Distributed computing enables Internet of vehicle (IoV) services by collaboratively utilizing the computing resources from the network edge and the vehicles. However, the computing interruption issue caused by frequent edge network handoffs, and a severe shortage of computing resources are two problems in providing IoV services. High altitude platform station (HAPS) computing can be a promising addition to existing distributed computing frameworks because of its wide coverage and strong computational capabilities. In this regard, this paper proposes an adaptive scheme in a new distributed computing framework that involves HAPS computing to deal with the two problems of the IoV. Based on the diverse demands of vehicles, network dynamics, and the time-sensitivity of handoffs, the proposed scheme flexibly divides each task into three parts and assigns them to the vehicle, roadside units (RSUs), and a HAPS to perform synchronous computing. The scheme also constrains the computing of tasks at RSUs such that they are completed before handoffs to avoid the risk of computing interruptions. On this basis, we formulate a delay minimization problem that considers task-splitting ratio, transmit power, bandwidth allocation, and computing resource allocation. To solve the problem, variable replacement and successive convex approximation-based method are proposed. The simulation results show that this scheme not only avoids the negative effects caused by handoffs in a flexible manner, it also takes delay performance into account and maintains the delay stability.

翻译：分布式计算通过协同利用网络边缘和车辆的计算资源，实现车联网服务。然而，频繁的边缘网络切换导致的计算中断问题，以及计算资源的严重短缺，是提供车联网服务面临的两大挑战。高空平台站（HAPS）计算凭借其广覆盖性和强大计算能力，有望成为现有分布式计算框架的有益补充。为此，本文提出一种基于HAPS计算的新型分布式计算框架中的自适应方案，以应对车联网的上述两大问题。该方案根据车辆的多样化需求、网络动态性以及切换的时间敏感性，灵活地将每个任务划分为三部分，并分别分配给车辆、路侧单元（RSUs）和HAPS执行同步计算。同时，方案约束RSUs上的任务计算必须在切换前完成，以避免计算中断风险。在此基础上，我们构建了一个考虑任务拆分比例、传输功率、带宽分配和计算资源分配的时延最小化问题。为求解该问题，提出了变量替换和基于逐次凸逼近的方法。仿真结果表明，该方案不仅能够灵活避免切换带来的负面影响，还能兼顾时延性能并保持时延稳定性。