📄 中文摘要
令牌通信（TokenCom）作为一种新兴的有效通信范式，利用令牌作为多模态通信和计算的统一单元，能够在未来的无线网络中实现高效的数字语义和目标导向通信。为了建立共享的语义潜在空间，令牌通信中的发送方和接收方需要就相同的令牌器模型和代码本达成一致。为此，在每个通信回合中，发送方和接收方需合作从一组预训练的令牌器模型和代码本中选择，以实现高效的令牌通信。该研究集中于多用户下行无线令牌通信场景中的令牌器协议（TA）过程，探讨了如何在多用户环境中优化令牌器的选择和协作机制。
无线令牌通信：基于强化学习的多用户无线令牌器协议
📄 English Summary
Wireless TokenCom: RL-Based Tokenizer Agreement for Multi-User Wireless Token Communications
Token Communications (TokenCom) has emerged as an effective paradigm where tokens serve as unified units for multimodal communications and computations, facilitating efficient digital semantic and goal-oriented communications in future wireless networks. To establish a shared semantic latent space, transmitters and receivers in TokenCom must agree on an identical tokenizer model and codebook. An initial Tokenizer Agreement (TA) process is conducted in each communication episode, where transmitters and receivers collaborate to select from a set of pre-trained tokenizer models and codebooks available to them for efficient TokenCom. This study focuses on the Tokenizer Agreement process in a multi-user downlink wireless TokenCom scenario, investigating how to optimize the selection and collaboration mechanisms of tokenizers in a multi-user environment.