Discrete-Space Generative AI Pipeline for Semantic Transmission of Signals

TL;DR

This paper presents Discernment, a semantic communication system that uses GenAI models to transmit the meaning of signals over noisy channels, dynamically adapting to impairments to preserve semantic integrity.

Contribution

The introduction of Discernment, a novel semantic communication system that switches between autoregressive and diffusion models based on channel conditions to maintain signal meaning.

Findings

Maintains semantic integrity under severe channel degradation.

Adapts dynamically to erasure patterns with minimal performance decline.

Operates efficiently with low model complexity for IoT applications.

Abstract

We introduce Discernment, a semantic communication system that transmits the meaning of physical signals (baseband radio and audio) over a technical channel using GenAI models operating in discrete spaces. Discernment dynamically adapts to channel impairments - modeled as erasure channels - by switching between an autoregressive or a diffusion-based generative algorithm, depending on the erasure pattern. Our results show that Discernment maintains semantic integrity even as channel capacity severely degrades, exhibiting very small and graceful performance decline in both classification accuracy and statistical fidelity of the reconstructed meaning. These findings demonstrate Discernment's ability to adjust to diverse physical channel conditions while maintaining spectral efficiency and low model complexity, making it well suited for IoT deployments and strongly motivating further research on this semantic channel paradigm.