| “Automatic Identification System(AIS)” has already attracted much attention from many countries and organizations and becomes one of the most important & hot development directions in maritime monitoring fields,due to its immense value in civilian and potential value in military.With the vigorous development and increasing maturity of space technology,especially the micro/small satellite technology,using the micro/small satellite-based AIS for wide-area,all time and all weather global maritime surveillance has been evolved into possibility and it has the remarkable advantage such as low cost,quick test,fast assemblage,rapid launch,rapid in use and easy for maintenance and upgrade.However,due to its large coverage,long-range transmission and high speed in orbit compared with the traditional ground AIS system,new challenges emerge such as multi-cell signal collisions,Doppler frequency shift,long transmission delay and low Signal Noise Ratio(SNR),which severely limite the outstanding performance of the space borne AIS.Moreover,when using a single AIS satellite to realize large area maritime surveillance,the real-time performance is poor and is hard to put into practice.Aiming at addressing these above problems,this paper has conducted a profound research on key technologies of enhancing performance of the system based on modeling and anyalysis of space borne AIS system;and design and optimization on AIS satellite constellation for regional coverage have been carried out at the background of maritime surveillance on our nation's major sea waters.The main research activities of the thesis are:Firstly,the system characteristic is studied and the observation model and detection probability model are established for satellite-based AIS,respectively.The Doppler frequency shift,path delay,multi-cell signal collisions and link budget are analysed,among which the multi-cell signal collisions are discussed in detail and two types of signal collisions are classified.By introducing a non-uniform distribution factor,the detection probability model based on ship density function is established,and the detection probability of ships in uniform distribution and non-uniform distribution are compared with each other,which validates this model.After that,the effects of the two-dimension ship distributions,reporting interval,orbit height,antenna swath and Class B ships on the detection probability are studied through simulations to direct the development of key technologies and enhancement of detection ability.Secondly,three key technologies are carried out from the aspects of AIS receiving antenna,AIS receiver and blind source separation for the satellite-based AIS in detail as follows.(1)The design of AIS receiving antenna and beamforming techniques are investgated.The design requirement of receiving antenna was analysed in advance for the global ship detection in 2018.To overcome the contradictions between large coverage and low detection probability,the “Tian Tuo 1” AIS system is taken as an example to analyse the effect of receiving antenna and its beam on the detection ability.Considering the platform limitations of micro/small satellite AIS,a beam scanning method based on the directional antenna is proposed.By applying this method,a narrow swath can be obtained to achieve high detection probability instantaneously.Meanwhile,a large coverage can be obtained to keep the wide coverage of satellite-based AIS through beam scanning.Based on the above ideas,one beam scanning method based on the designed helical antenna and the other beam scanning method based on the designed ESPAR antenna are discussed in detail.Then,the beam scanning strategy,scanning range,scanning rate and optimal swath are studied,respectively.The results show that the proposed method can obviously improve the detection probability and is effective for the global ship detection.(2)The technology of satellite-based AIS receiver are studied.To address the poor performance of bit error resulting from Doppler frequency shift and co-channel signals interference,a new kind of AIS receiver architecture based on multiple channels and parallel processing is proposed.The structure and software processes of the revised AIS receiver are introduced and the demodulation algorithm is studied in detail.Besides the primary channels,two candidate channels operating in the range of(159~163)MHz are deployed in the new AIS receiver,which can not only be taken as the backup of the primary channels but also as the special channel for long-range AIS transmissions.By parallel processing in the multiple sub-channels,the influence of Doppler frequency shift is relaxed and the demodulation performance is also improved.Thanks to the combined N bit differential demodulation scheme,the revised AIS receiver gets strong ability in counteracting the Doppler frequency shift and co-channel signals interference.Therefore,great improvement is achieved for the demodulation performance of the AIS receiver,further enhancing the detection ability of the satellite-based AIS.(3)The Blind Source Separation(BSS)method for separating the overlapped AIS signals is theoretically studied.For the inevitable AIS signal collisions,a separation algorithm based on the differences of signal amplitude and signal reconstruction & cancellation is proposed to separate two overlapped AIS signal.At first,the feasibility of this separation algorithm for the improvement on detection probability is verified.Then,the impact of parameter estimation error on the performance of signal reconstruction & cancellation is theoretically deduced.More emphases are put on the parameter estimation of the overlapped AIS signal and a two-time estimation method is proposed to improve the estimation accuracy.Moreover,the real-time parameter tracking and performance limits of parameter estimation are analysed.At last,the demodulation performance of the weak signal is computed through simulations which validated the effectiveness of the proposed separation algorithm.Finally,in order to further enhance the detection ability and improve the real-time performance of the satellite-based AIS,an AIS satellite constellation for regional coverage is designed and optimized based on the genetic algorithm.The coverage characteristic of the AIS satellite is analysed,and the grid distributions and coverage model of the target region are established by using point coverage analysis method.Taking into account of the characteristic of the satellite-based AIS,the orbital elements and its influence on the performance of satellite coverage are designed and analysed through numerical simulation.According to the mission requirement,a constellation optimization model is established.It is a multi-objective optimization problems and the solution is to seek the minimum number of satellites and maximum the average coverage time.For the above objectives,a hierarchy optimization strategy based on the adaptive GA is proposed and an optimal constellation is worked out by combining Matlab and STK tool. |
PDF Full Text Request