More accuracy approach for signal subspace-based algorithms in bistatic EMVS-MIMO radar

Abstract In this article, we rediscuss the parameter estimation in a bistatic multiple-input multiple-output radar system with electromagnetic vector sensor (EMVS) and propose an improved approach for all signal subspace-based estimation algorithms. First, the signal subspace is obtained by directly performing SVD or high-order SVD on the signal matrix or the signal tensor. Then, the elevation angle is automatically estimated by exploiting the rotation invariance of the receive-transmit array manifold and the Joint diagonalization technology. Then, the estimated elevation angle is taken as prior information, and the spatial response vector is recovered from the entire signal subspace using least-squares by exploring the property of the Kronecker product. Next, the azimuth angle is estimated by the ‘Vector Cross-Product’ strategy. Finally , the polarization parameters are calculated based on the least-squares method. The proposed algorithm is analyzed from the aspects of identifi-ability and complexity. The proposed signal subspace acquisition method is less computationally intensive. The improved parameter estimation approach can realize automatic parameter pairing and have a better parameter estimation performance than the previous corresponding algorithms when it works on the subspace obtained in different ways. Simulation results verify the performance improvement of the proposed algorithm.