Source Number and DOA Estimation Method Based on Eigen-Beam MUSIC for Closely Spaced Signals

Li Liu,Xiaopeng Yang,Sheng Gao,Shuai Li
DOI: https://doi.org/10.1109/icsidp47821.2019.9173374
2019-01-01
Abstract:The estimation accuracy would degrade severely when two signal sources with different power are located closely by conventional DOA methods. In order to solve this problem, a estimation method based on Eigen-Beam MUSIC is proposed to obtain the DOA and source number jointly. Firstly, the source spectrum is constructed, and then the source number can be estimated by comparing the peak value of each peak and the noise power. Secondly, the DOA spatial spectrum is constructed by combining MUSIC spatial spectrum and eigen-beam spatial spectrum, which can enhance the spectral peak of weak sources in the signal subspace, and then estimate the DOA by searching the angle corresponding to the peak. Compared with the existing method, the proposed method has the simper computation and better estimation performance. Some results of numerical experiments are provided to demonstrate the effectiveness of the proposed method.
What problem does this paper attempt to address?