Clustering-Based Mitigation of Anonymity Attacks in Wireless Sensor Networks

The use of wireless sensor networks (WSNs) can be advantageous in applications that serve in hostile environments such as security surveillance and military battlefield. The operation of a WSN typically involves collection of sensor measurements at an in-situ Base-Station (BS) that further processes the data and either takes action or reports findings to a remote command center. Thus the BS plays a vital role and is usually guarded by concealing its identity and location. However, the BS can be susceptible to traffic analysis attack. Given the limited communication range of the individual sensors and the objective of conserving their energy supply, the sensor readings are forwarded to the BS over multi-hop paths. Such a routing topology allows an adversary to correlate intercepted transmissions, even without being able to decode them, and apply attack models such as Evidence Theory (ET) in order to determine the position of the BS. This paper proposes a technique to counter such an attack by reshaping the routing topology. Basically, the nodes in a WSN are grouped in unevenly-sized clusters and each cluster has a designated aggregation node (cluster head). An inter-cluster head routes are then formed so that the BS experiences low traffic volume and does not become distinguishable among the WSN nodes. The simulation results confirm the effectiveness of the proposed technique in boosting the anonymity of the BS.