Novel DOA estimation based on human behavior algorithm

The famous MUSIC algorithm performs well in estimating the direction of arrival (DOA). However, the spectral peak search step in MUSIC is computationally expensive for real-time applications because it uses the grid method, in which the spectral value at each grid point must be computed. Because DOA estimation must find several maximum spectral peaks, it is a typical problem in global optimization. According to the Human Behavior Algorithm, this study proposes a novel DOA estimation approach for searching for MUSIC spectral peaks that possess the property of global convergence. Experiments were conducted to demonstrate that the proposed approach has a higher resolution or lower computational time.