Research On Key Techniques Of Universal Filtered Multi-carrier

The 5~thh generation of mobile communication(5G)will satisfy the demand of communication in the next few years.However,Orthogonal Frequency Division Multiplexing(OFDM)has the disadvantages of high out-of-band(OoB)leakage and sensitivity to carrier frequency offset(CFO),it can't no longer satisfy the requirements of 5G networks.Therefore,a new type of air-interface waveform is needed to adapt new requirements.Universal Filtered Multi-carrier(UFMC)is one of the candidate air-interface waveforms for 5G.Compared with OFDM,UFMC technology adopt flexible sub-band filtering which significantly reduced OoB leakage and improve spectrum utilization.In addition,UFMC is better applied to short burst communication because of its shorter filter order and has more flexibility with the parameter configuration.This article mainly studied OoB suppression and multi-frequency offset estimation of UFMC.Firstly,this paper described the system model of UFMC and introduced three kinds of receivers,zero forcing(ZF)detection?matched filter(MF)detection and 2N-point FFT receiver.Then analysis the three receivers'BER performance,the 2N-point FFT receiver has lower complexity and better BER performance and introduced CP-UFMC.Furthermore,compared the two waveforms with OFDM.Simulation results showed that the UFMC has higher tolerance to carrier frequency offset(CFO),lower OoB leakage and higher time-frequency efficiency.Secondly,this paper studied OoB suppression of UFMC in depth.The frequency domain characteristic of the UFMC subband filter determined the OoB leakage.In order to suppress OoB leakage and improve the spectral efficiency,it is necessary to reduce the relative amplitude of the side lobe of the sub-band filter.In this paper,we compared the performance of UFMC with different window function filters and proposed a convex optimization method to optimize the prototype filter of UFMC.Numerical simulations showed that,compared with the Dolph-Chebyshev filter in conventional UFMC system,the proposed filter has superiority in terms of OoB emission and BER performance in high mobility scenarios.At last,the multi-frequency offset estimation algorithm in cooperative multi-point transmission(CoMP)scenario is studied.Firstly,we analysis the influence of CFO on UFMC system and derived the system model of the CoMP-UFMC.Then we proposed the CFOs estimation algorithm of CoMP-UFMC system.Compared with the maximum likelihood CFO estimation algorithm with multi-dimensional search,the proposed algorithm only need one dimensional search to estimate the CFO of the primary user which reduced the complexity.In addition,this algorithm reach convergence after three iterations and the CFOs estimation performance is better than that of Moose algorithm.