The Estimation Of NC-OFDM And FBMC-OQAM Signal Parameters

With the development of the age,the demand for data transmission rate is increasing day by day,and the spectrum resources are increasingly scarce.Non-Continuous Orthogonal Frequency Division Multiplexing(NC-OFDM)is used as the OFDM signal of cognitive radio.It can make full use of existing spectrum resources and its subcarriers are orthogonal to each other with a good spectrum utilization.However,the problem of the side lobe leakage of NC-OFDM is serious.Filter Bank Multi-Carrier(FBMC)which is one of the candidates for 5G transmission technology,can improve the serious problem of sidelobe leakage of NC-OFDM,and the spectrum efficiency is better.With the advancement of NC-OFDM and FBMC techniques,the study of signal parameter estimation and modulation recognition on subcarriers has certain research significance.The research objects of this thesis are NC-OFDM and FBMC signals.Combined with the existing signal parameter estimation algorithm,the parameter estimation of signals and modulation identification on subcarriers are carried out by analyzing the characteristics of NC-OFDM and FBMC signal parameters.The main work of the thesis is as follows:(1)Aiming at the important problem of parameter estimation of NC-OFDM signals,the algorithms of cyclic autocorrelation and fourth-order cyclic cumulant are studied.Firstly,according to the cyclic autocorrelation of the NC-OFDM signal,the parameters are estimated with the characteristics of the discrete spectral lines in the cyclic frequency?(28)0 and the time delay?(28)T_u cutaway.Then the subcarrier frequency interval is estimated according to the fourth-order cyclic cumulant characteristics of NC-OFDM.Finally,the cyclic autocorrelation of the NC-OFDM signal and the fourth-order cyclic cumulant are numerically simulated.The experimental results show that when SNR is low,both algorithms can effectively estimate multiple parameters of the NC-OFDM signal.(2)With the regard to the blind estimation of FBMC-OQAM signal of symbol period,autocorrelation second-order moments and cyclic autocorrelation algorithms are studied.Firstly,the symbol period of FBMC-OQAM signal is estimated by the discrete property of the value of the autocorrelation second moment of FBMC-OQAM signal.Secondly,according to the cyclic autocorrelation of the FBMC-OQAM signal,the symbol period estimation is estimated on the cyclic frequency?(28)0 slice with discrete spectral features.Then,the symbol periods estimated by the two algorithms is compared and analyzed.Finally,the performance of the symbol period estimation algorithm of FBMC-OQAM signals is demonstrated.The experimental results suggest that both algorithms can effectively estimate the symbol period of FBMC-OQAM signals under low SNR,and the feasibility of the proposed method is verified.(3)Considered the subcarrier modulation identification of FBMC signals,this thesis studied a combination of spectral correlation and singular value decomposition algorithms.Firstly,the modulation and recognition of the FBMC signal is performed according to the spectral line difference of the cyclic autocorrelation?(28)0 slice of the QAM-FBMC and OQAM-FBMC signals.Then,a signal matrix is composed by MOQAM(MQAM)-FBMC,and the singular value decomposition algorithm is used to decompose the eigenvalue of the signal matrix.Finally,the corresponding numerical simulation is carried out.The experimental results indicate that the proposed algorithm can effectively identify the subcarrier modulation method in FBMC signal with low SNR,which has certain theoretical research value.