Bi-iterative algorithm for joint localization and time synchronization in wireless sensor networks

This paper efficiently solves the problem of joint localization and synchronization in wireless sensor networks by using two-way message exchange mechanism. Employing an affine model of unsynchronized local clocks, we first establish the famous Maximum Likelihood (ML) estimator for joint localization and synchronization, which derives the best consistent solution. Unfortunately, achieving directly the ML estimator is usually difficult because the ML cost functions are highly non-linear and non-convex. In order to effectively obtain the ML estimator, a novel bi-iterative algorithm is proposed by alternately searching the sensor node location and clock parameters (clock skew and clock offset). We prove that the proposed algorithm is convergent, and analyze that its estimation accuracy for the sensor node locations and clock parameters can approximately reach the Cramer-Rao Lower Bound (CRLB) under the mild noise condition. Compared with some previous methods, the proposed bi-iterative algorithm is more computationally efficient, and takes fewer anchor nodes and less communication overhead. Simulation results verify theoretical analysis and show that the proposed algorithm has the better performance than the several existing algorithms.