In this paper, we present a thorough performance analysis of three decision schemes recently introduced in in [1] the context of Multiple-Input Multiple-Output radars. Such algorithms, referred to as I-GLRT-D, A-GLRT-D, and TS-GLRT-D, have been derived using the Generalized Likelihood Ratio Test (GLRT) and address joint detection and localization in the delay-Doppler domain by exploiting target energy spillover between adjacent matched filter samples. Herein, we prove that I-GLRT-D, A-GLRT-D, and TS-GLRT-D possess the Constant False Alarm Rate property with respect to the power of the disturbance. In addition, we show by Monte Carlo simulations that they are substantially insensitive to mismatches between the nominal and the actual structure of the clutter covariance matrix under several illustrative scenarios. Finally, we study their selectivity behavior showing that they are more robust than the natural competitors introduced in [1] to possible mismatches between the actual and the nominal target steering vector.
Analysis of MIMO Radar Detection Algorithms With Location Capabilities: CFAR Property and Selectivity
D. Orlando;
2024-01-01
Abstract
In this paper, we present a thorough performance analysis of three decision schemes recently introduced in in [1] the context of Multiple-Input Multiple-Output radars. Such algorithms, referred to as I-GLRT-D, A-GLRT-D, and TS-GLRT-D, have been derived using the Generalized Likelihood Ratio Test (GLRT) and address joint detection and localization in the delay-Doppler domain by exploiting target energy spillover between adjacent matched filter samples. Herein, we prove that I-GLRT-D, A-GLRT-D, and TS-GLRT-D possess the Constant False Alarm Rate property with respect to the power of the disturbance. In addition, we show by Monte Carlo simulations that they are substantially insensitive to mismatches between the nominal and the actual structure of the clutter covariance matrix under several illustrative scenarios. Finally, we study their selectivity behavior showing that they are more robust than the natural competitors introduced in [1] to possible mismatches between the actual and the nominal target steering vector.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
