Research On Channel Estimation Algorithm Of IRS Communication System Based On Residual Network And Path Symmetry

The Intelligent Reflecting Surface(IRS)regulates the incident electromagnetic waves through software,and reconfigures the wireless communication environment with lower power consumption,thereby realizing precise 3D beamforming,showing great potential in 5G communication,Internet of Things and other fields.The premise of beamforming is to obtain accurate channel state information,and the IRS-assisted communication system currently lacks an effective channel estimate solution.This paper investigates channel state estimation in IRS-assisted communication systems,not only provides a general solution,but also provides an effective channel estimation method for the special scenario of cross-band communication.The following are the key research findings of this paper:Aiming at the high complexity of the high-precision estimation algorithm of the IRS-assisted communication system,this paper proposes the RCE(Resnet-Channel Estimation)channel estimation model,which combines the least squares method and the residual network.The leastsquares method with low complexity is used to initially estimate the entire reflection link passing through the IRS,and then it is enhanced by the residual network to obtain a more accurate channel state.Meanwhile,this paper discusses a method for separately acquiring the state information of the IRS-base station channel and the user-IRS channel.The method in this paper does not require additional work on the reflecting surface to participate in the calculation,and the algorithm complexity is low.Simulation experiments show that when compared to the least square method,the estimation error of the reflection channel is reduced by more than ten times,and the estimation accuracy is better than that of the linear minimum mean square error estimator with lower complexity.In the cross-band communication scenario,since the intelligent reflecting surface introduces an additional communication channel,the feedback overhead between uplink and downlink is even greater.In this regard,this paper proposes a method for estimating the downlink channel through the uplink channel.By analyzing the channel state expression,using the path symmetry between the uplink and the downlink,the channel estimation problem is transformed into a transmission distance estimation problem,and then the uplink channel is used to predict the downlink channel state operating at different frequencies.The proposed method does not require any feedback information,thus saving channel resources.Simulation experiments show that in a communication environment with a signal-to-noise ratio higher than-10 d B,the channel with a frequency interval of 10 MHz can be accurately estimated.