Machine Learning Prediction of Time-Varying Rayleigh Channels

TL;DR

This paper evaluates RNN and LSTM neural networks for predicting time-varying Rayleigh channels, demonstrating their superiority over traditional Wiener predictors in certain scenarios, especially with short observation windows and noisy conditions.

Contribution

It provides a comparative analysis of neural network predictors versus Wiener predictors for Rayleigh channel prediction, highlighting the effectiveness of shallow RNNs in high mobility scenarios.

Findings

Neural network predictors outperform Wiener predictor for small observation windows.

RNNs are more robust under weak channel correlation and noise.

Shallow RNNs effectively model Rayleigh channels across Doppler shifts.

Abstract

Channel state information (CSI) rapidly becomes outdated in high mobility scenarios, degrading the performance of wireless communication systems. In these cases, time series prediction techniques can be applied to combat the effects of outdated CSI. Recently, it has been shown that recurrent neural networks (RNNs) exhibit outstanding performance in time series prediction tasks. In this paper, we investigate the performance of RNN and long short term memory (LSTM) predictors in a simple Rayleigh flat-fading channel. We conduct numerical experiments to evaluate whether these machine-learning (ML)-based predictors can outperform the optimal linear minimum mean square error Wiener predictor. Our simulation results indicate that the considered neural network predictors outperform the Wiener predictor for small observation window lengths and are more robust under weak channel correlation as well as in the presence of noise. Furthermore, we show that simple shallow RNNs are sufficient to model Rayleigh channels over a wide range of Doppler shifts.