MIMO Radar Sensitivity Analysis of Antenna Position for Direction Finding

Abstract

Sensitivity analysis plays an integral role in many engineering design problems. The purpose of this paper is to investigate the direction finding sensitivities (DFSs) with respect to antenna position uncertainties (APUs) for multiple-input multiple-output (MIMO) radar with colocated antennas. These uncertainties cause differences between the MIMO radar virtual array manifold used by direction finding (DF) algorithms and the true array manifold, which is named as “calibrated errors.” To evaluate the effects of such errors on DF, the DFSs relative to APUs are considered from two following approaches. First, we use the first-order sensitivity analysis for MIMO radar. For a given arbitrary antenna geometry, the formulas of DFSs using maximum likelihood (ML) algorithm are developed for relatively small APUs. In addition, the formula for computing the ambiguity thresholds of the ML algorithm as a function of target separation and other DF system parameters are derived for relatively large APUs. Alternatively, the DFSs are only concerned with antenna geometry, i.e., the virtual array manifold, being regardless of any certain DF algorithm. Herein, we extend Manikas's method to MIMO radar. To assess the importance of each antenna in a given MIMO radar system, we derive the antenna importance function (AIF) which is defined as the amount of varieties of manifold vectors from the APUs. Furthermore, to compare the robustness to APUs for different antenna geometries, we derive the overall system sensitivity (OSS) for MIMO radar systems. In numerical example section, we show the previous DFS analysis results by several representative MIMO radar antenna geometries. The presented sensitivity analysis could be as the guideline of MIMO radar system analysis and design.