Edge and Central Cloud Computing: A Perfect Pairing for High Energy Efficiency and Low-Latency
Edge and Central Cloud Computing: A Perfect Pairing for High Energy Efficiency and Low-Latency
复制标题
DOI：
10.1109/twc.2019.2950632
复制
发表时间：
2020-02-01
影响因子：
10.4
通讯作者：
Zheng, Zhongbin
中科院分区：
文献类型：
--
作者：
Hu, Xiaoyan;Wang, Lifeng;Zheng, Zhongbin
在本文中，我们研究了边缘和中心云计算在异构蜂窝网络（HetNet），其中包含一个多天线宏基站（MBS），多个多天线小基站（SBS）和多个单天线用户设备（UE）之间的共存和协同。SBS由为UE提供有限计算服务的边缘云授权，而MBS经由到其相关联的SBS的受限多输入多输出（MIMO）回程向UE提供高性能中央云计算服务。在中心和边缘网络的处理延迟约束下，我们的目标是最小化用于任务卸载和计算的系统能耗。该问题是通过联合优化云选择、UE的发射功率、SBS的接收波束形成器和SBS的发射协方差矩阵来制定的，这是一个混合整数和非凸优化问题。基于分解方法和逐次伪凸方法，通过迭代算法得到了一个易于处理的解。仿真结果表明，我们提出的解决方案可以实现很大的性能增益比传统的方案，单独使用边缘或中央云。此外，在MBS处具有大规模天线的情况下，大规模MIMO回程可以显著降低所提出的算法的复杂度并且获得甚至更好的性能。
In this paper, we study the coexistence and synergy between edge and central cloud computing in a heterogeneous cellular network (HetNet), which contains a multi-antenna macro base station (MBS), multiple multi-antenna small base stations (SBSs) and multiple single-antenna user equipment (UEs). The SBSs are empowered by edge clouds offering limited computing services for UEs, whereas the MBS provides high-performance central cloud computing services to UEs via a restricted multiple-input multiple-output (MIMO) backhaul to their associated SBSs. With processing latency constraints at the central and the edge networks, we aim to minimize the system energy consumption used for task offloading and computation. The problem is formulated by jointly optimizing the cloud selection, the UEs' transmit powers, the SBSs' receive beamformers, and the SBSs' transmit covariance matrices, which is a mixed-integer and non-convex optimization problem. Based on the methods such as decomposition approach and successive pseudoconvex approach, a tractable solution is proposed via an iterative algorithm. The simulation results show that our proposed solution can achieve great performance gain over conventional schemes using edge or central cloud alone. Also, with large-scale antennas at the MBS, the massive MIMO backhaul can significantly reduce the complexity of the proposed algorithm and obtain even better performance.