Research On Key Technology Of Receivers For Multi-carrier And Multi-antenna Systems

As key technologies in the physical layer of mobile communication systems,multi-carrier and multi-antenna technologies can significantly improve the system spectrum efficiency and signal transmission reliability,and are widely used in mobile communication systems.Facing the new-generation wireless communication,such as underwater communications,vehicle-to-everything network,and integration of air and earth,the complex non-Gaussian noise transmission environment,ultra-high terminal movement rate,and ultra-high-speed transmission put forward higher requirements for physical layer technologies.This paper conducts in-depth research on pulse interference cancellation technology,orthogonal time frequency and space(OTFS)technology,and Massive Multipleinput Multiple-output(M-MIMO)decentralized baseband processing(DBP)technology,and designs the corresponding low-complexity and high-performance receiver schemes.The specific research contents are as follows:1)To tackle the performance degradation of traditional Gaussian model-based receivers due to ICI,energy attenuation,and non-Gaussian interference caused by pulse blanking,this dissertation proposes a non-Gaussianity-aware iterative receiver scheme.This scheme mitigates the effects of impulse blanking and residual non-Gaussian interference through iterative ICI removal,energy equalization,and log-likelihood ratio(LLR)compression.Simulation results illustrate that the proposed non-Gaussianity-aware iterative receiver can significantly improve the performance of the traditional pulse blanking method.Furthermore,the proposed receiver can achieve a better performance-complexity trade-off compared with compressive sensing-based sparse recovery receivers.2)To solve the problem that the current receivers have high computational complexity for the OTFS system with fractional Doppler in high-speed scenarios,this dissertation proposes a channel coefficient-aware approximate expectation propagation(AEP)receiver based on the sizes of associated channel coefficients.In addition,a novel OTFS-based pattern division multiple access(PDMA)scheme,shortened as OTFS-PDMA,is proposed to meet the low-latency,high-reliability,and massive connection requirements in high-mobility scenarios.To mitigate the detrimental effect of BS antenna correlation,this dissertation designs a vector EP(V-EP)receiver by jointly processing the received signals of multiple antennas on each delay-Doppler domain resource.A pattern-aware serial mechanism is proposed to facilitate the convergence of the proposed V-EP receiver by exploiting reliable updated messages in a timely manner.An iterative detection and decoding(IDD)structure with parallel interference cancellation(PIC)is incorporated into the proposed receivers to further improve the system performance.Simulation results illustrate that the proposed EP-based receivers provide good performancecomplexity trade-offs,particularly in spatially correlated multiple-input multipleoutput(MIMO)channels.3)To solve the problems,i.e.,excessive interconnect,chip I/O data rates,and power consumption in M-MIMO systems,this dissertation studies the highperformance receiver design scheme for DBP architectures.To mitigate the performance deterioration caused by the constrained information-sharing between antenna clusters,this dissertation proposes a decentralized group-wise detection paradigm for the star architecture by dividing users into multiple user groups,which can be effectively exploited by the factor graph-based message-passing algorithms,such as the expectation propagation(EP)algorithm.Then,the proposed method is extended to the daisy-chain architecture and the corresponding message fusion rule is designed based on the multivariate complex Gaussian probability density product principle.Simulation results demonstrate that the proposed GW-EP receiver achieves significant performance gains over the conventional decentralized methods of the star architecture,especially for scenarios with lower BS antenna-touser ratios and/or spatially correlated channels.For the daisy-chain architecture,the proposed detectors achieve comparable performance to that of the centralized detector with lower front-haul bandwidth requirements.