Deep Learning Based Channel Estimation for 5G
Xingyu Guo, Xiaoxu Liu, Xianfeng Gong · 2023
In this paper, we propose a channel estimation method based on deep learning and image Super-Resolution for an orthogonal frequency division multiplexing (OFDM) system in 5G scenario. To be specific, Cascading Residual Channel Estimation Network (CRCEN) and Resource Block Network (RBnet) are introduced in order to meets different needs. Because of well designed network architecture, the computational cost and the number of parameters of CRCEN can be greatly reduced. Moreover, CRCEN shows outstanding performance compared with other models. RBnet is a denoising network which can run on any resource block to support multi-user requirement. Both models have great potential in 5G applications.