Energy Efficient Approach to Target Localisation in Wireless Sensor Networks

This paper investigates a general class of centralized-distributed optimization algorithm for in-network target localization in wireless sensor networks, WSNs. The main objective is reducing the amount of energy and bandwidth used for computations as well as for communications. The proposed algorithm is based on incremental gradient optimization technique. Applying results from signal detection theory, a criteria for sensor self activation and participation is formulate. The optimization scheme is extended to the case wherein signal measurements are corrupted with additive white Gaussian noise. Results demonstrated the potential of the proposed algorithm for applications in practical wireless sensor networking.