Research On CSI Feedback Optimization And Channel Calibration In Wi-Fi 7

As a major network access method,Wireless Fidelity(Wi-Fi)is closely related to people’s lives.We are exploring high-speed,high-reliability,low-latency,and low-cost wireless networks.To meet these demands,access points(AP)are required to continue increasing the number of antennas and better spatial multiplexing capabilities.Therefore,techniques such as multi-AP cooperation,16×16 multiple-input multiple-output(MIMO),multi-link operation,and multi-resource unit(RU)allocation are proposed in Wi-Fi 7 to improve the performance of Wi-Fi systems.Among them,multi-AP cooperation and massive MIMO are hot issues of current research,and the performance of the entire multi-AP cooperative system can be greatly improved by using the cooperation among multiple APs.However,the premise of effective multi-AP cooperation is to obtain channel state information(CSI).Therefore,for multi-AP cooperation technology,channel detection under multi-AP cooperation is a key point.However,as the scale of MIMO increases gradually,the burden of channel detection becomes heavier and heavier,which reduces the efficiency of channel detection and affects the overall performance of the system.Therefore,the research goal of this thesis is to optimize the two aspects of CSI feedback and channel calibration in channel sounding to reduce the burden of multi-AP cooperative channel sounding,thereby improving the performance of the entire WiFi system.First of all,this thesis examines the CSI feedback phase in the explicit channel sounding in Wi-Fi7: 1)An improved explicit channel sounding process is proposed and the related frame structure is designed,and the CSI information between all APs and station(STA)is aggregated in a CSI feedback packet and transmitted to the master-AP,and then the master-AP distributes the processed CSI to each slave-AP through wired or wireless means.Through theoretical and simulation analysis,it is verified that the CSI feedback process proposed in this thesis can effectively reduce the time overhead of channel sounding;2)Considering that the CSI feedback by STA when multi-AP collaboration contains CSI information between STA and multiple APs,and the CSI information has strong spatial correlation in the same environment,in order to reduce the CSI feedback overhead,a differential feedback method is used.The CSI feedback overhead can be effectively reduced by means of complete CSI of master-AP and difference of CSI between master-AP and slave-APs and verified by simulation analysis that the differential feedback algorithm can effectively reduce the overhead of CSI feedback;3)Aiming at the problem that in the process of CSI feedback based on principal component analysis(PCA),the computational complexity of the actual calculation of the channel vector covariance matrix becomes difficult to achieve as the number of antennas increases significantly,a two-directional two-dimensional PCA-based channel feedback scheme is proposed.The simulation results show that the feedback scheme proposed in this thesis not only has the characteristics of lower complexity than the traditional PCA channel feedback,but also has better compression performance.And then,the implicit channel sounding method introduced in Wi-Fi 7 is studied,mainly for optimizing the channel reciprocity calibration method of implicit channel sounding: 1)Based on the traditional single-AP channel calibration algorithm,it is extended to multi-AP scenarios,a multi-AP cooperative reciprocity calibration algorithm is proposed,and the corresponding frame exchange process and frame structure are designed;2)Argos-based multi-channel calibration algorithm is proposed.The AP cooperative channel reciprocity calibration algorithm performs relative calibration on the channel reciprocity between multiple APs participating in the multi-AP network,without the need for STAs to participate in the calibration process.Then the corresponding process and related frame structure are designed.Finally,the effectiveness of the two algorithms is verified by simulation.It is shown that the multi-AP cooperative channel reciprocity calibration algorithm based on Argos has better performance than the multi-AP cooperative reciprocity calibration algorithm.