TokenCom-UEP: Semantic Importance-Matched Unequal Error Protection for Resilient Image Transmission

Token communication (TokenCom), an emerging semantic communication framework powered by Large Multimodal Model (LMM), has become a key paradigm for resilient data transmission in 6G networks. While the concept of leveraging non-uniform token importance for unequal error protection (UEP) in Token Communication (TokenCom) has been previously suggested, existing implementations predominantly rely on equal error protection (EEP), which is suboptimal. To address this, this letter proposes TokenCom-UEP, a novel semantic importance-matched UEP framework designed for resilient image transmission. TokenCom-UEP integrates rateless UEP coding with the non-uniform semantic importance of tokens by partitioning source tokens into nested expanding windows, assigning higher selection probabilities to windows containing critical tokens to ensure their prioritized recovery. Simulation results demonstrate that TokenCom-UEP outperforms EEP schemes in terms of three core semantic restoration metrics and achieves higher bandwidth utilization.