ChanEstNet: A Deep Learning Based Channel Estimation for High-Speed Scenarios

Abstract

Aiming at the problem that the downlink channel estimation performance is limited due to the fast time-varying and non-stationary characteristics in the high-speed mobile scenarios, we propose a channel estimation network based on deep learning, called ChanEstNet. ChanEstNet uses the convolutional neural network (CNN) to extract channel response feature vectors and recurrent neural network (RNN) for channel estimation. We use a large amount of high-speed channel data to conduct offline training for the learning network, fully exploit the channel information in the training sample, make it learn the characteristics of fast time-varying and non-stationary channels, and better track the features of channels changing in high-speed environments. The simulation results show that in the high-speed mobile scenarios, compared with the traditional methods, the proposed channel estimation method has low computational complexity and significant performance improvement.