Source Localization Using TDOA Measurements From Underwater Acoustic Sensor Networks

Abstract

In this letter, we investigate the problem of TDOA localization with unknown signal propagation speed in underwater acoustic sensor networks (UASNs). Although the state-of-the-art semidefinite programming (SDP)-based algorithm has good accuracy, its computational complexity is huge. We develop a two-step algorithm to estimate the source position and signal propagation speed. It not only has better accuracy than the SDP-based algorithm but also has much less computational complexity. The first step of the proposed algorithm is a least squares (LS) estimator, and the second step of the proposed algorithm is a fixed point iteration (FPI) estimator. Numerical simulations validate that the proposed algorithm is superior to the SDP algorithm and reaches the Cramér-Rao lower bounds (CRLB).