ParamNet: A Multi-Layer Parametric Network for Joint Channel Estimation and Symbol Detection

TL;DR

This paper introduces ParamNet, a multi-layer parametric neural network designed for joint channel estimation and symbol detection in impaired communication systems, optimizing a soft thresholding parameter during training.

Contribution

It presents a novel parametric network architecture that effectively models hardware impairments and optimizes detection parameters in a unified framework.

Findings

Improved joint channel estimation and data detection accuracy

Effective handling of hardware impairments and noise

Parameter optimization enhances detection performance

Abstract

This paper proposes a parametric-based network architecture for joint channel estimation and data detection in communications systems with hardware impairments. This architecture is composed of a data-augmented layer, a custom soft thresholding function, and several linear layers modeling the effect of channel effects and hardware impairments. In the proposed network, the soft thresholding function softly constrains the detected data to be within the considered constellation. The latter depends only on one one parameter that is optimized during training. The benefit of the proposed approach is illustrated through a communication chain corrupted by multiple impairments and noises.