Research On Key Technology Of Terahertz Transmission Based On Intelligent Reflecting Surface

With the explosive growth of the data flow in wireless communications,the demand of the high-speed data rate in communication networks is more and more significant.Due to the extremely rich spectrum resources,terahertz communication is effective to solve the problem of data congestion in wireless transmissions.The terahertz channel is composed of the line-of-sight path and the reflection path,in which the line-of-sight path plays a major role since the path loss of it is smaller than the relfection path.When the line-of-sight transmission in the terahertz communication is broke down by obstocles,the transmission quality of the terahertz communication link drops sharply.Thus,it is necessary to enhance the performance of the terahertz system with the aid of relays.Compared with traditional relays,intelligent reflecting surface has many advantages,such as no self-interference effect on electromagnetic signals,no amplication effect on noise power,low cost,low power consumption,low manufacturing difficulty,and easy for preparation,installation and application.The intelligent reflecting surface is composed of a large number of passive elements,and each element can independently adjust the phase of the electromagnetic waves on it.The reflected electromagnetic waves of all elements can form beams based on the interference principle,and the direction of the beams can be flexibly adjusted.In this paper,we consider the joint beamforming design for the ideal intelligent reflecting surface enhanced terahertz multi-input multi-output communication system.For the ideal case of the intelligent reflecting surface,the value of the relfection coefficient is 1,and the phase is continuous between zero and 360 degree.For maximizing the spectral efficiency of the tarehertz system,we make an approximate expression of it,and propose some methods for optimizing the phase shift matrix of the intelligent reflecting surface,including conjugate gradient algorithm,semidefinite programming algorithm based on eigen-decomposition and mapping,semidefinite programming algorithm based on random Gauss,and alternating direction method of multipliers.And then,we joint design the precoding matrix of the base station and the combining matrix of the user by orthogonal matching pursuit algorithm.Simulation results show that when the channel between the base station and the user is composed of the line-of-sight path and the reflecting path or the number of data streams is 1,the spectrum efficiency performance of the proposed algorithms can reach to the optimal performace of the exhaustive algorithm.In addition,we consider the joint beamforming design for the non-ideal intelligent reflecting surface enhanced terahertz multi-input multi-output communication system.For the non-ideal case of the intelligent reflecting surface,the value of the relfection coefficient is smaller than 1,and the phase is discrete.In order to maximize the spectral efficiency of the system,we use the hybrid beamforming method based on orthogonal matching pursuit algorithm for designing the precoding matrix of the base station and the combining matrix of the user,and we use the cross entropy algorithm and coordinate ascent algorithm for designing the phase shift matrix of the intelligent reflecting surface.Simulation results show that whether the channel between the base station and the intelligent reflecting surface is composed of line-of-sight path and the reflecting path or only composed of the line-of-sight path,and whether the number of the data stream is 1or more than 1,the spectral efficieny performance of the proposed algorithm can reach to the optimal performance of the exhaustive algorithm.Finally,we consider the security rate optimization for ideal intelligent reflecting surface enhanced terahertz multi-input single-output wiretap system.For the system,we first establish the mathematical formulation,and then analyze the security rate maximization problem according to the formulation.For maximizing the security rate,we propose an alternative method for designing the phase shift matrix of the intelligent reflecting surface and the precoding matrix of the base station.Specifically,when the phase shift matrix is fixed,the precoding matrix can be obtained by solving the Rayleigh Ritz problem.And when the precoding matrix is fixed,the phase shift matrix can be obtained by the alternating direction method of multipliers.Simulation results show that the proposed method is able to achieve a better security rate performance than traditional algorithms.