Recursive moving target detection with distributed MIMO radar in clutter with non-homogeneous power

Our previous study addresses moving target detection (MTD) using a distributed multiple-input multiple-output (MIMO) radar in clutter with non-homogeneous power. The developed detector, referred to as the MIMO-GLRT detector, assumes perfect knowledge of the clutter subspace and uses the assumed clutter subspace to construct a projection matrix which is required to compute the test statistic. In this work, we take into account uncertainties on the clutter subspace, i.e., the subspace dimension is not known a priori, and develop a recursive version of the MIMO-GLRT detector by integrating a computationally efficient updating algorithm for the subspace projection matrix from one iteration to another, together with a generalized Akaike Information criterion (GAIC) for the subspace dimension selection. Simulation results with a synthesized dataset and a general clutter dataset are provided to demonstrate the performance degradation of the standard MIMO-GLRT detector when an over-estimated or under-estimated clutter subspace is used, and show that the recursive MIMO-GLRT detector is able to mitigate such degradation by choosing a proper clutter subspace directly from the received signals.