Abstract: In recent years, Automatic Dependent Surveillance-Broadcast has received extensive attention due to its excellent surveillance performance, and ICAO has made it the new generation of air traffic surveillance technology. Aiming at the overlapping problem of ADS-B signal, a method of overlapped signal processing under single antenna is proposed. The ADS-B overlapped signals received by the single antenna is adaptively decomposed by Empirical Mode Decomposition(EMD), and the number of sources is estimated based on the signal characteristics. On this basis, the phase space reconstruction method is used to eliminate the modal mixing phenomenon in the decomposition. In addition, the similarity coefficients between the decomposed components and the overlapped signals are calculated to form a new multidimensional signal, and then the ADS-B signal is processed by Independent Component Analysis(ICA). In this paper, the Empirical Mode Decomposition is applied to process overlapped signals, which realizes the self-detection of the number of overlapped signal sources, and the method is not sensitive to the relative delay of the overlapped signals. The results of simulation and data acquisition show that the algorithm can effectively solve the problem of source number estimation and separation of ADS-B overlapped signals.