Reseach On Time Difference Of Arrival Estimation And Direct Position Determination For Frequency-hopping Signals

Due to its outstanding performance for resistance of narrowband interference and anti-interception,Frequency-Hopping communication has become an important part of modern wireless communication,and found wide applications in Globe System for Mobile(GSM),bluetooth,drone communications and military wireless communications.The wide applications of Frequency-Hopping(FH)communication also brings a big challenge to the reconnaissance and monitoring of FH emitters.In accordance with this problem,this dissertation presents research on time difference of arrival(TDOA)estimation and direct geolocation for FH signals.Compared to the case of fixed frequency signals,the key point of the research on TDOA estimation and direct geolocation for FH signals is how to jointly utilize the multiple hops of FH signals to improve the estimation and geolocation performance.The main work and contibutions of this dissertation are as follows.1.The maximum likelihood estimator and the Cramer-Rao lower bound(CRB)of TDOA estimation for FH signals are derived under different channel models.The closed-form CRB expression distinctly shows how the parameters of FH signals affect the accuracy of TDOA estimation.The relations between the CRBs of different models and methods are also disclosed,which demonstrated that the performance of maximum likelihood method is much better than that of dual phase differential method.Then,the cause of periodical peaks of cross correlation for FH signals are analysed theoretically,and corresponding solutions are proposed.To solve the problem of that the data amount of wideband and slow hopping FH signals is huge,we present a fast method for such FH signals TDOA estimation.2.The direct TDOA geolocation methods of FH signals have been addressed.First,the direct TDOA geolocation model for FH signals in flat fading channels is established.Based of this,the maximum likelihood direct TDOA geolocation estimator and the least squared direct TDOA geolocation estimator for a single FH signal are derived,and the CRB is also analysed.Then,we analyse the resolution for a single FH signal direct TDOA geolocation.Further,an extension of the proposed method for multiple FH emitters direct TDOA geolocation has been presented based on the resolution analysis.For the cases of multiple FH signals have equal or unequal amplitudes,solutions are given respectively.3.The direct direction of arrival(DOA)geolocation algorithms for narrowband signals have been studied.First,the direct DOA geolocation model and algorithms of narrowband signals under synchronous observation are introduced,including the case of a single emitter and the case of multiple emitters,and the multiple signal classification(MUSIC)based direct position determination(DPD)method is proposed.Based on this,the direct DOA geolocation model and algorithms of narrowband signals under asynchronous observation are studied,and the conventional beamforming(CBF)based DPD method,minimum variance method(MVM)based DPD method and the MUSIC based DPD method are proposed.To resolve the problem of coherent signals,the forward-backward linear predictor(FBLP)based DPD method is proposed.4.The direct DOA geolocation algorithms for FH signals have been addressed.Based on the research of narrowband signals,combining the idea of incoherent signal subspace method(ISSM)and the characters of FH signals,the CBF based DPD method,the MVM based DPD method and the MUSIC based DPD method for FH signals are proposed both for the case of synchronous observation and the case of asynchronous observation.All these methods jointly utilize the multiple hops of FH signals to gain the total cost function.5.Outdoor experiments have been carried out for the direct TDOA geolocation method.Based on the equipments at hand,a three-station TDOA locating system has been built up for outdoor experiments.In a real environment,direct TDOA geolocation and two-step TDOA location tests have been carried out for the binary phase shift keying(BPSK)signal from E4438 C,the video signal from drone and the FH control signal from drone controller.The results demonstrated that the performances of the direct TDOA geolocation method are better than the two-step method.