arXiv：cs.LG· Haowen Wan, Qianqian Yang·· 7 小时前AI 评分30
基于 MoE 机制的自适应语义通信无线图像传输
Adaptive Semantic Communication for Wireless Image Transmission Leveraging Mixture-of-Experts Mechanism
AI 导读
研究者提出一种面向 MIMO 信道的多阶段端到端图像语义通信系统，核心是基于自适应 MoE Swin Transformer 块构建的动态专家门控机制。该机制同时评估实时 CSI 与输入图像 patch 的语义内容，联合计算自适应路由概率，仅激活部分专家，从而突破单驱动路由的瓶颈。仿真结果显示，该方法在保持传输效率的同时显著提升了图像重建质量。
正文
Abstract:Deep learning based semantic communication has achieved significant progress in wireless image transmission, but most existing schemes rely on fixed models and thus lack robustness to diverse image contents and dynamic channel conditions. To improve adaptability, recent studies have developed adaptive semantic communication strategies that adjust transmission or model behavior according to either source content or channel state. More recently, MoE-based semantic communication has emerged as a sparse and efficient adaptive architecture, although existing designs still mainly rely on single-driven routing. To address this limitation, we propose a novel multi-stage end-to-end image semantic communication system for multi-input multi-output (MIMO) channels, built upon an adaptive MoE Swin Transformer block. Specifically, we introduce a dynamic expert gating mechanism that jointly evaluates both real-time CSI and the semantic content of input image patches to compute adaptive routing probabilities. By selectively activating only a specialized subset of experts based on this joint condition, our approach breaks the rigid coupling of traditional adaptive methods and overcomes the bottlenecks of single-driven routing. Simulation results indicate a significant improvement in reconstruction quality over existing methods while maintaining the transmission efficiency.
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2604.02691 [cs.LG] |
| (or arXiv:2604.02691v2 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2604.02691 arXiv-issued DOI via DataCite |
Submission history
From: Haowen Wan [view email]
[v1]
Fri, 3 Apr 2026 03:35:09 UTC (10,890 KB)
[v2]
Tue, 6 Oct 2026 13:16:55 UTC (9,181 KB)
来源：arXiv：cs.LG · arxiv.org