Study On Blind Estimation Of TDDM-BOC And CBOC Signals

Along with the rapid development of global economy and technology,satellite navigation system has been importantly advancing across the world and largely used in the area of science and technology as well as the civilian and military field.Meanwhile,as navigation systems and their number of satellites and navigation services continue to increase,the navigation frequency is getting gradually over-crowded.To reduce the interference between the signals in the same frequency band and the signal compatibility problem,Binary Offset Carrier(BOC)modulation signal is proposed and a variety of BOC signals are derived.To improve the tracking accuracy and anti-multipath ability,a Composite Binary Offset Carrier(CBOC)modulation signal is derived from the BOC modulated signal.To increase the information confidentiality and the tracking and capturing precision of the signal,the Time Division Data Modulation Binary Offset Carrier(TDDM-BOC)signal is proposed.At present,the Global Position System(GPS)from the US,the Galileo satellite navigation system from the EU,and the "Beidou" navigation system from China have already selected the two new BOC derivative signals of CBOC and TDDM-BOC.Therefore,information management and control are particularly important,so the importance of the blind estimation research on TDDM-BOC and CBOC signal sequences is becoming increasingly prominent.In this paper,the modulation principle and characteristics of CBOC and TDDMBOC signals are studied first,and then the blind estimation algorithm of the two signals is further investigated and discussed.The main work is as follows:(1)Aiming at the issue of blindly estimating CBOC modulation signal sequences,an principal component analysing neural network algorithm based on Hebbian learning rules is studied.Firstly,the algorithm uses the singular value decomposition algorithm which is segmented by the two-fold information symbal period to verify the feasibility of the blind estimation of CBOC modulated signal sequences.Then,to solve the disadvantage of singular value decomposition algorithm that highly requires memory and complexity,an principal component analysing neural network algorithm based on Hebbian learning rules is introduced,and the blind estimation of the,combined code sequence of CBOC signal is finally realised.(2)For the issue that the singular value decomposition algorithm cannot handle long data vector and needs high complexity during the blind estimation of combined code sequence of TDDM-BOC signal,a blind estimation method based on Sanger Neural Network(Sanger NN)is studied.Let one-cycle TDDM-BOC signal after segmentation to be input signal in algorithm.Let the sign function of weight vector in algorithm stand for the combined code sequence of TDDM-BOC signal.Input signal continuously to achive the target of repetitively training weight vector until algorithm converge completely.Finally use the sign function of weight vector to reconstruct the combined code sequence of TDDM-BOC signal.Besides,this paper introduces variable step length convergence model based on Recursive Least Squares(RLS),which improves the convergence rate of neural network to a great extent.(3)Regarding to the low convergence rate of Sanger NN in the blind estimation of the combined code sequence of TDDM-BOC signal,a blind estimation method of the combined code sequence of TDDM-BOC signal based on LEAP NN(on line unsupervised LEArning neural network for adaptive feature extraction via Principle component analysis).LEAP NN is an improved algorithm based on Sanger NN.It increases convergence rate by using segmented signal as input signal,using the weight vector which have been trained to reconstruct the combination code sequence of signal,and improving weight updating formula.