A Nested Array Wideband Signal Angle Estimation Algorithm Based on Sparse Bayesian Learning

To estimate the direction of arrival (DOA) of wideband signals from nested arrays, a Wideband Smooth Reconstruction Block Sparse Bayesian Learning (WSR-BSBL) algorithm is proposed. Firstly, a segmented broadband signal frequency domain single measurement vector DOA estimation model is established. The single-measurement vector DOA estimation model is transformed into a multi-measurement vector model by smooth reconstruction, which reduces the dimension of the compressed perceptual dictionary matrix. The multi-measurement vector model is then solved by utilizing the block sparse Bayesian learning algorithm to estimate the broadband signal’s baud direction. This reduces the solution complexity of the sparse Bayesian learning algorithm. According to the experimental results, the technique significantly improves estimation performance over the state-of-the-art approaches and effectively addresses the issue of the high complexity of the DOA estimation algorithm for nested array broadband signals.