Approximating the Void: Learning Stochastic Channel Models from Observation with Variational Generative Adversarial Networks

TL;DR

This paper introduces a variational GAN architecture to accurately learn stochastic wireless channel models from observations, improving the representation of channel response PDFs for better communication system design.

Contribution

The paper presents a novel variational GAN architecture and loss function specifically designed to accurately learn the probability distribution functions of stochastic wireless channels.

Findings

Successfully captures complex channel PDFs

Outperforms previous GAN-based methods in accuracy

Effective across various realistic channel distributions

Abstract

Channel modeling is a critical topic when considering designing, learning, or evaluating the performance of any communications system. Most prior work in designing or learning new modulation schemes has focused on using highly simplified analytic channel models such as additive white Gaussian noise (AWGN), Rayleigh fading channels or similar. Recently, we proposed the usage of a generative adversarial networks (GANs) to jointly approximate a wireless channel response model (e.g. from real black box measurements) and optimize for an efficient modulation scheme over it using machine learning. This approach worked to some degree, but was unable to produce accurate probability distribution functions (PDFs) representing the stochastic channel response. In this paper, we focus specifically on the problem of accurately learning a channel PDF using a variational GAN, introducing an architecture and loss function which can accurately capture stochastic behavior. We illustrate where our prior method failed and share results capturing the performance of such as system over a range of realistic channel distributions.