Target localization sensitivity using MIMO radar in the presence of antenna position uncertainties

In the context of widely distributed multiple-input-multiple-output (MIMO) radar, target localization accuracy is very sensitive to the knowledge of radar antenna positions. However, perfect knowledge of antenna positions is difficult to acquire even after calibration. This paper aims at investigating the impact of receiver antenna position uncertainties (APUs) on target localization performance. We model the APUs as random variables and derive the mean square error (MSE) as well as Cramér-Rao lower bound (CRLB) of location estimate. Simulations are provided to compare the obtained MSE and CRLB with those in the absence of receiver APUs cases in different antenna geometries. The results show that the target localization sensitivity (TLS) of MIMO radar is dependent not only on the bistatic geometry, but also on the APU power level as well as the signal-to-noise ratio (SNR).

I. Introduction

Active target localization has been the focus of considerable research efforts due to its usefulness in various applications. Particularly in [1]–[3], multiple-input-multiple-output (MIMO) radar systems with widely distributed antennas take advantage of the geographical spread of the deployed sensors and support target localization through either direct or indirect multilateration [4]. Therein, the impact of radar antenna position uncertainties (APUs) on the performance of localization algorithms was not taken into account, while the antenna geometry was mainly considered [5], [6]. However, in practice, the exact positions of antennas are not available quite often even after calibration and hence the localization performance may deteriorate significantly in the presence of antenna positions imperfections.