Mobile Ad Hoc networks intrusion detection system against packet dropping attacks

Abstract

Due to the extreme lack of a stable infrastructure, also self-organization of network components, unpredictable network topologies, and the lack of a central authority for routing, security assurance in mobile ad hoc networks (MANETs) is an important and difficult challenge. Among the famous threat that MANETs suffer from: blackhole, grayhole, and selfishness attacks, because the target of these attacks is to drop packets and disturb the routing operation of the network. A scalable, reliable, and robust network intrusion detection system (NIDS) should be created to effectively combat these families of network layer routing assaults in order to offer high availability for MANETs. In this paper, we present a MANETs-IDS based on machine learning algorithm against blackhole, grayhole, and selfishness attacks with Ad Hoc on-demand distance vector (AODV) routing protocol (RFC 3561) and optimized link state routing (OLSR) potocol (RFC 3626), using ns-3 simulation platform. Our simulation took into consideration the density of the network and a random mobility model of nodes. The obtained experimental results show that the proposed detection algorithm reached very promoting performances (in term of accuracy, processing time, time to build the model, precision, recall, F-measure).