The objective of the proposed research is to develop and benchmark signal processing techniques that enable multiple-input, multiple-output (MIMO) approaches for ground moving target indication (GMTI) radar with non-orthogonal signals. The primary contributions of preliminary research herein have been to show how nonzero waveform cross-correlations degrade the effective virtual array and that inclusion of these terms is necessary in the derivation of optimal weights to maximize signal-to-interference-plus-noise ratio (SINR). The proposed research plans to expand current benchmarking to include space-time adaptive processing (STAP) performance prediction for non-orthogonal MIMO radar. A novel framework for MIMO clutter simulation is presented that allows for arbitrary matched filtered characteristics to be utilized without the additional complex task of waveform generation. The developed clutter simulation tools will be leveraged to characterize MIMO STAP performance under non-ideal collection environments (e.g., heterogeneous clutter and array errors) and to suggest processing approaches to recoup performance. The proposed research addresses gaps in MIMO STAP benchmarking with realistic non-orthogonal signals which may result in performance that does not match expectations.