Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Multi-User Semantic Fusion for Semantic Communications over Degraded Broadcast Channels

Tong Wu, Zhiyong Chen, Meixia Tao, Bin Xia, Wenjun Zhang

TL;DR

This work tackles semantic communications over two-user degraded broadcast channels by introducing a CSI-aware, Transformer-based fusion framework that jointly encodes and fuses both users' semantic features. A DBC-Aware JSCC backbone embeds CSI into the encoder/decoder and a semantic fusion (SF) module combines the two users' features into a shared latent representation for broadcast, with a fusion ratio α balancing performance between users. A tailored loss L3 is proposed to maximize the semantic performance region, and experiments on CIFAR10, STL-10, and CelebA demonstrate that the SF scheme with DBC-Aware channel adaptation outperforms traditional TD/PA and JPEG2000+Capacity baselines, while incurring modest resource overhead. The approach shows robustness to channel estimation errors and varying SNR gaps, offering a practical pathway for efficient multiuser semantic broadcasting in wireless networks and potential extension to more users.

Abstract

Degraded broadcast channels (DBC) are a typical multiuser communication scenario, Semantic communications over DBC still lack in-depth research. In this paper, we design a semantic communications approach based on multi-user semantic fusion for wireless image transmission over DBC. In the proposed method, the transmitter extracts semantic features for two users separately. It then effectively fuses these semantic features for broadcasting by leveraging semantic similarity. Unlike traditional allocation of time, power, or bandwidth, the semantic fusion scheme can dynamically control the weight of the semantic features of the two users to balance the performance between the two users. Considering the different channel state information (CSI) of both users over DBC, a DBC-Aware method is developed that embeds the CSI of both users into the joint source-channel coding encoder and fusion module to adapt to the channel. Experimental results show that the proposed system outperforms the traditional broadcasting schemes.

Multi-User Semantic Fusion for Semantic Communications over Degraded Broadcast Channels

TL;DR

This work tackles semantic communications over two-user degraded broadcast channels by introducing a CSI-aware, Transformer-based fusion framework that jointly encodes and fuses both users' semantic features. A DBC-Aware JSCC backbone embeds CSI into the encoder/decoder and a semantic fusion (SF) module combines the two users' features into a shared latent representation for broadcast, with a fusion ratio α balancing performance between users. A tailored loss L3 is proposed to maximize the semantic performance region, and experiments on CIFAR10, STL-10, and CelebA demonstrate that the SF scheme with DBC-Aware channel adaptation outperforms traditional TD/PA and JPEG2000+Capacity baselines, while incurring modest resource overhead. The approach shows robustness to channel estimation errors and varying SNR gaps, offering a practical pathway for efficient multiuser semantic broadcasting in wireless networks and potential extension to more users.

Abstract

Degraded broadcast channels (DBC) are a typical multiuser communication scenario, Semantic communications over DBC still lack in-depth research. In this paper, we design a semantic communications approach based on multi-user semantic fusion for wireless image transmission over DBC. In the proposed method, the transmitter extracts semantic features for two users separately. It then effectively fuses these semantic features for broadcasting by leveraging semantic similarity. Unlike traditional allocation of time, power, or bandwidth, the semantic fusion scheme can dynamically control the weight of the semantic features of the two users to balance the performance between the two users. Considering the different channel state information (CSI) of both users over DBC, a DBC-Aware method is developed that embeds the CSI of both users into the joint source-channel coding encoder and fusion module to adapt to the channel. Experimental results show that the proposed system outperforms the traditional broadcasting schemes.
Paper Structure (12 sections, 7 equations, 19 figures, 1 table)

This paper contains 12 sections, 7 equations, 19 figures, 1 table.

Table of Contents

  1. INTRODUCTION
  2. SYSTEM MODEL
  3. System Overview
  4. The Architecture of the DBC-Aware JSCC
  5. FUSION BASED BROADCASTING SCHEME FOR SEMANTIC COMMUNICATIONS
  6. Semantic Fusion Broadcasting Scheme
  7. Loss Function Design
  8. EXPERIMENTS RESULTS
  9. Experimental Setup
  10. Visualization Results
  11. Performance Analysis
  12. CONCLUSION and FUTURE WORK

Figures (19)

  • Figure 1: The overall architecture of the proposed fusion-based multi-user semantic communication system over DBC
  • Figure 2: The overall structure of the proposed DBC-Aware JSCC
  • Figure 3: The architecture of DBC-Aware Transformer
  • Figure 4: The $RV_2$-coefficient of semantic features from images in CelebA dataset
  • Figure 5: The Pearson correlation coefficient of semantic features from images in CelebA dataset
  • ...and 14 more figures