Sparsity-Inducing Direction Finding for Nested Array: A Bayesian Inference Treatment

This paper aims at formulating a hierarchical Bayesian framework and variational inference perspective for direction-of-arrival (DOA) estimation using nested array. An overcomplete basis model of the array covariance vectors is constructed first to exploit the spatial sparsity of the incident signals. Then we utilize a three-stage hierarchical prior to model the sparsity of the unknown signal, and derive the estimation procedure with this prior using variational inference for fully Bayesian estimation. Finally, we conduct numerical simulations to demonstrate the performance of the proposed method in DOA estimation precision as compared with other state-of-the-art algorithms.