Estimating the parameters of a moving target in MIMO radar with widely separated antennas

In this paper, we develop a new maximum likelihood (ML) moving target parameter estimation technique for multiple-input multiple-output (MIMO) radar. It is required for this technique that different receive antennas have the same time reference, but no synchronization of initial phases of the receive antennas is needed and, therefore, the estimation process is noncoherent. The target motion within a certain processing interval is modeled as a second-order polynomial whose coefficients are given by the initial location, velocity, and acceleration of the target. The proposed ML estimator is able to jointly process the data collected from multiple consecutive radar pulses. It is shown that the considered ML problem simplifies to the classic “overdetermined” nonlinear least-squares problem. The proposed ML estimator requires multi-dimensional search over the unknown location, velocity, and acceleration parameters. The performance of the proposed estimator is validated by simulation results.