Research On The Theory And Key Technology Of Real-time Detection Of AIS Autonomous Positioning Information

As the application of GNSS has gradually penetrated into all areas of the society,its inherent vulnerabilities have also become prominent gradually,which has attracted the attention of the international community.In the field of maritime,in order to ensure the navigation safety of coastal vessels,the IMO has proposed that all vessels should be equipped with both space-based and land-based dual backup positioning and navigation systems.The AIS is a real-time interactive navigation assistance system.The SOLAS stipulates that all vessels with a gross tonnage of 300 tons or more must install the AIS compulsorily.The positioning information required to realize the automatic identification function of vessels comes from its built-in GNSS receiver.Once the GNSS is suddenly and continuously interrupted,the AIS will be paralyzed,which will seriously,threaten the navigation safety of vessels.Therefore,the AIS ranging mode has been explicitly listed as one of the backup land-based radio navigation systems in the World Wide Radio Navigation Plan.At present,both GNSS and other land-based positioning systems are using continuously periodic carrier signals of the same frequency in the positioning process.However,the AIS is essentially a communication system,and it uses the SOTDMA method,that is,the AIS signal is non-periodic.Moreover,the carriers of AIS adopt the GMSK modulation which are completely different from the biphase modulated carriers of the traditional positioning systems.This means that the traditional carrier measurement techniques cannot be applied to extract positioning information from AIS signals.Therefore,in order to use the AIS as a positioning system,the first technical problem that needs to be solved is how to obtain positioning information from the AIS signals in the time of one slot,that is,the positioning information measurement of non-periodic AIS signals.The solution proposed in this dissertation is to apply the theory of signal sparse representation to the detection AIS real-time signals.Based on the in-depth study of AIS carriers,dictionary training algorithms,sparse decomposition algorithms and transmission losses of AIS signals,a novel sparse representation algorithm that suitable for real-time signal processing is proposed.It uses a small number of atoms to represent the AIS signal to obtain the holographic data of AIS signal,and reduces the error caused by noise.Then,by analyzing and co-sensing processing of the carrier characteristics,time slot,and baseband waveforms of AIS signal,a timestamp detection algorithm of AIS signal is further proposed to detect the positioning information of AIS signal and finally realize the measurement of positioning information.The main contents of this dissertation are as follows:Firstly,in order to obtain the holographic data of AIS signals,the theory of signal sparse representation is applied to AIS,real-time signals.Aiming at the structure and properties of AIS signal,the dictionary training algorithms and the sparse decomposition algorithms are studied.Based on these,a sparse representation algorithm of AIS signals based on a completed dictionary is proposed,and the holographic data of AIS signals is obtained.Secondly,in order to solve the problem of real-time detection of holographic data of AIS signals in the time of one slot due to the non-periodic characteristics of AIS signals,a fast denoise sparse representation algorithm of AIS signal that based on the segmented representation and the pre-calculation is proposed by the mathmetical transformation of the objective function of the sparse representation and the optimization of the realization process of sparse decomposition.Compared with the sparse representation algorithm of AIS signal proposed in Chapter 1,the sparse representation accuracy of this algorithm is improved by about 60%,the bit error rate is reduced by about 70%,and the calculation speed is increased by at least 30%.Thirdly,in order to solve the problem of the increase of computational complexity due to the signal segment processing and further enhance the real-time performance of the algorithm,the transmission loss of AIS signal over sea surface is studied.The transmission loss of AIS signal under different conditions such as propagation distance,seawater temperature,seawater salinity,and wind speed over sea surface are estimated and analyzed.On this basis,an adaptive real-time detection method of AIS signals based on the transmission loss is proposed,and the corresponding detection module of AIS signal based on the FPGA is developed.Finally,in order to solve the problem of the positioning information measurement of non-periodic AIS signal,a method for measuring the propagation delay of AIS signal based on the in-depth research and analysis of the baseband waveform of AIS signal.This method can detect the timestamp information that required for positioning using AIS signals,and then perform the measurement of propagation delay.The established fieldtest platform verifies the effectiveness of the method.In summary,the research on the theory and key technology of real-time detection of AIS positioning information proposed in this dissertation has important theoretical significance and practical value for the realization of AIS autonomous positioning,and has made a certain contribution to the research of this field.