Overlapping ADS-B Signals Separation With Single Antenna Based On Matrix Reconstruction

Automatic Dependent Surveillance-Broadcast(ADS-B)is a kind of surveillance technology which is applied to air traffic management,its on board communication equipment sends out the position information of the aircrafts in the same frequency band by random broadcast,so that the ground station can monitor them.Therefore,the ADS-B system can effectively track a variety of flight state parameters of the aircrafts when appiled to the civil aviation.However,because of the ADS-B signals have the characteristics of random broadcast transmission,the signal overlapping will inevitably occur between different ADS-B signals.If there is no effective overlapped signal separation algorithm,the overlapping problem will lead to decoding error and monitoring information loss,which will cause significant degradation of ADS-B system monitoring performance.Comparing with the array antenna receiver,the single antenna ADS-B receiver is more common,lower cost and more easy to maintain.So considering using the single antenna receiver to solve the problem of ADS-B signal overlapping.The results are as follows:An improved Project Algorithm Single Antenna(PASA)algorithm which separates the overlapping ADS-B signals is proposed.The PASA algorithm transforms single antenna model into virtual multi-channel model by matrix reconstruction,and achieves the purpose of separating overlapped signals by using the Projection Algorithm in array signal processing.However,the PASA algorithm needs the relative time delay of the two signals should be greater than 10 ?s,so as to accurately estimate the virtual steering vectors of the two signals.Therefore,a new matrix reconstruction method to reduce the relative time delay requirment between two signals is proposed.That is,when the relative time delay between two signals is greater than 0.5 ?s,the overlapping signals can be separated.Because of the improved PASA algorithm can not process the overlapped signals with the relative time delay less than 0.5 ?s.Therefore,the Minimum Variance Distortless Response(MVDR)algorithm is applied to separate overlapping ADS-B signals received by single antenna,which has no requirement for relative time delay.The single antenna model is transformed into virtual multi-channel model by matrix reconstruction,and then the overlapped signal is separated by MVDR algorithm which is used in array signal processing.However,this method needs to estimate the frequency of the signal accurately.Because of the problem that the frequency error of the Fast Fourier Transform(FFT)estimation is large,the improved frequency estimation method can reduce the frequency estimation error of the signal,so as to effectively separate the overlapped signals.The simulation results show that the improved PASA algorithm can effectively separate overlapped signals,but the relative time delay of two signals is required to be greater than 0.5 ?s.The MVDR algorithm can separate the overlapped signals without the requirement of the relative time delay between two signals under the premise of accurately estimate the frequency of signals.