Energy-Efficient Task Offloading Under E2E Latency Constraints

In this paper, we propose a novel resource management scheme that jointly allocates the transmitpower and computational resources in a centralized radio access network architecture. The networkcomprises a set of computing nodes to which the requested tasks of different users are offloaded. Theoptimization problem minimizes the energy consumption of task offloading while takes the end-to-end latency, i.e., the transmission, execution, and propagation latencies of each task, into account. We aim toallocate the transmit power and computational resources such that the maximum acceptable latency ofeach task is satisfied. Since the optimization problem is non-convex, we divide it into two sub-problems,one for transmit power allocation and another for task placement and computational resource allocation.Transmit power is allocated via the convex-concave procedure. In addition, a heuristic algorithm isproposed to jointly manage computational resources and task placement. We also propose a feasibilityanalysis that finds a feasible subset of tasks. Furthermore, a disjoint method that separately allocatesthe transmit power and the computational resources is proposed as the baseline of comparison. A lowerbound on the optimal solution of the optimization problem is also derived based on exhaustive searchover task placement decisions and utilizing Karush–Kuhn–Tucker conditions. Simulation results showthat the joint method outperforms the disjoint method in terms of acceptance ratio. Simulations alsoshow that the optimality gap of the joint method is less than 5%.本文提出一种新颖的资源管理方案，可在集中式无线接入网（Centralized Radio Access Network）架构下联合分配发射功率与计算资源。该网络包含一组计算节点，不同用户的请求任务均将卸载至这些节点执行。所构建的优化问题以最小化任务卸载能耗为目标，同时需兼顾各项任务的端到端时延——即传输时延、执行时延与传播时延。我们的目标是合理分配发射功率与计算资源，确保满足各项任务的最大可接受时延约束。由于该优化问题属于非凸问题，我们将其拆解为两个子问题：其一为发射功率分配子问题，其二为任务放置与计算资源分配子问题。发射功率分配可通过凹凸过程（Convex-Concave Procedure, CCP）求解。此外，本文还提出一种启发式算法，用于联合管理计算资源与任务放置策略。本文同时提出一种可行性分析方法，可筛选出可行的任务子集。此外，本文还提出一种分离式分配方法，即单独分配发射功率与计算资源，作为对比基准方案。本文还基于任务放置决策的穷举搜索与卡罗需-库恩-塔克（Karush-Kuhn-Tucker, KKT）条件，推导得到该优化问题最优解的下界。仿真结果表明，联合分配方法在任务接受率方面优于分离式分配方法；同时仿真结果显示，联合分配方法的最优性差距小于5%。