Off-Grid DOA Estimation Via Real-Valued Sparse Bayesian Method in Compressed Sensing

摘要

A novel real-valued sparse Bayesian method for the off-grid direction-of-arrival (DOA) estimation is proposed in compressed sensing (CS). The off-grid model is reformulated by the second-order Taylor expansion to reduce modeling error caused by mismatch. To apply the Bayesian perspective in CS conveniently, complex data are addressed to yield a real-valued problem by utilizing a unitary transformation. By assuming that sources among snapshots are independent and share the same sparse prior, joint sparsity is exploited for DOA estimation. Specifically, a full posterior density function can be provided in the Bayesian framework. The convergence rate and convergence stability of the proposed method can be guaranteed in the iterative procedure. Simulation results show superior performance of the proposed method as compared with existing methods.