Cluster-Based Minimum Mean Square Estimation for Secure and Resilient Localization in Wireless Sensor Networks

Abstract

To support a wide variety of applications ranging from military surveillance to health care clinic monitoring, a wireless sensor network must obtain accurate location for each sensor. A number of localization schemes have been developed to allow each sensor node to acquire its location. However, most of these techniques assume benign environments, and thus cannot survive malicious attacks in hostile environments where external and/or compromised nodes may launch attacks. This paper proposes a new computationally efficient and resilient localization scheme based on the clustering of benign location reference anchors. Moreover, this paper reports both simulation and field experiments using a test-bed of MICAz motes performed to compare the proposed approach with several recent secure localization schemes. The experimental results demonstrate that the proposed scheme has the fastest execution time among all resilient localization schemes that can be used for the current generation of sensor platforms (e.g., MICA series of motes).