Research On Channel Estimation And PAPR Reduction Techniques For GFDM Systems Based On FrFT

The application scenarios of the 5th generation mobile communication are very extensive,including Internet of Things,virtual reality,Vehicle-to-everything,etc.Therefore,the waveform applied in 5G should meet the diverse requirements of different application scenarios.Orthogonal Frequency Division Multiplexing(OFDM)technology,as the main waveform technology of 4G,has advantages such as high transmission speed and strong resistance to frequency selective fading.However,its strict orthogonality makes it sensitive to frequency offset,and it has large out-of-band(OOB)leakage and high peak-to-average power ratio(PAPR),making it unsuitable as the waveform for 5G.Generalized Frequency Division Multiplexing(GFDM)technology,as an improved form of OFDM,has insensitivity to frequency offset due to the non-orthogonality between subcarriers.Meanwhile,the introduction of filters reduces OOB leakage,and the PAPR can be adjusted,making it one of the 5G candidate communication waveforms.However,if there is Doppler frequency shift(DFS)in fading channels,the GFDM system may experience communication issues due to inter-carrier interference(ICI)and inter-symbol interference(ISI).This paper discusses the Fractional Fourier Transform Generalized Frequency Division Multiplexing(FrFT-GFDM)system,which is inspired by the Fractional Fourier Transform Orthogonal Frequency Division Multiplexing(FrFT-OFDM)system,uses Chirp signals as the basis replace the sine cosine signals basis in the GFDM system.Since Chirp signals are broadband signals with good resistance to DFS,the FrFT-GFDM system can adapt to different fading channels by changing the transformation order of FrFT.This paper mainly focuses on three aspects to study the FrFT-GFDM system:Firstly,Introducing the signal time-frequency analysis tools needed in the FrFT-GFDM system,and based on this,explaining the system structure and how to design the filters for Zero Forcing(ZF)reception and Minimum Mean Square Error(MMSE)reception using fractional Gabor transform.Then,simulation is performed on the performance of FrFT-GFDM system in terms of OOB,bit error rate(BER),and PAPR,demonstrating its superiority in DFS channels.Since the FrFT-GFDM system itself has limited resistance to frequency offset and its PAPR does not change with the order,this paper proposes two algorithms to improve the system performance from the perspectives of BER and PAPR.Secondly,in order to enable the FrFT-GFDM system to communicate effectively in fading channels,the causes of self-interference under matching filter(MF)reception are analyzed,as well as the impacts of DFS,multipath,double-selective fading channels,and the expressions for interference are given.From the expressions,it can be seen that the interference of fading channels can be divided into ICI and ISI,which ICI caused by DFS and ISI caused by overlapping spectra of filters in different time slots.When channel estimation is used to equalize the fading channel,the overlapping spectra of filters in different time slots will contaminate the pilot information.Therefore,this paper proposes a data-encoded orthogonal pilot channel estimation algorithm for the FrFT-GFDM system under MF reception in fading channels.This algorithm eliminates the self-interference in MF reception through preprocessing,and makes the fractional domain spectra of the first sub-symbol filter orthogonal to the spectra of the remaining sub-symbol filters,thus eliminating the impact of ISI.The impact of ICI is eliminated through data coding,and the design of the coding matrix is inspired by the interference cancellation algorithm in OFDM system.Finally,simulation verify that the proposed algorithm can completely eliminate self-interference in MF reception,and through a comparison of the BER performance between the algorithm and the system itself in three types of fading channels,the superiority of the algorithm in fading channels is demonstrated.Finally,in order to reduce the system’s PAPR performance,the principles of some commonly used PAPR reduction algorithms in multicarrier systems are introduced,including Clipping,partial transmit sequence(PTS),and Discrete Fourier Transform(DFT)precoding.Due to the issues such as OOB and BER emissions,this paper proposes an algorithm that uses sub-carrier reduction PAPR(PRSC)based on Clipping,which reduces the system’s PAPR without changing the OOB and BER,and combines it with the PTS algorithm.Finally,simulation verify that the proposed algorithm has good performance in terms of PAPR,OOB,and BER.