Learning the Wireless Channel: A Deep Neural Network Approach

Abstract

In this paper, we propose a new deep neural network (DNN)-based channel estimation method for the Rayleigh fading channel model. While deep learning has been considered for estimating channels in many communication scenarios, direct estimation of the basic wireless single-input single-output (SISO) communication channel coefficients has not been considered. The proposed DNN-based method can efficiently estimate the channel in real time. Extensive simulation results demonstrate that the proposed channel estimator outperforms conventional least square (LS) estimators in terms of bit error rate (BER) and mean square error (MSE). In addition, the proposed channel does not need channel statistics information or complex matrix computation, thereby reducing the amount of calculation significantly.