Wireless Sensor Network Internal Attacker Identification with Multiple Evidence by Dempster-Shafer Theory

Abstract

Wireless sensor Network (WSN) is known to be vulnerable to variety of attacks due to the construction of nodes and distributed network infrastructure. In order to ensure its functionality especially in malicious environments, security mechanisms are essential. Malicious or insider attacker has gained prominence and poses the most challenging attacks to WSN. Many works has been done to secure WSN from internal attacker but most of it relay on either training data set or predefined threshold. Without a fixed security infrastructure WSN need to find the internal attacker. Normally, internal attacker node behavioral pattern is different from the other neighbor good nodes in the system, but neighbor node can be attacked as well. In this paper, we use Dempster-Shafer theory (DST) of combined multiple evidence to identify the malicious or internal attacker in WSN. This theory reflects with the uncertain event or uncertainty as well as uncertainty of the observation. Moreover, it gives a numerical procedure for fusing together multiple pieces of evidence from unreliable neighbor with higher degree of conflict reliability.