Generalized neuron based secure media access control protocol for wireless sensor networks

Abstract

Security plays a pivotal role in most applications of wireless sensor networks. It is common to find inadequately secure networks confined only to controlled environments. The issue of security in wireless sensor networks is a hot research topic for over a decade. This paper presents a compact generalized neuron (GN) based medium access protocol that renders a CSMA/CD network secure against denial-of-service attacks launched by adversaries. The GN enhances the security by constantly monitoring multiple parameters that reflect the possibility that an attack is launched by an adversary. Particle swarm optimization, a popular bio-inspired evolutionary-like optimization algorithm is used for training the GN. The wireless sensor network is simulated using Vanderbilt Prowler, a probabilistic wireless network simulator. Simulation results show that the choice of threshold suspicion parameter impacts on the tradeoff between network effectiveness and lifetime.