Research On Channel Estimation And Equalization For UFMC System

Universal Filtered Multi-Carrier(UFMC)technology is a new non-orthogonal multi-carrier modulation technology,which is considered to replace Orthogonal Frequency Division Multiplexing(OFDM)and become one of the candidates for transmitting waveforms in 5G some communication scenarios.UFMC can be seen as a combination of OFDM and FBMC waveform techniques while retaining the respective advantages of both.The main application scenarios of UFMC are short data packet burst,low latency and wide synchronization.This article mainly studies the channel estimation and equalization technology in UFMC system.First of all,the basic principle of UFMC system is described,including the realization of signal modulation,transmission and receiving modules and mathematical expressions.On this basis,simulation software is used to build the UFMC emulation system.Then,performance analysis and result simulation of UFMC system from anti-time-frequency deviation performance,peak-to-average ratio,spectral efficiency,filter length,etc.,which is compared with OFDM and demonstrates UFMC application advantages in the Internet of things,sensor network.Regarding the channel estimation of UFMC system,the performance of UFMC symbols under Rayleigh multipath channel is first analyzed.The UFMC channel estimation can reuse the existing OFDM channel estimation methods.Multiplexing mode and channel estimation expression of the traditional LS channel estimation,MMSE channel estimation,the DFT-based conversion domain channel estimation algorithm in the UFMC are studied.UFMC uses sub-band filters,and the mean squared error of the channel estimation in the frequency domain will become larger.In addition,the traditional DFT channel estimation algorithm can reduce the influence of noise.However,UFMC does not use the cyclic prefix and the filter length is resistant to multi-path fading of the system.The optimal channel sampling length is difficult to determine.Aiming at the above problems,a UFMC channel estimation scheme,which smooth the channel estimation information in the frequency domain to reduce the influence of the filter and designs the optimal sampling channel length in the time domain to filter out the noise.Simulations show that this scheme can effectively reduce the influence of noise and improve the channel estimation accuracy.In the pilot-based channel estimation method,the design of the pilot is very important.The position,quantity,and value of the pilot will all affect the channel estimation error.Assuming that the UFMC system uses LS channel estimation and linear interpolation to recover the channel frequency domain response,the UFMC uniform comb pilot expression based on channel estimation error minimization is solved.Because the UFMC modulates signals in units of sub-band,a new pilot pattern is proposed in this paper.The pattern is used for channel estimation in sub-band.Simulations show that the overall MSE of the pilot pattern is smaller than that of the OFDM uniform distribution channel estimation and is more suitable for channel estimation of UFMC systems.The sub-band filter makes the UFMC system itself have certain anti-ISI and ICI characteristics,but when the channel delay spread is large,this characteristic becomes insufficient,which affects the UFMC application range.On the study of ISI and ICI generation in the UFMC,channel shortened time domain equalization methods are introduced to improve the original system structure and channel adaptive equalization scheme is designed,which improves the UFMC system's communication performance.For the MSSNR algorithm,a search window is set to reduce the ergodic range of the algorithm,which reduces the computational complexity of the original algorithm.