Fast PARAFAC Algorithm for Target Localization in Bistatic MIMO Radar in the Co-Existence of Unknown Mutual Coupling and Spatially Colored Noise

Junjie Sui,Fang Ye,Xinhai Wang,Fangqing Wen

doi:10.1109/access.2019.2961454

Junjie Sui, Fang Ye + Show 2 more

Open Access

https://doi.org/10.1109/access.2019.2961454

Copy DOI

Abstract

Target positioning using multiple-input multiple-output (MIMO) radar system has aroused extensive attention in the past decade. However, most of the existing positioning algorithms are only suitable for ideal scenarios (e.g., well-calibrated sensors, orthogonal waveforms, Gaussian white noise). In this paper, we focus on the target localization problem in a bistatic MIMO system in the co-existence of mutual coupling and spatially colored noise, i.e., to estimate the direction-of-arrival (DOA) and direction-of-departure (DOD) in such non-ideal scenario. To tackle this issue, a parallel factor (PARAFAC) analysis algorithm is proposed. Firstly, the de-noising operation is carried out to suppress the spatially colored noise. Then a PARAFAC analysis model is constructed to explore the tensor nature of measurement. After computing PARAFAC decomposition on the new tensor, the factor matrices associate with DOD and DOA are obtained. Thereafter, the de-coupling operation is followed to eliminate the mutual coupling. Finally, the idea of least squares is utilized to recovery DOD and DOA from the factor matrices. The proposed algorithm can achieve closed-form solution for DOD and DOA estimation without pairing calculation. Moreover, it provides better estimation performance than the state-of-the-art approaches. Detailed analyses are illustrated and simulation results verify the effectiveness of the proposed algorithm.

Highlights

INTRODUCTIONMultiple-input multiple-output (MIMO) radar has drawn massive attention [1], [2]
In the past decade, multiple-input multiple-output (MIMO) radar has drawn massive attention [1], [2]
We focus on target positioning using a bistatic MIMO radar, which is a special kind of colocated MIMO radar and can be extended to multistatic MIMO radar case

Summary

INTRODUCTION

Multiple-input multiple-output (MIMO) radar has drawn massive attention [1], [2]. J. Sui et al.: Fast PARAFAC Algorithm for Target Localization in Bistatic MIMO Radar proposed, such as multiple signal classification (MUSIC) [5], [6], maximum-likelihood (ML) [7], [8], estimation method of signal parameters via rotational invariance technique (ESPRIT) [9], sparsity-aware approaches [10], [11] and tensor-based methods [12]–[15]. Tensor decomposition has been turned out to be a highly efficient tool to explore the tensor algebra Owing to their superior de-noising performance, tensor approaches often provide better estimation accuracy than matrix-based methods. Speaking, the spatial cross-correlation method suffers from the lose of array DOF, the matrix completion is computationally inefficient, the covariance differencing method must to solve the ambiguity problem, while the temporal cross-correlation can avoid the above drawbacks. The superscript (X)T ,(X)H ,(X)−1 and(X)† stand for the operations of transpose, Hermitian transpose, inverse and pseudo-inverse, respectively; ◦, ⊗, and ⊕ represent, respectively, the outer product, the Kronecker product, the Khatri-Rao product (column-wise Kronecker product) and the Hadamard product; diag (·) denotes the diagonalization operation, toeplitz (t) returns a symmetric Toeplitz matrix constructed by the vector t; angle (·) denotes the angle of a vector

TENSORS AND PARAFAC ANALYSIS

MC MATRIX OF A ULA

DATA MODEL

COMPLEXITY

SIMULATION RESULTS

CONCLUSION