Direction-finding Algorithm for Noncircular Signals through Direction Domain Filtering

In some practical applications, the directions of arrival (DOAs) of impinging noncircular signals are in a roughly known range. A new algorithm based on direction domain filtering is proposed to improve the resolution and accuracy of the traditional MUSIC (multiple signal classification) algorithm for noncircular signals (NC-MUSIC). Firstly, get the spatial spectrum of NC-MUSIC algorithm and filter it by the roughly known range. Secondly, use the filtered spatial spectrum to reconstruct the extended covariance. Thirdly, find DOAs of impinging noncircular signals followed the steps of traditional NC-MUSIC algorithm based on the reconstructed extended covariance. Simulation results show that the proposed algorithm holds the ability of array extension that the traditional NC-MUSIC algorithm already has, improves the resolution and the accuracy which are inversely proportional to the magnitude of the roughly known range. The proposed algorithm is of practical value.