A robust algorithm for multiple disjoint moving sources localization with erroneous sensor locations

This paper considers the problem of multiple disjoint sources localization using time difference of arrivals (TDOA) and frequency difference of arrival (FDOA) with sensor position and velocity errors. By exploiting the observation that the multiple disjoint sources exhibit the same sensor location errors, the influence of sensor location errors can be greatly alleviated. The CTLS technique, as a natural extension of LS when noise occurs in all coefficients and the noise components of the coefficients are algebraically related, is more appropriate for the above problem, and a localization algorithm based on constrained total least-squares (CTLS) for estimating the positions and velocities of multiple disjoint moving sources with sensor location errors is proposed. Simulation results show that the proposed estimator achieves remarkably better performance than the two-step weighted least squares (WLS) approach, which makes the Cramér-Rao lower bound (CRLB) at a sufficiently high noise level before the threshold effect occurs.