Model-based learning for multi-antenna multi-frequency location-to-channel mapping

TL;DR

This paper introduces a model-based neural network architecture for accurately learning the complex, rapidly varying location-to-channel mapping in wireless communications, overcoming spectral bias issues inherent in classical neural models.

Contribution

It proposes a problem-specific neural architecture derived from a propagation channel model that efficiently learns high-frequency variations in location-to-channel mapping.

Findings

Outperforms classical INR architectures on synthetic data

Achieves higher accuracy in modeling rapid channel variations

Demonstrates explainability through model-based approach

Abstract

Years of study of the propagation channel showed a close relation between a location and the associated communication channel response. The use of a neural network to learn the location-to-channel mapping can therefore be envisioned. The Implicit Neural Representation (INR) literature showed that classical neural architecture are biased towards learning low-frequency content, making the location-to-channel mapping learning a non-trivial problem. Indeed, it is well known that this mapping is a function rapidly varying with the location, on the order of the wavelength. This paper leverages the model-based machine learning paradigm to derive a problem-specific neural architecture from a propagation channel model. The resulting architecture efficiently overcomes the spectral-bias issue. It only learns low-frequency sparse correction terms activating a dictionary of high-frequency components. The proposed architecture is evaluated against classical INR architectures on realistic synthetic data, showing much better accuracy. Its mapping learning performance is explained based on the approximated channel model, highlighting the explainability of the model-based machine learning paradigm.