Target Localization in Bistatic EMVS-MIMO Radar Using Tensor Subspace Method

Abstract

Target localization is a fundamental task of multiple-input multiple-output (MIMO) radar systems with numerous applications. In this paper, we investigate into the localization problem in a bistatic MIMO radar with electromagnetic vector sensors (EMVS). Unlike the traditional scaler sensors, an EMVS is able to offer two dimensional (2D) direction finding, and it can provide additional polarization characteristics of the source. Therefore, target localization in bistatic EMVS-MIMO radar system involves 2D direction-of-departure (2D-DOD) and 2D direction-of-arrival (2D-DOA) estimation. Besides, we can obtain transmit polarization characteristics as well as polarization characteristics of the targets. To exploit the tensor nature of the array measurement after matched filters, a tensor subspace algorithm is developed, which estimates the target parameters via cross-product technique from tensor subspace. The proposed algorithm, which obtains closed-form solutions for parameters estimation, shows more accurate performance than the existing algorithm. Numerical simulations verify the effectiveness and improvement of the proposed algorithm.

Keywords

• Multiple-input multiple-output radar
• electromagnetic vector sensors
• higher order singular value decomposition analysis
• estimation