Research On Intelligent Algorithm Of Channel Equalization And Adaptive Decoding Of Fast-than-nyquist Wireless Communication

Because of the limited and scarce wireless spectrum resources,it is becoming more and more important to improve the transmission rate and bandwidth efficiency of wireless communication.In the range of Mazo,FTN transmission system can,without increasing the energy per bit and BER cases,significantly increase the wireless communication transmission rate and bandwidth efficiency in the same bandwidth,compared with the traditional Nyquist system.But it also artificially introduces infinitely long ISI.The paper first briefly describes the principle and application of FTN system,and analyzes the process and trend of the development of FTN technology.Then,the signal characteristics and transmission characteristics are analyzed in the transmission of different waveform and different acceleration factor in FTN wireless communication.The basic principles and the framework of FTN system are introduced,and more details of FTN receiver,are discussed.Because FTN people introduce infinitely long ISI,it is difficult to design the pilot symbol and the effective channel estimation scheme,so the paper abandons the channel estimation.On the one hand,establish an intelligent equalization system model of FTN wireless channel,and use LSSVM intelligent algorithm to form the new scheme of channel equalization with low computational complexity.The scheme does not need to estimate the channel,and it can get the parameters of the equalizer directly.This paper focuses on the reciever of BPSK modulation FTN system.Under the condition of 0d B to 50 d B different signal-to-noise ratios AWGN channels,it compares LSSVM linear kernel equalization algorithm with LSSVM nonlinear RBF kernel equalization algorithm with different training sequence lengths in FTN system.The experimental results show that the equilibrium effect of the FTN LSSVM nonlinear RBF equalization algorithm is excellent.Therefore,the FTN wireless channel intelligent equalization system is balanced by the nonlinear RBF LSSVM kernel function.Furthermore,based on the Bayesian framework criterion and the grid search algorithm optimizing LSSVM algorithm,the effect of channel equalization is studied respectively.The grid search LSSVM equalization algorithm is determined,based on 50 bits as the length of the training sample sequence.On the other hand,the paper studies the adaptive decoding algorithm of TPC in the FTN system.The adaptive channel decoding of FTN wireless communication is studied.Based on Chase algorithm,the complexity of channel decoding operation is reduced and its decoding processing speed is improved.Finally,based on the intelligent algorithm equalization,study on adaptive channel decoding deeplier,and obtain better BER and WER performance.This paper focuses on LSSVM intelligent equalization algorithm of FTN wireless communication receiver and TPC adaptive decoding of Chase algorithm.The effectiveness of FTN wireless communication transmission,such as increasing the transmission rate and bandwidth efficiency of wireless communication,is achieved,and the reliable transmission of FTN wireless communication is also ensured.