Moving Target Detection for Distributed MIMO Radar under Clutter Model Mismatch

In this paper, we consider the problem of moving target detection (MTD) in distributed multiple-input multiple-output (MIMO) radar in non-homogeneous clutter environments. Due to the non-homogeneous environment and the azimuth-selective back-scattering of the background, the clutter varies from one cell to another and from one transmit-receive pair to another. We consider two types of clutter models. The first model assumes Gaussian clutter distribution in the Doppler space. The mean and variance are different from different transmit-receive paths. The second model characterizes the clutter as the superposition of multiple discrete scatterers of unknown powers and Doppler frequencies. To mitigate the effect of mismatch in the first model, we represent the clutter as a weighted linear combination of multiple basis vectors and use maximum likelihood techniques to estimate the optimum weight values. For the latter case, we address off-grid discrete clutter mismatch via jointly estimating the discrete clutter powers and Doppler frequencies. For both cases, moving target detection is performed using primary data (cell under test), thus avoiding the use of secondary data and associated stationarity issues. The performance of the proposed moving target detectors is validated using simulation examples.