Abstract: Automatic Dependent Surveillance-Broadcast（ADS-B）ADS-B is the monitoring technology of aircraft based on the global positioning system.Due to the random access of ADS-B signal and the increasing of airspace traffic, signal overlapping was inevitable.Overlapped signal would seriously affect the decoding of ADS-B signals, resulting in erroneous decoding information or loss. In this paper, a robust and high gain algorithm of overlapped signal processing based on Project Algorithm (PA) was proposed, which used gain from antenna array.Firstly,the steering vector of source signals were estimated by PA.Then,combining the Capon beamforming algorithm,the output SNR of the separated signal is improved.In addition, considering the estimation error of the steering vector, this paper applied the Robust Capon Beamforming (RCB) algorithm to further improve the robustness of the separation algorithm of ADS-B.Through simulation and test experiments, it is verified that the proposed algorithm can improve the output SNR of the signal more robust..