The article investigates aerial computing terrestrial-support-free multi-layer multi-access edge computing (AC-TMMEC) systems within the context of massive machine-type communications (mMTC) beyond 5G (B5G). With its four layers, namely space platform (S), high-altitude platform (HAP), low-altitude platform (LAP), and ground mobile access point platform (G), task offloading of ground users takes place in a sparsely distributed and dynamic environment mainly due to the high mobility of the AC-TMMEC entities. The article begins by reviewing the state-of-the-art, then identifies data routing and trajectory planning as new challenges in the design of AC-TMMEC offloading schemes. A simple yet effective task offloading scheme under AC-TMMEC, called iterative maximum flow algorithm (IMFA), is introduced, aiming to jointly solve the problems of source-destination pairing and data routing in a single shot via iterative maximum flow search. Extensive case studies are conducted to gain insights on the offloading latency performance of the proposed IMFA and compare it with a couple of counterparts with respect to two key parameters, namely the number of task requests and data size of each request.

中文翻译：

---

无地面支持的多层多接入边缘计算中的任务卸载


本文研究了 5G (B5G) 之外的大规模机器类型通信 (mMTC) 背景下的空中计算无地面支持的多层多接入边缘计算 (AC-TMMEC) 系统。通过空间平台（S）、高空平台（HAP）、低空平台（LAP）和地面移动接入点平台（G）四层，地面用户的任务卸载发生在稀疏分布的空间中。动态环境主要是由于AC-TMMEC实体的高移动性。本文首先回顾了最先进的技术，然后将数据路由和轨迹规划确定为 AC-TMMEC 卸载方案设计中的新挑战。提出了一种 AC-TMMEC 下简单而有效的任务卸载方案，称为迭代最大流算法（IMFA），旨在通过迭代最大流搜索联合解决单次源-目的地配对和数据路由问题。我们进行了广泛的案例研究，以深入了解所提出的 IMFA 的卸载延迟性能，并将其与几个对应项的两个关键参数（即任务请求数量和每个请求的数据大小）进行比较。