Two-Dimensional Off-Grid DOA Estimation Based on Sparse Bayesian Learning

In this paper, a two-dimensional off-grid direction of arrival (DOA) estimation strategy based on sparse Bayesian learning (SBL) is proposed for improving the two-dimensional DOA estimation accuracy of millimeter-wave multiple-input multiple-output (MIMO) radar in low signal-to-noise ratio (SNR) environment. First, the grid is employed alongside Taylor expansion to formulate the signal model. After that, the expectation maximization (EM) algorithm is employed in conjunction with the corresponding marginal probability density function to iteratively estimate the hyperparameters. Finally, the conditional probability density function of the received signal is utilized, from which the error weighting coefficients for off-grid DOA estimation are derived. The simulation combined with the data of Cascade MIMO radar demonstrates the effectiveness of the proposed method relative to other algorithms.