Research On Reception And Processing Technologies Of Satellite-based ADS-B Signals

As a crucial component of the next-generation air transportation system and free flight,automatic dependent surveillance-broadcast(ADS-B)has been of wide application.ADS-B is an emerging technology with functions of communication,navigation,and surveillance,which significantly improves the ability and efficiency of air traffic management.However,in some extreme areas,e.g.oceans and polar regions,the air traffic control is troublesome for the lack of ground stations.To this end,the satellite-based ADS-B application has attracted a great deal of interest as a promising solution.The reception of ADS-B signals from a satellite makes it possible to realize global air traffic monitoring and inevitably faces two obstacles as well.Firstly,within the satellite coverage,ADS-B signals from different aircraft may reach the satellite receiver at about the same time,leading to a mixture of messages.Secondly,the signal-to-noise ratio(SNR)of ADS-B signals received by the satellite receiver is relatively low due to the long transmission distance.It is thus of great theoretical significance and practical value to study the reception and processing technologies of satellite-based ADS-B signals.This dissertation concentrates on the reception and processing technologies of satellitebased ADS-B signals.The main contributions are summarized as follows:(1)An improved constant false alarm rate(CFAR)time support estimation method is proposed for overlapping ADS-B signals,which is suitable for cases with low SNR.Based on eigenvalue decomposition,three estimation methods of noise variance are presented,which are applicable for constant and slowly-varying noise variance respectively.The proposed algorithm solves the problem of threshold determination by realizing the CFAR processing.According to the characteristic of ADS-B signals,we design a matched filter for received signals and propose two time support estimation schemes of direct form and indirect form.Compared with the existing algorithm,the proposed algorithm improves the estimation accuracy and detection probability of arrival times of superimposed signals dramatically at low SNRs,which provides the possibility for the application of other source separation algorithms.Simulation results show that the proposed time support estimation method provides a strong support for the excellent separation performance of overlapped signals at low SNR.(2)A multi-criteria CFAR preamble detection algorithm is proposed for ADS-B signals.The preamble detection is the prerequisite for subsequent demodulation and decoding and is of great importance.The proposed algorithm consists of CFAR detection,deterministic symbol match,consistent power test,and null symbol validation.The CFAR detection exploits the fixed preamble and downlink format(DF)field of ADS-B signals and correlates the received signal with the reference sequence.Based on the correlation property,the ratio of peak value to corresponding valley values is taken as the test statistic and compared with the predetermined threshold to decide the presence or absence of ADS-B signals.The other three criteria are aimed to further confirm that the received signal is ADS-B message and suppress the false alarm as well.A detailed theoretical analysis of the proposed algorithm is described and the formula of detection probability is derived,as well as the theoretical basis for threshold determination.Simulation results show that the proposed method outperforms the existing algorithm in terms of detection performance and is resistant to frequency offset to some extent.(3)With the noncoherent demodulation scheme,N-confidence labelling method and descending confidence labelling method are proposed.The correct confidence labelling is the key to the success of error correction.We point out that there is a trade-off between the performance of error correction and complexity.In view of error correction,the labelling threshold should be as large as possible.The larger the threshold,the lower the probability that erroneous bits are declared high confidence.For complexity,the labelling threshold should be as small as possible because the number of low confidence bits in an ADS-B message decreases as the threshold declines.Through theoretical analysis,we derive the optimal threshold and the corresponding maximum probability of correct reception(PCR)for the N-confidence labelling method.Besides,a simplified method termed descending confidence labelling method is proposed,with a satisfactory performance and relatively lower complexity.(4)With the coherent demodulation scheme,an optimal-threshold confidence labelling method is proposed for known and unknown amplitudes.In the case of unknown amplitude,we estimate the pulse amplitude via the preamble and DF field and derive the optimal threshold for confidence labelling through theoretical analysis.Therefore,the maximum performance improvement can be achieved.Moreover,a dynamic-threshold confidence labelling method is proposed,in which a specific threshold is unnecessary.We also present the theoretical formula of PCR for the dynamic-threshold labelling method.Simulation results show that the proposed algorithm still has a high PCR at low SNRs and has a performance gain of at least 2d B over the noncoherent demodulation scheme with the optimal threshold.(5)A prototype of high-sensitivity noncoherent receiver is designed and built for satellitebased ADS-B application.Considering the dynamic range of the receiver and practical implementation,the proposed multi-criteria CFAR preamble detection algorithm is modified and applied in the receiver prototype to detect the ADS-B signals.In combination with matched filter-based decoding and 1-bit error correction,the processing for ADS-B signals can be accomplished.The probabilities of detection and correct reception are analytically formulated for the receiver prototype and validated by Monte Carlo simulations.The availability of algorithms is evidenced by feeding the real ADS-B signals collected near the Nanjing Lukou International Airport.The prototype test in the laboratory shows that the sensitivity of the receiver prototype is approximately-98 d Bm,corresponding to the PCR equal to 0.93,which is consistent with the theoretical value.