Model-based learning for location-to-channel mapping

TL;DR

This paper introduces a model-based neural network architecture that effectively maps user location to channel coefficients by separating frequency components, improving accuracy over standard methods in wireless communication systems.

Contribution

It proposes a novel hypernetwork architecture that isolates high frequency components, addressing spectral bias in neural networks for location-to-channel mapping.

Findings

Outperforms standard neural networks on synthetic data

Effectively separates low and high frequency components

Enhances channel estimation accuracy

Abstract

Modern communication systems rely on accurate channel estimation to achieve efficient and reliable transmission of information. As the communication channel response is highly related to the user's location, one can use a neural network to map the user's spatial coordinates to the channel coefficients. However, these latter are rapidly varying as a function of the location, on the order of the wavelength. Classical neural architectures being biased towards learning low frequency functions (spectral bias), such mapping is therefore notably difficult to learn. In order to overcome this limitation, this paper presents a frugal, model-based network that separates the low frequency from the high frequency components of the target mapping function. This yields an hypernetwork architecture where the neural network only learns low frequency sparse coefficients in a dictionary of high frequency components. Simulation results show that the proposed neural network outperforms standard approaches on realistic synthetic data.