Research On Key Techniques Of Universal Filter Multi-Carrier

In recent years,with the rapid development of mobile Internet and the growing popularity of wireless intelligent terminals,the accelerated growth in the number and variety of intelligent terminals and the density of terminal connections has led to diversified demands from data businesses,showing an explosive growth trend in the global data business.All of these are key indicators of the 5G communication system being richer than other generation communication systems.The performance of the 5G communication system in terms of energy efficiency,spectral efficiency,and transmission rate is higher,and the requirements of the 5G communication system in terms of system security,coverage,and transmission delay have improved significantly.As of now,extensive research is being conducted on 5G communication systems.The Universal-Filtered Multi-Carrier(UFMC)system features high restraint outside the band,allows for more flexible allocation of spectrum resources,and is robust to carrier frequency offset(CFO),more relax synchronization.It is suitable for massive machine communication small packet application scenarios.The introduction of the sub-filters brings new problems,such as inter-band interference,more complex,and whether the solutions used for the traditional OFDM system can work effectively in case of the UFMC system.Many scholars have made effective researches on the above problems,but some key technical problems have not been fully solved.Especially,Some key technical problems relating to high power LED chip manufacturing have not been fully solved,especially,influence of interval of sub-bands on the energy properties of the UFMC system,influence of the parameter selection of filters on the PAPR of the UFMC system,improving power efficiency,and improving the channel estimation performance on high speed mobile environment.The paper studies deeply about these problems and the main research and innovation achievements are presented as follows:1.In this study,the authors have thoroughly studied the energy properties of the UFMC system and compared them with the energy properties of OFDM,which can effectively reduce inter-band interference and improve spectrum efficiency.The results show that the energy of UFMC is composed of three parts: the useful energy of the signal,energy of inter-carrier interference(ICI),and energy of inter-band interference(IBI).The useful energy of the signal and the energy of ICI do not change with the interval of sub-bands,whereas the energy of IBI changes with the interval of sub-bands and the parameters of the FIR filters.The transmission performance of UFMC is verified by simulation in a two sub-band scenario.The simulation results show that intervals of subbands affect the system transmission performance when the transmission powers remain unchanged.In the two adjacent sub-bands scenario,when SNR is high(>10 dB)and the SER of system meets certain condition,the spectrum efficiency can be increased by reducing the filter length.In addition,the influence of CFO on the performance of the UFMC system is analyzed theoretically.The simulation result shows that MSE decreases rapidly with increasing SNR when CFO is large.The UFMC system is robust to CFO.2.The time variation of the channel is enhanced in the high-speed mobile scenario,resulting in performance degradation.By making complete use of the temporal correlation and the sparse property of the radio channel,in this thesis,we establish a sparse channel model,and propose an improved priori information assistance CoSaMP algorithm,which significantly reduces the number of poilts and computational complexity,improves the system spectrum efficiency,and achieves better CSI performance.First,the main delays and paths of the channel are detected as priori information based on autocorrelation of the PN sequence.The CIRs of several successive symbols are averaged in the coherence time to improve the precision of rough estimation with no extra calculation burden.Second,the accuracy of path delays is improved and the computational complexity dramatically is reduced using the proposed PPI-CoSaMP algorithm.Finally,the path amplitude is estimated precisely using LS algorithm.3.The sparsity of channel is not possible to obtain in most communication systems.To solve this problem,a priori information assistance SAMP improved algorithm is proposed.The main ideas of the proposed SAMP improved algorithm are similar to those of the CoSaMP improved algorithm,which also increases the estimation performance by priori information.The PPI-SAMP algorithm is proposed to overcome the limitation of the PPI-CoSaMP algorithm.The PPI-SAMP algorithm adjusts the iterative step automatically based on the prediction residuals.Meanwhile,the size of the supporting set changes as well,which ensures that the estimated signals match the original signals to the best extent possible.The improved SAMP algorithm solves the problem of sparsity.Compared to the conventional SAMP algorithm,the improved SAMP algorithm significantly reduces the computational complexity and improves the estimation accuracy.4.The PAPR of the UFMC system is researched theoretically and a PAPR reduction scheme based on filter optimization is proposed to increase power efficiency.First,the mathematical model of PAPR and CCDF of UFMC are deduced.According to the expression of the PAPR of UFMC system,the PAPR performance depends not only on the number of sub-carriers but also on the parameter selection of filters.Then,a PAPR reduction scheme based on filter optimization is proposed.Based on the CCDF mathematical model,CCDF optimization is translated into the standard minimization problem of Lagrange multipliers,and the theoretical expression of the optimal CCDF and filter parameters are demonstrated.5.To reduce the PAPR of the UFMC system,an improved PTS algorithm is proposed.A comparison between the proposed method and the conventional PTS method shows that the former significantly reduces the complexity.The improved PTS algorithm can effectively reduce the PAPR value and significantly enhance the BER performances of the UFMC system as compared to the clipping algorithm.First,on the basis of the classical PTS theory of the OFDM system,the sub-bands are divided into several subblocks.Second,the threshold value of the minimum possible peak power is preset;thus,the number of sampling points in the candidate set can be reduced.Finally,the relational of the phase-weighting sequences are used to decrease the cost of computing of the sampling points multiplied by the phase rotation factor vectors.Compared to the conventional PTS algorithm,the computational complexity is further reduced without degrading the PAPR of system capacity.