Semantic Change Driven Generative Semantic Communication Framework

TL;DR

This paper introduces a novel semantic change driven generative framework for semantic communication, enabling independent optimization of encoder and decoder, reducing energy consumption, and improving remote scene regeneration in AI-powered monitoring.

Contribution

It proposes a modular semantic encoder and a diffusion probabilistic model-based decoder for improved semantic communication and scene regeneration, addressing black-box issues and energy efficiency.

Findings

Effective semantic encoder and decoder demonstrated in simulations

Significant energy savings potential shown

Enhanced remote scene regeneration capability

Abstract

The burgeoning generative artificial intelligence technology offers novel insights into the development of semantic communication (SemCom) frameworks. These frameworks hold the potential to address the challenges associated with the black-box nature inherent in existing end-to-end training manner for the existing SemCom framework, as well as deterioration of the user experience caused by the inevitable error floor in deep learning-based SemCom. In this paper, we focus on the widespread remote monitoring scenario, and propose a semantic change driven generative SemCom framework. Therein, the semantic encoder and semantic decoder can be optimized independently. Specifically, we develop a modular semantic encoder with value of information based semantic sampling function. In addition, we propose a conditional denoising diffusion probabilistic mode-assisted semantic decoder that relies on received semantic information from the source, namely, the semantic map, and the local static scene information to remotely regenerate scenes. Moreover, we demonstrate the effectiveness of the proposed semantic encoder and decoder as well as the considerable potential in reducing energy consumption through simulation based on the realistic $\mathcal{F}$ composite channel fading model. The code is available at https://github.com/wty2011jl/SCDGSC.git.