AI 中文总结
针对干扰信道中FAS辅助的下行MEAN，该研究联合优化智能体位置、FAS端口选择与发射功率以最大化能效，提出的迭代算法在噪声严重时性能优于基准方案。
AI 中文摘要
本文研究干扰信道中流体天线系统（FAS）辅助的下行移动具身人工智能网络（MEAN），其中多对基站（BS）智能体重用相同频谱。基站采用FAS提升通信质量，而移动具身人工智能（AI）智能体可根据感知环境的信道信息（如信道-干扰加噪声图（CINM））调整自身位置。考虑同信道干扰以及通信和智能体移动带来的能耗，我们通过联合优化智能体位置、FAS端口选择和发射功率，构建能效（EE）最大化问题。为求解该混合整数非凸问题，我们先针对给定的智能体位置和FAS端口推导闭式最优发射功率，再开发一种迭代算法，该算法包含自适应FAS端口优化、序贯智能体位置优化，以及低复杂度功率更新方法。仿真结果表明，所提设计优于所考虑的基准方案，且在噪声严重的条件下具备更高的可行性。
英文摘要
In this paper, we investigate a fluid antenna system (FAS)-assisted downlink mobile embodied AI network (MEAN) over interference channels, where multiple base station (BS)-agent pairs reuse the same spectrum. The BSs employ FASs to improve the communication quality, while the mobile embodied artificial intelligence (AI) agents can adjust their positions according to environment-aware channel information, such as a channel-to-interference-plus-noise map (CINM). Considering both co-channel interference and the energy consumption caused by communication and agent movement, we formulate an energy efficiency (EE) maximization problem by jointly optimizing the agent positions, FAS port selections, and transmit powers. To solve this mixed-integer non-convex problem, we first derive the optimal transmit power in closed form for given agent positions and FAS ports. We then develop an iterative algorithm with adaptive FAS-port optimization and sequential agent-position optimization, together with a low-complexity power-update method. Simulation results demonstrate that the proposed design outperforms the considered benchmark schemes and provides improved feasibility under severe noise conditions.
