Estimation of Direction of Arrival(DOA) Based on the Spatial Kurtosis Spectrum

Low signal-to-noise ratio (SNR) and weak signal direction of arrival (DOA) estimation have important implications in target orientation and signal separation. However, in the case of unknown number of source signals, traditional DOA estimation requires source number estimation in advance. At low SRN and weak signals, the source number estimation accuracy is low, leading to DOA estimation errors. In response to this problem, the signal kurtosis was used as a cost function to construct a Spatial Kurtosis Spectrum under a uniform circular array constraint, and the DOA estimation was achieved by searching spectral peaks. At the same time, the statistical independence of source signals and the independence of spatial distribution were fully utilized during the calculation operation. In the calculation process, the DOA can be directly estimated without the need for source signal number estimation or signal space decomposition. The simulation results showed that searching the peaks of the Spatial Kurtosis Spectrum can estimate the DOA of the unknown source number. In the case of low signal-to-noise ratio and low SNR, it also had better robustness.