标题： 融合陆地与非陆地网络的可持续6G运营：一种时延感知的多层小区切换方法 显示英文标题

标题： Integrating Terrestrial and Non-Terrestrial Networks for Sustainable 6G Operations: A Latency-Aware Multi-Tier Cell-Switching Approach

摘要： 可持续性在现代蜂窝网络中至关重要，这些网络面临着移动流量增加和无线技术进步带来的显著能耗挑战。 单元切换在文献中已被确立为一种有效解决方案，但在通过陆地网络（TN）实施时，会遇到容量不足和覆盖范围有限等限制。 本研究通过将非陆地网络（NTN）集成到TN中，从而增强了单元切换，这些非陆地网络包括卫星（首次用于单元切换）、高空平台站（HAPS）和无人飞行器（UAV）。 这种集成通过扩展容量、增强覆盖范围和提高操作灵活性，显著提高了节能效果。 我们引入了一种多层单元切换方法，通过在网络层之间动态卸载用户来有效管理能源并最小化延迟，采用上下文感知策略满足多样化的用户需求。 此外，我们探讨了人工智能（AI），特别是生成式AI，在通过数据压缩、不同网络层之间的切换优化以及提高设备兼容性来优化网络效率中的作用，进一步提高了单元切换操作的适应性和能效。 案例研究证实了网络功耗和用户满意度的显著提升，展示了我们方法在未来网络中的潜力。 显示英文摘要

摘要： Sustainability is paramount in modern cellular networks, which face significant energy consumption challenges from rising mobile traffic and advancements in wireless technology. Cell-switching, well-established in literature as an effective solution, encounters limitations such as inadequate capacity and limited coverage when implemented through terrestrial networks (TN). This study enhances cell-switching by integrating non-terrestrial networks (NTN), including satellites (used for cell-switching for the first time), high altitude platform stations (HAPS), and uncrewed aerial vehicles (UAVs) into TN. This integration significantly boosts energy savings by expanding capacity, enhancing coverage, and increasing operational flexibility. We introduce a multi-tier cell-switching approach that dynamically offloads users across network layers to manage energy effectively and minimize delays, accommodating diverse user demands with a context aware strategy. Additionally, we explore the role of artificial intelligence (AI), particularly generative AI, in optimizing network efficiency through data compression, handover optimization between different network layers, and enhancing device compatibility, further improving the adaptability and energy efficiency of cell-switching operations. A case study confirms substantial improvements in network power consumption and user satisfaction, demonstrating the potential of our approach for future networks.