Research On Channel Estimation And Beamforming For Intelligence Reflecting Surface-aided Two-way Relaying Network

Intelligent reflecting surface(IRS)is a passive communication device for 6G,consisting of low-cost fully passive reflecting units,which can be used to adjust the phase and amplitude of the incident signal in real time through IRS controllers,to reconfigure the wireless propagation environment in a green and smart manner and improve spectrum efficiency.The conventional two-way relaying network(TWRN)can effectively improve the communication quality of cell edge,provide information exchange between users and reduce disconnection rate,but has the disadvantage of high-power consumption and circuit cost compared to IRS.The combination of IRS and TWRN can be used to achieve a good balance between circuit cost,spectral efficiency and energy consumption.This thesis focuses on IRS-assisted TWRN and the main research and innovation points are as follows.1)To address the problem of channel estimation and pilot optimization for IRSassisted decode-and-forward TWRN,using the symmetry of the received signal,firstly,a novel pilot training structure is proposed,which can subtly separate the directly and cascaded channel components.Subsequently,a channel estimation method based on minimizing the mean square error is designed.The method not only gives the suboptimal closed-form solution of the pilot and phase shift training matrix,but also estimates the channel state information.Finally,the performance impact of on the channel estimation of IRS equipped with finite discrete phase shifters is analyzed and an approximate closedform solution of the theoretical performance loss factor is derived.The simulations show that the optimal performance can be achieved with only 1-bit phase shifter when the Hadamard matrix is used for the phase shift training matrix,while the discrete Fourier transform matrix requires 3～4-bit phase shifters to achieve the approximate optimal performance.2)In order to improve the sum rate of IRS-assisted amplify-and-forward TWRN,a joint beamforming and phase shift matrices optimization method is proposed.The method optimizes the following four variables alternately iteratively with the criterion of maximizing the minimum rate: relay beamforming matrix,reflection phase shift matrix of the first time slot,reflection phase shift matrix of the second time slot.In which,the beamforming matrix is obtained as a low-complexity semi-closed-form solution by a singular value decomposition-based method,and the phase shift matrices of the first and second time slots are solved separately by successive convex approximation.Simulations show that the IRS-assisted system model can achieve significant rate gain over the noIRS model by the proposed alternating optimization algorithm.