Millimeter Wave Channel Estimation Based on Clustering Block Sparse Bayesian Learning

Author(s):  
Jiawen Liu ◽  
Xiaohui Li ◽  
Kun Fang ◽  
Tao Fan
2020 ◽  
pp. 826-832
Author(s):  
Jicheng Dong ◽  
◽  
Wei Zhang ◽  
Bowen Yang ◽  
Xihong Sang

In millimeter wave (mmWave) communication systems, Channel State Information(CSI) is extremely essential for beamforming. The traditional Successive Support Detection (SSD) algorithm can achieve high wideband channel estimation accuracy, but it only used least square (LS) algorithm to recover the detected channel part, the estimation accuracy is low under low SNR regions. To tackle this problem, in this paper, inspired by the classic Support Detection (SD) channel estimation scheme in narrowband, we propose an efficient Wideband Support Detection Sparse Bayesian Learning (WSDSBL) channel estimation scheme. For every subcarrier, we first detect the support of the wideband beamspace channel of the subcarrier, then we use the Sparse Bayesian Learning (SBL) scheme to recover it. Simulation results show that the proposed WSDSBL channel estimation algorithm is better than conventional wideband channel estimation schemes in MSE performance and achievable sum-rate performance, especially in low SNR regions.


Author(s):  
Xu Shuang

With the explosive growth in the number of communication users and the huge demand for data from users, Limited low-frequency resources have been far from being satisfied by users. The combination of Massive MIMO technology and millimeter-wave technology has brought new hope to users. In this paper, several basic algorithms are placed under the millimeter wave large-scale antenna channel for simulation research.


Sign in / Sign up

Export Citation Format

Share Document