Overlapping ADS-B signals separation algorithm based on MUSIC

Abstract

In order to improve the reception performance of SSR replies in high-density signal environment, a separation algorithm is proposed, which constructs the separating matrix with estimating the direction of arrival of signal. Firstly, determine the number of overlapping signals with the eigenvalues distribution of the measurements. Secondly, reconstruct the mixing matrix with the DOA of signals, which is estimated by peak value searching in MUSIC algorithm. Finally, estimate the separating matrix by calculating the Moore-Penrose inverse of the reconstructed mixing matrix to separate the overlapping signals. Simulation is done based on uniform linear array with 6 elements. The results show that the proposed separation algorithm can achieve more than 90 percent success rate to separate two short Mode S replies, and the separating performance is similar to independent component analysis algorithm better than projection algorithm. The amount of calculation is less than 10 percent of ICA algorithm, so the proposed separation algorithm is easier to engineering application.