An Attack-Resistant Localization Algorithm in Wireless Sensor Networks

Many sensor network applications require sensor´s location information. Despite the great advances, most of the existing localization algorithms for sensor networks are vulnerable in hostile environments. Almost all the localization algorithms need the location information of reference nodes to locate the unknown nodes. When the location information is tampered by the attacks, the accuracy of these algorithms will degrade badly. This paper presents a novel algorithm to tolerate malicious attacks against reference nodes. The algorithm filters out malicious location information by Metropolis-Hasting (MH) filter rule. A simulation circumstance which might be attacked is constructed to compare the improved algorithm with original one. The experiment results demonstrate that the proposed algorithm can effectively survive malicious attacks even if the attacks bypass authentication.