Estimation of 2D-DOA for Multiple Frequency Hopping Signals Based on Blind Source Separation

In order to realize two-dimensional direction of arrival (DOA) estimation of multiple frequency hopping signals, a two-dimensional DOA estimation algorithm based on underdetermined blind source separation is proposed. Firstly, multiple frequency hopping signals are received through L-shaped array antenna, and STFT&WVD combined time-frequency method is used for time-frequency transformation operation to improve time-frequency resolution and cross term suppression ability of time-frequency map. Then, single time-frequency source points are detected by using azimuth angle difference of each source's frequency hopping signals. Finally, improved K-means clustering algorithm is used to complete time-frequency single source point and frequency hopping rate clustering of each source's frequency hopping signals, The mixed matrix is calculated according to the hopping frequency and the corresponding time-frequency single source point, so as to realize the azimuth and pitch angle estimation. The simulation results show that under the condition of low signal-to-noise ratio, the algorithm has high precision of two-dimensional DOA estimation, good robustness and certain anti noise ability.