Two-dimensional DOA estimation algorithm based on time-frequency point selection

The key problem of estimating the direction of arrival (DOA) of non-stationary signals based on spatial time-frequency distribution (STFD) matrix is to select appropriate time-frequency points. The purpose of this paper is to improve the estimation accuracy of DOA by selecting the time and frequency points of source signals. Firstly, we analyze the principle of time-frequency distribution matrix optimization and the principle of selecting time-frequency points of signal terms. Then, the STFD matrix trace is used to construct the decision quantity to select the time-frequency points of the signal items. Moreover, we establish the STFD matrices at all selected time-frequency points. Finally, in view of the high complexity of the traditional time-frequency multiple signal classification (TF-MUSIC) algorithm, this paper uses the TF-MUSIC algorithm of dimensionality reduction to realize DOA estimation. The effectiveness of the proposed algorithm is verified by theoretical analysis and simulation experiments.