Compressive parametric GLRT detector for airborne MIMO radar

In this paper, an improved target detection algorithm for MIMO airborne radar has been proposed namely, Compressive Parametric Generalized Likelihood Ratio Test (CP-GLRT). The Parametric Generalized Likelihood Ratio Test (P-GLRT) and Generalized likelihood ratio test (GLRT) are also studied considering the availability of secondary data. The signal is assumed to be spatially and temporally colored. The noise signal is modeled as an auto-regressive process. Performance of the proposed CP-GLRT detector is compared with existing P-GLRT and GLRT detectors. Detailed comparative results have been obtained for different compression ratios as well as different number of samples. It is observed that by using compressive parametric GLRT, the computational burden on the system can be reduced. However this reduction in computational burden is achieved as a trade off with the performance of the detector. Moreover, the performance of the CP-GLRT can be improved by choosing an adaptive measurement matrix from the transmitted signal subspace.