Design Of Subband Filter And Research On ICI Cancellation Technology In F-OFDM System

Filtered Orthogonal Frequency Division Multiplexing is a new multi-carrier modulation technology to meet the higher communication requirements of 5G.F-OFDM system adds subband filters on the basis of OFDM system.The design of subband filters will affect the complexity of system implementation and the power leakage level outside the subband.Therefore,the design of subband filters is the focus of F-OFDM system research.However,the current design of subband filters can't flexibly adjust the filter parameters according to communication scenarios.At the same time,ICI occurs when F-OFDM system is subjected to CFO,which seriously affects the system performance.However,there is still no research on ICI cancellation of F-OFDM system.Therefore,this dissertation takes F-OFDM as the research object and studies the design of subband filters and ICI cancellation technologies.Firstly,aiming at the problem that subband filters can't adjust filter parameters flexibly according to communication scenarios,this dissertation presents a window function method based on Kaiser window to design F-OFDM subband filters.According to different communication requirements,the filter designed in this dissertation can adjust the level between main lobe width and sidelobe amplitude by adjusting the parameters of filter length and controling window function shape.Hence,it overcomes the shortcomings of other sub-band filters,which can't change the stopband attenuation after selecting the specific window function,and has more flexibility.Secondly,aiming at the problem that CFO will have a significant impact on the reliability of the original F-OFDM system,based on the analysis of the causes of CFO and the effects of integer and decimal frequency offset on the F-OFDM system,this dissertation proposes an ICI self-cancellation algorithm for the MCSR of the F-OFDM system.The algorithm Maps symbols which have conjugate relationship to mirror symmetric subcarriers,by exploiting the conjugated relationship of the interference coefficients of the received signals,the ICI interference caused by the CFO and IQ imbalance of the F-OFDM system is reduced.The simulation results show that the BER performance and CIR performance of the proposed MCSR self-cancellation algorithm in F-OFDM are better than other ICI self-cancellation algorithms in AWGN channel.Finally,aiming at the problem that the error rate performance can't be improved obviously with the increasing CFO by using MCSR self-cancellation algorithm,two improved MCSR algorithms are proposed in this dissertation.The first improved algorithm based on frequency offset estimation in time domain,which uses the correlation between CP and its repetitive parts to estimate frequency offset in F-OFDM system,and feedback the estimated frequency offset directly to the receiver for frequency correction.The second improved algorithm based on frequency offset estimation in frequency domain,which estimates the CFO of the system by using the conjugate correlation of the transmitted data on the mirror symmetric subcarriers after FFT transform of the received signal.The estimated carrier frequency offset is fed back to the receiver of the system to compensate the phase of the received signal.The simulation results show that the improved algorithm improves the computational complexity of the receiver,but reduces or even cancels the ICI caused by carrier offset.