An Area-Based Approach for Node Replica Detection in Wireless Sensor Networks

Abstract

Typical wireless sensor networks have restricted resources on memory capacity, computing/processing power, and energy supply. However, wireless sensor network has been increasingly applied to various applications and security in wireless sensor networks has become an essential and challenging issue. There is a wide range of security attacks in wireless sensor networks. A node replication attack is a type of attacks in wireless sensor networks. An attacker can be easily disguised in a wireless sensor network and acts as an intermediate node to intercept data packets. The attacker could be a clone node in the network by capturing nodes' ID, the pair-wise key etc. The base station may not be able to distinguish between good nodes and malicious nodes. In this paper, we present a new method, named Area-Based Clustering Detection (ABCD) method, for detecting node replication attacks. Our simulation results show that the proposed ABCD method can achieve high successful detection rate while decrease the communication overheads when compared with the Line-Selected Multicast (LSM) method previously proposed in the literature. The proposed ABCD method can also maintain the network lifetime and decrease the number of stored messages when compared with a centralized approach.