra Algorithm and improvement has been observed.