The energy-efficient task offloading problem of a massive multiple-input multiple-output (MIMO)-aided fog computing system is solved, where multiple task nodes offload their computational tasks to be solved via a massive MIMO-aided fog access node to multiple processing nodes in the fog for execution. By considering realistic imperfect channel state information (CSI), we formulate a joint task offloading and power allocation problem for minimizing the total energy consumption, including both computation and communication power consumptions. We solve the resultant non-convex optimization problem in two steps. First, we solve the computational task allocation and computational resource allocation for a given power allocation. Then, we conceive a sequential optimization framework for determining the specific power allocation decision that minimizes the total energy consumption of the fog access node. Given the computational tasks, the computational resources, and the power allocation, we propose an iterative algorithm for the system optimization. The simulation results show that the proposed scheme significantly reduces the total energy consumption compared to the benchmark schemes.

中文翻译：

---

大规模MIMO辅助多对雾计算网络中的节能任务卸载

解决了大规模多输入多输出（MIMO）辅助雾计算系统的高能效任务卸载问题，其中多个任务节点通过大规模MIMO辅助雾访问节点将要解决的计算任务卸载到多个处理中雾中执行的节点。通过考虑现实的不完美信道状态信息（CSI），我们制定了一个联合任务卸载和功率分配问题，以最大程度地减少包括计算和通信功耗在内的总能耗。我们分两步解决了由此产生的非凸优化问题。首先，我们解决给定功率分配的计算任务分配和计算资源分配。然后，我们设想了一个顺序优化框架，用于确定特定的功率分配决策，以最大程度地减少雾访问节点的总能耗。给定计算任务，计算资源和功率分配，我们提出了一种用于系统优化的迭代算法。仿真结果表明，与基准方案相比，该方案显着降低了总能耗。