Key Technologies Of Line-of-sight Multi-antennas Wireless Communications

With the development of the mobile internet and the internet of things(IoT),new communication scenarios emerge in endlessly and line-of-sight(LoS)multi-input and multi-output(MIMO)wireless communication scenario is an important one.In fixed wireless access and backhaul links,the line-of-sight multi-antennas system can be used to solve the last kilometer broadband access problem in communications.It is used as an alternative technology to fiber and digital subscriber line(DSL).The line-of-sight multi-antennas system can be used in some mountains,rivers with complex terrain and places where it is difficult to lay fibers or where construction costs are relatively high.In addition,LoS MIMO is also suitable for high-speed railway communication systems.Since the high-speed train travels in the wilderness most of time,the scatters are very sparse in this scenario.What's more,with the high propagation loss,poor diffraction ability,and other inevitable limitations,the sparse scattering property in high frequency channel makes LoS transmission become dominant.Different from the non-line-of-sight transmission mode in Rayleigh fading channel,the channel matrix in the line-of-sight scenario is usually singular,which will greatly affect the capacity of systems.In addition,due to the different channel environments,the transmission schemes will be different.The channel state in the line-of-sight scenario is restricted by the angle information of the propagation channel,so the angle estimation should be paid more attention.Therefore,this thesis mainly focuses on the key technologies in LoS MIMO wireless communication,including the design of antenna array configuration and the estimation of direction of arrival(DOA)in the separated digital-analog hybrid beamforming system.The main contents of this thesis are summarized as follows.Firstly,the MIMO technology is introduced,including spatial diversity and multiplexing technologies,channel capacity and effective degrees of freedom.Then,the characteristics and modelings of wireless channel are elaborated in detail,focusing on the Rayleigh fading channel,Rice channel and pure line-of-sight channel.Next,the optimization of linear array configuration in pure line-of-sight channel is studied,and the corresponding numerical results are given.Then,the transmission schemes of the separated digital-analog hybrid antenna system in line-of-sight scenario are described.At last,the traditional DOA estimation algorithms,including Bartlett algorithm and multiple signal classification(MUSIC)algorithm are discussed,and the limitations and shortcomings of these traditional DOA estimation algorithms in the application of the separated digital-analog hybrid beamforming system are analyzed,which further confirm the main research contents of this thesis.Secondly,we investigate the design of multiple panel arrays(MPAs)in LoS MIMO communication systems.We utilize the spherical-wave model for the LoS channel component.The fundamental relations among key design parameters in MPAs LoS MIMO system are studied to obtain the maximum spatial multiplexing gain according to the effective degrees of freedom and channel capacity.We further analyze the performance of MPAs system when the array size and transmission distance are not matched and propose an algorithm to optimize the location of antennas.Numerical results indicate that the proposed algorithm can reduce the computational load compared with the exhaustive approach for the similar system performance.Moreover,the optimized panel antenna array is robust to the geometrical change,misplacement,and signal-to-noise ratio.Thirdly,aiming at the separated digital-analog hybrid beamformin systems,two new DOA estimation algorithms are proposed,which are curve fitting method and beam scanning and optimal beam migration method.Then,the complexities of these two algorithms and traditional DOA estimation algorithms such as MUSIC are analyzed,and the corresponding numerical simulations are carried out.The results show that the curve fitting method is simple and intuitive,and has better performance when the number of sampling points is small.In addition,the DOA estimation algorithm based on beam scanning and optimal beam migration has low computational complexity and high accuracy.Finally,an intelligent DOA estimation method driven by model and data is proposed for the separated digital-analog hybrid systems,which combines the traditional wireless communication algorithm with artificial intelligence.The method can be divided into off-line training process and angle prediction process.In the off-line training process,three learning algorithms including support vector regression(SVR),gradient boosting regression(GBR)and a new neural network architecture named as MicroFishNet are used to train the model,respectively.The simulation results show that the intelligent DOA estimation algorithm can achieve higher accuracy while reducing the complexity than the traditional DOA estimation algorithms.