Canscaded Channel Estimation And Beamforming Design Of Reconfigurable Intelligent Surface Systems

Reconfigurable Intelligent Surface(RIS)is a communication device made of programmable metamaterials.Each reflective unit on it can flexibly and independently adjust the phase and amplitude of incident electromagnetic waves,realizing controllable wireless propagation.environment.RIS has many significant advantages.For example,the operating energy consumption generated by RIS is very low,only the controller part consumes a small amount of energy,and other components are almost passive devices,which do not need to provide additional energy.Compared with traditional amplifiers,it has obvious advantages in energy saving.And RIS has the characteristics of lightweight and easy deployment,which can be conveniently and flexibly deployed on the surface of various objects.The principle of the RIS-assisted communication system is to generate a new communication link between the base station(BS)and the user(UE),which can enhance the strength of the received signal to a certain extent.RIS supports full-duplex communication,so it is easy to be compatible with other communication networks.At the same time,RIS technology also brings many new challenges.For example,its passive characteristics increase the complexity of channel estimation exponentially,and it is imminent to explore new solutions.The channel estimation and beamforming design problems of the RIS-aided single-user system are studied first.The channel state information(CSI)between the BS and the RIS is calculated based on the respective location information of the BS and the RIS.Under the condition that the CSI between the BS and the RIS is obtained,the orthogonal matching pursuit(OMP)algorithm is used to estimate the CSI between UE and RIS.According to the acquired two-segment CSI,the active beamforming of BS and the passive beamforming of RIS are jointly optimized.Then the channel estimation and beamforming design problems of RIS-assisted single-antenna multi-user systems are studied.The UE-RIS-BS cascade channel of a random user is estimated by the Sparsity Adaptive Matching Pursuit(SAMP)algorithm,and then the UE-RIS-BS of the same user is estimated by the variable step size adaptive matching pursuit(Vss AMP)algorithm.BS cascade channel.Then,based on the fact that all users have the same RIS-BS,the concatenated CSI of other users is obtained from the estimated concatenated channel using the linear relationship between the channels.Finally,the active beamforming of the BS and the passive beamforming of the RIS are jointly optimized with the goal of maximizing the received signal-to-noise ratio(SNR)of the user.Finally,the channel estimation and beamforming design problems of dual RIS-assisted single-user mm Wave systems are studied.It is estimated that there are two single RIS cascaded channels and one dual RIS cascaded channel in the dual RIS system,respectively.This paper proposes an alternate estimation scheme based on the Vss AMP algorithm and uses the relationship between the dual RIS cascaded channel and the single RIS cascaded channel to estimate the dual RIS cascaded channel.Beamforming at the BS and RIS is designed according to the estimated CSI.