FGLS-Based Optimal Performance Approximation Method for TDOA Localization System

In this paper, a localization algorithm based on feasible generalized least squares (FGLS) is proposed to solve the problem of time difference of arrival (TDOA) localization performance degradation caused by unknown measurement error correlation. First, the nonlinear least squares is used to obtain the unbiased estimation of the position. Subsequently, the estimated position is used to estimate the covariance matrix, with some structural assumptions. Finally, the estimated covariance matrix is used to carry out the generalized least squares to achieve high-precision localization. The simulation of static experiments show that the proposed method reduces the localization error and obtains a performance closer to the Cramer-Rao lower bound (CRLB). In addition, the method proposed in this paper is expected to be extended to other passive location systems and has broad application prospects.