Performance Evaluation And Analysis Of New Waveform Transmission For 5G Communication Systems

With the rise of the mobile Internet and the Internet of Things(IoT),the fifth generation(5G)mobile communication system puts forward higher requirements,including network traffic,energy efficiency,number of connections and so on.For the reason that the ability of Orthogonal Frequency Division Multiplexing(OFDM)technology to deal with some 5G new scenarios is limited,such as strict synchronization requirements harmful to the machine-to-machine(M2M)communication requiring lower power consumption,this thesis mainly focuses on new multi-carrier modulation techniques for 5G communication systems,including the performance analysis and optimization design of generalized frequency division multiplexing(GFDM)in non-ideal environments,and low complexity system models.The main contents of this thesis are summarized as follows:Firstly,basic concepts and developments of different multi-carrier modulation technologies are introduced.And the comparison is made from structure realization,low delay and channel equalization.Then,the performance of GFDM is analyzed from power spectrum,the compatibility between GFDM and MIMO,nonlinearity of power amplifier and IQ imbalance.The research status at home and abroad and features of GFDM can be known through numerical simulations,which further confirm the main research contents of this thesis.Secondly,combined with the structure of GFDM time-frequency resource block,an algorithm for IQ imbalance compensation at GFDM receiver is proposed.In the absence of channel state information(CSI),IQ imbalance parameters are directly estimated by using well-designed training sequences,which greatly reduces the training consumption.Compared with the first estimate of the channel and then obtaining IQ imbalance parameters,the proposed algorithm can relax the requirement on time-invariant channel over adjacent symbols and is suitable for different systems.Simulation results show that the proposed algorithm can obtain more accurate IQ imbalance parameters compared with some existing algorithms.Thirdly,the training sequence for channel estimation and IQ imbalance compensation in GFDM systems is designed.Combined with the GFDM system model based on sparse frequency domain processing,the received signal distorted by IQ imbalance is analyzed in detail.Let signals at particular frequency points meet the proposed constraints to estimate channel and IQ imbalance parameters without increasing training cost,and thus the expression of the designed training sequence can be derived.Then,the rest of the channel frequency domain response and IQ imbalance parameters are obtained by the linear interpolation and the leastsquares equalization(LSE)technique,respectively.Simulation results show that the proposed algorithm can estimate channel and IQ imbalance parameters with same training overhead.Finally,considering that the good properties of GFDM are traded at the cost of higher complexity,a low-complexity GFDM system model is studied in this thesis.By exploiting the feature that a prototype filter performs the cyclic shifting operation in time and frequency,the modulation matrix in the transmitter can be represented as the product of two simple matrices,and then the computational complexity of the GFDM transmitter is reduced.Following the reduced-complexity transmitter processing,low complexity MF?ZF and MMSE receivers are designed in this thesis and a unified demodulator structure is derived.Simulation results show that the proposed GFDM system significantly reduces the computational burden without losing any system performance.