Mobile edge computing (MEC) has been identified as a key technique of next-generation wireless networks, which supports cloud computing along with other compelling service capabilities at the network’s edge with the objective of reducing the system delay. As one of the prospective candidates for new spectrum in next-generation networks, millimeter wave (mmWave) communications has been gaining significant attention as a benefit of its high rate. Hence we conceive a joint hybrid beamforming and resource allocation algorithm for mmWave MEC. Explicitly, we jointly optimize the analog beamforming vectors at the users, the analog and digital beamforming matrices at the base station (BS), the computation task offloading ratios and resource allocation at the MEC server for minimizing the maximum system delay subject to the affordable communication and computing budget. We conceive a powerful algorithm for solving this challenging nonconvex optimization problem with coupled constraints based on the penalty dual decomposition (PDD) technique. The proposed algorithm can be implemented in a parallel and distributed fashion. Our numerical results demonstrate the superiority of the proposed algorithm by quantifying the benefits of intrinsically amalgamating MEC with mmWave communications.

中文翻译：

---

移动边缘计算遇到毫米波通信：联合波束成形和资源分配以最小化系统延迟

移动边缘计算（MEC）已被确定为下一代无线网络的关键技术，它支持云计算以及网络边缘的其他引人注目的服务能力，目的是减少系统延迟。作为下一代网络中新频谱的潜在候选者之一，毫米波 (mmWave) 通信因其高速率的优势而备受关注。因此，我们设想了一种用于毫米波 MEC 的联合混合波束成形和资源分配算法。明确地，我们联合优化了用户处的模拟波束成形向量、基站 (BS) 的模拟和数字波束成形矩阵，MEC 服务器上的计算任务卸载率和资源分配，以在可负担的通信和计算预算下最小化最大系统延迟。我们构思了一种强大的算法来解决这个具有挑战性的非凸优化问题，该问题具有基于惩罚对偶分解 (PDD) 技术的耦合约束。所提出的算法可以以并行和分布式方式实现。我们的数值结果通过量化将 MEC 与毫米波通信本质上合并的好处，证明了所提出算法的优越性。所提出的算法可以以并行和分布式方式实现。我们的数值结果通过量化将 MEC 与毫米波通信本质上合并的好处，证明了所提出算法的优越性。所提出的算法可以以并行和分布式方式实现。我们的数值结果通过量化将 MEC 与毫米波通信本质上合并的好处，证明了所提出算法的优越性。