Technology for Localization Attack Detection in Wireless Sensor Networks
|School||Harbin Institute of Technology|
|Course||Computer Science and Technology|
|Keywords||wireless sensor network localization location verification|
Wireless Sensor Networks have been proposed in various applications, such as environment monitoring, target tracking, geographical routing and so on. Many small, multi-function and low-cost sensor nodes are deployed in a special area. These nodes compose a large, application-related and self-organized network which has a dynamic topology.The localization of sensor nodes is an important application and it is the base of the numerous applications in wireless sensor networks. Many applications of WSN depend on the nodes’location information, such as supervision task distribution, route rules, overlay information, load equilibrium, topology control and so on. Therefore, locations of senor nodes play an important role in many location-based applications.When a sensor network is deployed in a hostile environment, adversaries can apply various attacks against the localization scheme, such as range enlargement or reduction attacks, wormhole attack and so on. Because location information of sensor nodes plays a critical role, the security of localization becomes a key problem in sensor networks.In this paper, we propose a novel location verification scheme LPV. Unlike other schemes, LPV works in a distributed manner and it doesn’t require any high-capability hardware. In LPV scheme, sensors utilize evaluations of the cooperative neighbors to complete the verification process. Another important contribution of our method is that we analyzed the impact of localization error on location verification and guarantee the accuracy of LPV. The results of simulation confirm the good performance of our method in different situations and show the resilience against attack.Furthermore, we also study the re-localization problem of anomaly nodes and introduce the centroid algorithm and MSL algorithm. In centroid algorithm, we find the area which covers the unknown node at first. Then we calculate the centroid of this area utilizing and choose the centroid as the location of the unknown node. MSL algorithm is based on Monte Carlo method. It doesn’t require that sensors are equipped with hardware to measure signal strengths, angles of arrival of signals or distances to other sensors. Like the LPV protocol, MSL algorithm completes the re-localization of the anomaly nodes utilizing neighborhood of the sensor nodes.