Title: Malicious node detection system in wireless sensor networks: a decentralised approach

Authors: Maryam Fanaeepour, Mohammad Reza Kangavari

Addresses: Computer Engineering Department, Iran University of Science and Technology, Narmak, Tehran, Iran. ' Computer Engineering Department, Iran University of Science and Technology, Narmak, Tehran, Iran

Abstract: The large deployment of wireless sensor networks (WSNs) increases their exposure to intrusions on sensor nodes. One of the most challenging issues is how to detect compromised or malicious nodes. Our goal in this paper is to review different techniques for detecting malicious nodes in WSNs and introducing a new system that we call HyPIDS. The main idea is using additional sensor nodes, called |monitor nodes|, inserted and distributed into the network to detect malicious nodes so that they monitor the propagated messages of common sensors through the network. We have proposed a distributed scheme by applying aggregator nodes but since we just work on monitor nodes at the simulation phase, we used a decentralised approach. We have modelled the solution with Petri nets (PNs). Also, the decentralised approach and detection of malicious nodes has been simulated with a new Java-based tool called SIDnet which supports heterogeneous sensor networks.

Keywords: wireless sensor networks; WSNs; security; malicious nodes; compromised nodes; intrusion detection; sensor nodes; generalised stochastic Petri nets; GSPNs; heterogeneous networks; SIDnet; node detection; wireless detection; simulation.

DOI: 10.1504/IJITST.2010.031473

International Journal of Internet Technology and Secured Transactions, 2010 Vol.2 No.1/2, pp.88 - 108

Published online: 05 Feb 2010 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article