Eigenvalue domain-based channel estimation in massive MIMO systems
In massive multiple input, multiple output (MIMO) systems, the base station can pair spatially orthogonal multiple users by collecting statistical data regarding the channel covariance function of each user in advance, and channel estimation using the Bayesian criterion can effectively remove the uplink pilot contamination of paired users.However, the real-time estimation of a users channel covariance function must then be conducted again if the base station occupies too much time conducting negotiation and collecting statistical data, or if the users position changes greatly due to high-speed motion.Therefore, this paper proposes a method for the real-time estimation of the multiuser channel covariance function based on the eigenvalue domain.Firstly, the subspace method is employed to determine the eigenvalue range of a single user in the eigenvalue domain so as to confirm the corresponding channel covariance function with no a priori channel information.On this basis, the estimation of the multiuser channel covariance function is achieved by iterative cancellation.Simulation results demonstrate that the proposed method is both effective and robust,and its estimation performance degrades to a much lesser extent than conventional estimation methods with an increasing number of users, making it applicable to massive MIMO applications.
5G MIMO PAS AOA OFDM
Feng DING Xin WANG Min-zhao ZHAO
Beijing Aerospace YILAI Electronic Technology Co., Ltd.Beijing, China
国际会议
2018 International Conference on Defence Technology (2018国际防务技术会议)
北京
英文
1041-1049
2018-10-21（万方平台首次上网日期，不代表论文的发表时间）