The time-domain multiple signal classification DOA estimation method based on app-phase preprocessing

For the instability problem of noise subspace estimated result of frequency domain multiple signal classification(MUSIC) under low signal-to-noise ratio, a time-domain multiple signal classification DOA estimation method based on all phase preprocessing was proposed. ①the linear array receiving data was grouped; ②the data of each group was preprocessed by phase shift according to the search angle, and the preprocessing results of each group of data were added to get a new set of data; ③the phase-shifted covariance matrix was constructed in the time domain, the noise subspace was stably estimated under the shorter data, the corresponding spatial spectrum estimation was obtained by the unit matrix adder according to the orthogonal characteristics of noise subspace, and the direction of arrival estimation value was obtained.The results of numerical simulation and measured data processing show that, compared with the frequency domain MUSIC method, this method effectively improves the signal content and signal-to-noise ratio of the covariance matrix, and realizes the stable estimation of the noise subspace under the case of shorter data.It has better stability and detection performance, and improves the robustness of the MUSIC method. © 2020, Editorial Office of Journal of Vibration and Shock. All right reserved.