Mixture of Semantics Transmission for Generative AI-Enabled Semantic Communication Systems

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Ni, Junjie, Wu, Tong, Chen, Zhiyong, Xu, Yin, Tao, Meixia, Zhang, Wenjun
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915490198192128
author Ni, Junjie
Wu, Tong
Chen, Zhiyong
Xu, Yin
Tao, Meixia
Zhang, Wenjun
author_facet Ni, Junjie
Wu, Tong
Chen, Zhiyong
Xu, Yin
Tao, Meixia
Zhang, Wenjun
contents In this paper, we propose a mixture of semantics (MoS) transmission strategy for wireless semantic communication systems based on generative artificial intelligence (AI). At the transmitter, we divide an image into regions of interest (ROI) and reigons of non-interest (RONI) to extract their semantic information respectively. Semantic information of ROI can be allocated more bandwidth, while RONI can be represented in a compact form for transmission. At the receiver, a diffusion model reconstructs the full image using the received semantic information of ROI and RONI. Compared to existing generative AI-based methods, MoS enables more efficient use of channel resources by balancing visual fidelity and semantic relevance. Experimental results demonstrate that appropriate ROI-RONI allocation is critical. The MoS achieves notable performance gains in peak signal-to-noise ratio (PSNR) of ROI and CLIP score of RONI.
format Preprint
id arxiv_https___arxiv_org_abs_2509_09499
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Mixture of Semantics Transmission for Generative AI-Enabled Semantic Communication Systems
Ni, Junjie
Wu, Tong
Chen, Zhiyong
Xu, Yin
Tao, Meixia
Zhang, Wenjun
Information Theory
In this paper, we propose a mixture of semantics (MoS) transmission strategy for wireless semantic communication systems based on generative artificial intelligence (AI). At the transmitter, we divide an image into regions of interest (ROI) and reigons of non-interest (RONI) to extract their semantic information respectively. Semantic information of ROI can be allocated more bandwidth, while RONI can be represented in a compact form for transmission. At the receiver, a diffusion model reconstructs the full image using the received semantic information of ROI and RONI. Compared to existing generative AI-based methods, MoS enables more efficient use of channel resources by balancing visual fidelity and semantic relevance. Experimental results demonstrate that appropriate ROI-RONI allocation is critical. The MoS achieves notable performance gains in peak signal-to-noise ratio (PSNR) of ROI and CLIP score of RONI.
title Mixture of Semantics Transmission for Generative AI-Enabled Semantic Communication Systems
topic Information Theory
url https://arxiv.org/abs/2509.09499