Research On Phase Noise Suppression Algorithm For OFDM System

Orthogonal Frequency Division Multiplexing(OFDM)is a widely used modulation technique for high-data-rate communication over wireless links.It is regarded as one of the key technologies of mobile communication systems due to its unique ability to resist multipath interference and eliminate inter-symbol interference.However,OFDM systems are highly sensitive to phase noise,and the presence of phase noise reduces the effective signal-to-noise ratio(SNR)of the receiver,thereby limiting the bit error rate(BER).Thesis studies the effect of phase noise on OFDM systems,and proposes two phase noise suppression algorithms: a variational Bayesian(VB)-based phase noise suppression algorithm and a Turbo iteration-based phase noise suppression algorithm.For the VB-based phase noise suppression algorithm,thesis estimates the phase noise and the data vector jointly under the VB framework.The algorithm is mainly divided into two stages: the first stage uses pilot subcarriers to roughly estimate the phase noise,and then compensates to obtain the data estimator? the second stage uses the statistical characteristics of phase noise to expand it in the feature domain,and uses the segmentation processing method and the approximation of phase noise to reduce the number of quantities to be estimated,and the VB method is used to obtain the posterior mean of the data,and the phase noise and the data vector are processed in an iterative manner to gradually improve the estimation accuracy of the phase noise and the accuracy of the data vector,and finally the posterior mean of the data vector is used as an estimated value for data demodulation.For the phase noise suppression algorithm based on Turbo iteration,the joint estimation of phase noise and data vector is carried out under the framework of Turbo iteration.The processing method of phase noise in this algorithm is the same as that of the VBbased phase noise suppression algorithm,and the data processing is carried out under the Turbo iteration framework,that is,statistical methods are used between two Minimum Mean Squared Error(MMSE)modules,and module A uses the phase noise estimator and the observation variable generate coarse external information.Module B uses the prior information of the data vector to refine the external information of module A,and transmits the obtained external information to module A again.The two modules iterate each other until convergence,the estimated value of the data vector is finally output,and data demodulation is performed on it.In the thesis,the BER performance of the two phase noise suppression algorithms in the OFDM system is simulated many times.The simulation results show that the two schemes can effectively suppress the phase noise and improve the BER performance of the OFDM system.The BER performance of the proposed algorithms is better than the existing algorithms.