Secure Sensor Localization in Wireless Sensor Networks based on Neural Network

Localization of sensor nodes in wireless sensor networks (WSNs) is very important since it associates spatial context with the data collected by sensor nodes and used in applications. Rapid development of wireless sensor technologies and wide applications of wireless sensor networks have also made security in sensor localization a primary concern as well as a great challenge. Without adequate security measures, the performance of sensor localization, e.g., the accuracy of localization results, cannot be ensured in hostile environments. In this paper, we propose a trust-based secure sensor localization scheme (TSLS) for WSNs following the theory of neural network (NN). The proposed TSLS scheme can ensure that unknown sensor nodes will get credible information to perform localization through the evaluation of beacon nodes in a WSN. The evaluation model is comprised of the evaluation of both the identity and the behavior of beacon nodes as well as a filtering mechanism to deal with the slander behaviors between beacon nodes. Simulation results show that the proposed TSLS scheme can improve the accuracy of sensor localization in hostile environments for both static and dynamic WSNs.