Security Analysis of AODV Protocol in Wireless Sensor Networks

Abstract

WSN leads to the vision of a connected world of physical and virtual processes, services and objects which are capable of providing multiple services within a network. WSN is a major part of future that mainly integrates and enables numerous communication solutions and technologies. WSN networks are always under threat of malicious attacks because of opening deployments in various domains. It's heteroge neous and distributed characteristics make conventional intrusion detection methodologies hard to deploy. WSN is a combination of a variety of nodes w ithin the same network, which work on unique addressing schemes. They are able to communicate with each other and cooperate with their surrounding nodes to reach common goals. One of the most challenging topics in WSN networks is security. Hence, Intrusion detection proves itself as a necessary and useful technique to keep the security and availability of WSN networks. This technique can monitor the security condition within the network. Further, they make an alert when an intrusion behavior is detected. I DS are categorized into signature - based and anomaly - based detection on the basis of technique in detecting an intrusion. Signature - based IDS depends on a set of pre - defined malicious activity patterns and attack signatures to detect intrusions while anomaly - based IDS relies on deviations from normal behaviors to detect intrusions. Si gnature - based IDS is better than an anomaly - based IDS in detecting previously known attacks, but the former is i neffective against unknown or polymorphic attacks. On the other hand, anomaly - based IDS is capable of detecting unknown attacks in the absence of a predefined pattern. In this thesis w e present analysis of AODV networks u n der black hole and flooding attac k. The networks have been evaluated under evaluation metrics like packet status, jitter and throughput . Also, each network has been analyzed using machine learning algorithms.