Study On Blind Estimation Of Parameters For BOC And CBOC Signals

Since the 21 st century,with the continuous improvement and development of the global satellite navigation system,the global satellite navigation systems are not only widely used in aerospace research,but also provide convenience in serving human production and life.In order to solve the problem of severe spectrum congestion,a binary offset carrier(BOC)modulated signal is introduced,and the signal has better peak-to-peak characteristics and more accurate positioning accuracy.Based on the BOC signal,a series of BOC derived signals are generated,for example,composite binary offset carrier(CBOC)modulation signal,etc.With the large number of these BOC signals used by navigation systems in various countries,it is meaningful to study the blind estimation of parameters for BOC and CBOC signals.In this thesis,the model and characteristics of BOC and CBOC signals are given firstly.Then,the in-depth research and analysis of the parameters blind estimation algorithm of the two signals are carried out.The main research contents are as follows:(1)Aiming at the problem of blind estimation of pseudo code rate,subcarrier rate and carrier frequency rate of CBOC signals,the cyclic spectrum algorithm is studied.Firstly,according to the characteristics of the CBOC signal and the cyclostationary characteristics of the signal,the cyclic autocorrelation expression of the signal is derived.Then the Fourier transform of the expression can be used to obtain the superposition of two BOC cyclic spectra.Finally,it can be blindly estimate the pseudo code rate,subcarrier rate and carrier frequency according to the peak characteristics of the cyclic frequency cross section of the superimposed CBOC signal.(2)Aiming at the traditional singular value decomposition(SVD)algorithm to estimate the BOC combination code sequence and the information code sequence have unitary fuzzy problem,an improved SVD algorithm is studied.Firstly,the linear transformation of the largest singular eigenvector and the eigenvector of the second largest singular value obtained by the SVD algorithm is performed.The constructed coefficient matrix is multiplied by two large eigenvectors obtained by the traditional SVD algorithm.Then the optimal coefficient matrix is optimized according to the characteristics of the coefficient matrix.Finally,the coefficient matrix is substituted into the eigenvector transformation equation,and the combination code sequence and information code sequence can be estimated at a lower signal-to-noise ratio,thus avoiding unitary fuzzy.(3)Aiming at the blind estimation of CBOC signal pseudo code period and combination code sequence,the secondary power spectrum algorithm and radial basis function(RBF)neural network algorithm are proposed respectively.First,the secondary power spectrum of the signal is derived,and the pseudo code period can be obtained according to the spectral peak interval of the secondary power spectrum.In the case of the estimated pseudo-code period and allowing a certain bit error rate,the combination code sequence of one cycle is overlapped and segmented firstly,and the learning coefficient is optimized and filtered in the RBF neural network.Then,each sequence is used as the RBF neural network with the input vector is supervised and adjusted.Finally,the original combination code sequence can be restored by inputting each of the intercepted data vectors multiple times and continuously training the weight vector.