Enhancing Sensor Position Accuracy Using CWLS Multisource Localization in Wireless Sensor Networks

This paper addresses the challenge of enhancing the accuracy of sensor positioning in a wireless sensor network by leveraging time difference of arrival (TDOA) measurements from multiple disjoint sources. We present a novel sensor position refinement method based on the constrained weighted least squares (CWLS) multi-source localization algorithm, which yields more accurate estimates of source positions. Utilizing these improved estimates, we generate a set of nonlinear range measurements for each sensor. These new observations, combined with the available prior information about sensor locations, form the basis of our final sensor refinement formulation, which follows the maximum likelihood estimation strategy. Although the resulting optimization problem is inherently non-convex and NP-hard, we develop an iterative algorithm to optimize the sensor positions that efficiently addresses this complexity. Additionally, our proposed algorithm is designed for distributed implementation, ensuring compliance with real-time requirements. Extensive Monte Carlo simulations demonstrate that our method significantly outperforms existing approaches, particularly in scenarios with moderately high TDOA noise level or ill-conditioned geometry.