Computation Offloading in MIMO Based Mobile Edge Computing Systems under Perfect and Imperfect CSI Estimation
Computation Offloading in MIMO Based Mobile Edge Computing Systems under Perfect and Imperfect CSI Estimation
复制标题
DOI：
10.1109/icc.2018.8422274
复制
发表时间：
2018-05
期刊：
影响因子：
--
通讯作者：
N. Ti;L. Le
中科院分区：
文献类型：
--
作者：
N. Ti;L. Le
将计算密集型任务智能卸载到移动的边缘计算服务器提供了扩展移动的设备的可用性并延长其电池寿命的有效手段。然而，在当今的多输入多输出（MIMO）无线系统中实现该技术需要联合计算卸载和其他通信功能（诸如信道状态信息（CSI）估计和资源分配）的复杂设计。本文研究了MIMO无线系统中考虑理想和非理想CSI估计的计算任务卸载和资源分配优化问题。我们的设计旨在最大限度地减少最大加权能耗（最小最大W.C.E）的计算和无线电资源和服务延迟的实际约束。提出了求解混合整数非线性问题（MINLP）的最优和次优算法。特别是，二分法搜索和凸（DC）优化方法的差异，以确定全球和次优的解决方案，分别为完美和不完美的CSI场景。数值结果证实了所提出的设计的优势，在处理计算量大的任务比传统的本地计算策略。
Intelligent offloading of computation-intensive tasks to a mobile edge computing server provides an effective mean to expand the usability of mobile devices and prolong their battery life. However, realization of this technology in today's multiple input multiple output (MIMO) wireless systems requires the sophisticated design of joint computation offloading and other communications functions such as channel state information (CSI) estimation and resource allocation. In this paper, we study the optimization of computation task offloading and resource allocation in MIMO wireless systems considering perfect and imperfect CSI estimation. Our design aims to minimize the maximum weighted energy consumption (Min-max W.C.E) subject to practical constraints on computing and radio resources and service latency. The optimal and sub-optimal algorithms are proposed to solve the underlying mixed integer non-linear problem (MINLP). In particular, the bisection search and difference of convex (DC) optimization methods are employed to determine the global and sub-optimal solutions for the perfect and imperfect CSI scenarios, respectively. Numerical results confirm the advantages of the proposed design over the conventional local computation strategy in handling computationally heavy tasks.