AN ARTIFICIAL INTELLIGENCE BASED LIGHT WEIGHT CRYPTOGRAPHIC ADDRESS GENERATION (LW-CGA) USING OPTIMIZATION FOR IPv6 BASED MANETs

Due to rising application requests and also for dependable data transfer, security concern is one the important research areas in the field of mobile ad hoc networks (MANET). As we know that MANET is formed by a number of mobile nodes and each node communicate with each other using their energy capabilities. To solve the security and routing protocol in IPv6 based MANET, the Secure Neighbor Discovery (SeND) routing protocol will be designed with concept of artificial intelligence techniques to overcome the security threats during autoconfiguration has proven to face security and technical issues in MANETs. The SeND routing protocol uses RSA and SHA-1 (Secure Hash Algorithm) implementation for ensuring privacy enabled auto-configuration. In Internet Protocol Version (IPv6) based MANETs, the neighbor discovery enables nodes to self-configure and communicate with neighbor nodes through auto-configuration. The Stateless address auto-configuration (SLAAC) has proven to face several security issues during the data transmission form source node to destination node. Even though the SeND uses Cryptographically Generated Addresses (CGA) to address these issues, it creates other concerns such as need for Cryptographic Address (CA) to authenticate hosts, exposure to CPU exhaustion attacks and high computational intensity. In this work, we will present empirically strong Light Weight Cryptographic Address Generation (LW-CGA) using entropy gathered from system states using artificial neural network techniques with route optimization. In this article, we present the effect of attackers in the IPv6 based MANET with their prevention technique. For preventing the network from attacker, genetic algorithm along with the artificial neural network is used and the parameters such as Throughput, Delay, BER and Energy consumptions are measured and compare with the simulator with using genetic algorithm and artificial neural network for prevention of network.