Single-witness-based distributed detection for node replication attack

Existing approaches for detecting node replication attacks in wireless sensor networks have low detection rates, high communication costs and high memory costs. This paper presents a single-witness-based distributed detection (SWDD) method to node replication attacks. This method consists of 5 steps: selecting a witness node, generating the declaration information, sending the declaration information, verifying the witness nodes, and detecting the replication node. The method has a single witness node selection mechanism, uses random numbers to describe the node location, verifies the witness nodes using a multiple mapping mechanism and finally detects the replication nodes by the ultimate witness node. Simulations on OMNeT++ comparing this method with the single deterministic cell (SDC) and parallel multiple probabilistic cells (P-MPC) methods show that the SWDD method has a better detection rate, less communications and less energy consumption.