DOA Estimation Using Sparse Representation of Beamspace and Element-Space Covariance Differencing
Fujia Xu,Aifei Liu,Shengguo Shi,Song Li,Ying Li
DOI: https://doi.org/10.1007/s00034-021-01846-y
2021-09-18
Abstract:In order to eliminate the effect of noise on the performance of the direction-of-arrival (DOA) estimation and reduce the computational complexity, a sparse representation (SR) DOA estimation method is proposed. The proposed method first utilizes the beamspace and element-space covariance differencing to eliminate noise. Afterward, it vectorizes the difference covariance matrix. In a sequence, it establishes a new SR model to complete DOA estimation. Compared to existing SR DOA estimation methods, the proposed method significantly reduces the computational complexity since the parameters to be solved in its SR cost function are regardless of the number of sources and the number of array elements. Simulation results show that in the case of the unknown number of sources and low signal-to-noise ratios (SNRs), the proposed method has high DOA resolution and estimation accuracy.
engineering, electrical & electronic