Low Complexity Implementation Technology Of Millimeter Wave Line-of-sight MIMO High Speed Communication System

The rapid development of mobile communication and information society has put forward high requirement on broadband high-speed data transmission.Millimeter wave(mmWave)multiple-input multiple-output(MIMO)has become an important technology to realize the high-speed data transmission system.Considering some special require-ments and application scenarios,such as the coverage of remote areas and the construction of emergency communication systems,especially military broadband tactical network,the mm Wave MIMO technology based on airborne mobile platform has become a hot issue in current research.At present,the focus of attention is the implementation of backbone communication network based on airborne platform and mm Wave MIMO technology.This paper mainly focus on this application background and studies the key technology to promote the engineering application of mm Wave MIMO system based on airborne plat-form.The mm Wave MIMO system based on airborne platform has the following main problems:1.The mm Wave airborne MIMO channel lacks scattering and reflection,it can be modeled as line-of-sight(LoS)MIMO channel.Therefore,in LoS MIMO channel,the correlation between antennas is strong,which reduces the performance of MIMO com-munication system.2.The hardware complexity and power consumption of the massive antenna array is high.Hence,the limited space and energy of airborne platform restricts the application of the massive antenna array.In this paper,relevant research work is car-ried out on the above issues,and the main work is summarized as follows:(1)In this paper,we establish the model of point-to-point mm Wave LoS MIMO channel by ray-tracing principal and investigate the fundamental characteristics of LoS MIMO channel.The parameters of the antenna array,including the antenna spacing,the distance between the receiver and transmitter and the angle of the antenna array,all affect the performance of the LoS MIMO system.As a consequence,in LoS MIMO channel,it is important to obtain the optimal layout of antenna array.(2)In this paper,we obtain the optimal layout of uniform linear antenna array(ULA)and uniform plane antenna array(UPA)by analytical method.However,not all types of antenna array have the analytical solution of the optimal layout.Therefore,we propose a more general Newton iterative method,and take the uniform circular antenna array(UCA)as an example to obtain the optimal layout of the antenna array by using Newton iterative method.The uniform circular antenna array is compact and has low sensitivity to mu-tual coupling.So it is suitable to be deployed in airborne platform with space constraint.In LoS MIMO communication system based on airborne platform,the distance between receiver and transmitter is changing constantly,but the parameters of antenna array are always fixed.Therefore,we propose an adaptive transmitting antenna selection algorithm based on channel state information,which can adaptively determine the number and in-dexes of selected antennas with the change of channel state information.The adaptive transmitting antenna selection algorithm based on channel state information can not only reduce the complexity of signal processing and the cost of feedback,but also obtain the performance close to that of the whole antenna array system.(3)Based on the characteristic of LoS MIMO channel,we propose the concept of effective degree of freedom(EDoF)and a hybrid beamforming design algorithm based on dictionary learning(DL).We derive the analytical expression between EDoF and the antenna array parameters.In LoS MIMO communication system,the number of transmit-ted data streams is determined by EDoF,which can be adaptively adjusted by the channel state information.Numerical results show that the adaptive hybrid beamforming based on DL algorithm can achieve the optimal performance of full-digital beamforming system.(4)In this paper,we study the hybrid beamforming of multi-user multiple-input multiple-out(MU-MIMO)communication system.We propose the Tomlinson-Harashima Precoding(THP)based on block diagonalization geometric decomposition(BD-GMD)and apply DL algorithm to the design of hybrid beamforming.Compared with the con-ventional THP based on block diagonalization singular value decomposition(BD-SVD),the proposed scheme can improve the performance of MU-MIMO communication sys-tem.Numerical results indicate that the performance of MU-MIMO system based on the proposed scheme can approach the optimal full-digital beamforming.