Path Selection Method for the Statistical Filtering-Based Sensor Networks Using a Security Evaluation Function

Many sensor network applications are dependent on the secure operation of sensor networks, and will have serious consequences if the network is compromised or disrupted. Fabricated reports can be injected through compromised nodes, which can lead not only to false alarms but also to the depletion of limited energy resources in battery powered networks. Ye et al. proposed the statistical en-route filtering scheme to detect and drop such fabricated reports during the forwarding process. In this scheme, every node has a limited amount of information that is used to verify against injected false data, the detection power is affected by the selection of routing paths. In this paper, we propose a path selection method to improve the detection power of the statistical filtering. Each node evaluates the detection power of each incoming path from the base station and chooses the most secure path for data delivery against false data injection attacks. We compare our proposed method with the statistical en-route filtering scheme and the simulation results show that our proposed method protects against false data injection attacks. Key words: Sensor networks, false data filtering, secure routing path selection. 1