A Closed-Form Location Algorithm Without Auxiliary Variables for Moving Target in Noncoherent Multiple-Input and Multiple-Output Radar System

This paper focuses on the moving target localization and velocity measurement in incoherent centralized multiple-input and multiple-output (MIMO) radar systems with widely separated antennas in 3D space. In this paper, we assume that parameters such as time-of-arrival (TOA), frequency-of-arrival (FOA), azimuth angles and elevation angles have already been measured. With these measurements, a final closed-form solution of the target position and velocity can be obtained via the weighted least squares (WLS) method. When the target is located in the far field, due to poor system observability, a first-order Taylor expansion based on the WLS solution is necessary to obtain a more accurate and unbiased solution. Unlike the preceding papers which were based on the two-stage weighted least squares (2SWLS) method [1]-[3], in this paper, the angle information is introduced into the time delay equations and the Doppler frequency equations, so that the intermediate variables in the estimator can be eliminated [4], [5]. Meanwhile, the time delay equations and the Doppler equations are transformed into linear equations only related to the position and the speed of the target. This method, unlike 2SWLS-based methods [1]-[3], does not introduce auxiliary variables, so it does not require the decorrelation procedure. Simulation results show that the root mean-square error (RMSE) of position and velocity can reach Cramer-Rao lower bound (CRLB) when the noise is at a moderate level before the thresholding effect occurs.