Sinkhole Attack Detection and Prevention using Agent Based Algorithm

Abstract

This study presents sinkhole attack detection and prevention using agent-based algorithm. In this algorithm, agents are used to provide information to all node from its reliable neighbors by negotiation in three steps, thus nodes may not be able to pay the attention to the traffic made by sinkhole attacker. In this work, network scale of 500×500 m2 square areas have been considered. Series of simulation are carried in each experiment. Every simulation run is being organized to work for 10mins. Network performance is evaluated in terms of throughput, packet delivery ratio, jitter, delay in packets delivery, data packets received, data packets drop using network simulations software. Network simulation results depicts that in proposed algorithm, throughput increases by 15 to 20 percent, packet delivery ratio increases by 30 to 40%, decrease in the jitter by 10 to 15 %, delay in packets delivery is decreased by 15 to 20 %, data packets received are increased by 15 to 20 % and number of the data packets drop are decreased by 5 to 15 %. Based on simulation results throughput, packet delivery ratio and data packets received increased in proposed agent-based algorithm. However, it is observed that, jitter, delay in packets delivery and data packets drop were decreased.