摘要
文章研究了基于多载波非正交多址接入(Multicarrier Non-Orthogonal Multiple Access, MC-NOMA)的协作移动边缘计算系统,该系统由一个用户端、一个协作节点与一个配备边缘计算服务器的接入点(Access Point, AP)组成,旨在给定任务执行时延约束下,最小化用户端与协作节点的通信与计算能耗。为此,首先提出一种新颖的MEC协议,使用户端和协作节点充分利用MC-NOMA技术以高频谱效率的方式进行任务卸载;然后在该协议下,建立系统优化问题,联合优化任务卸载时间与计算时间分配、子载波分配、各个子载波的发射功率分配、任务划分和MEC服务器CPU频率。该问题是一个非凸的混合整数规划问题,求解颇具挑战性。通过利用代数变换、大M表达法、分层优化、连续凸近似和黄金分割法,对该问题进行求解,从而提出一种高效的系统优化算法。仿真结果表明,与已有的仅采用多载波技术研究方案相比,采用MC-NOMA技术可节省50%~60%的能耗。
This paper studies a multicarrier non-orthogonal multiple access(MC-NOMA)-based mobile edge computing(MEC) system which consists of a user terminal, a cooperative node, and an access point equipped with an MEC server. The goal is to minimize the energy consumed for communication and computation at the user terminal and the cooperative node. To this end, a novel MEC protocol is first proposed, in which the user terminal and the cooperative node are enabled to offload their computation tasks with high spectral efficiency by employing the MC-NOMA technique. Then, under the proposed protocol, an optimization problem is formulated to jointly optimize the allocation of task offloading time and computation time, subcarrier allocation, power allocation over each subcarrier, task partition and the CPU frequency of the MEC server. The formulated problem is a highly non-convex and mixed integer programming problem, which is quite challenging to solve. By using algebraic transformation, big-M formulation, layered optimization, successive convex approximation and golden section techniques, the solution of the problem is achieved, and an efficient optimization algorithm is developed. Simulation results show that by employing the MC-NOMA technique, the proposed scheme can save 50 to 60 percent energy consumption for the MEC system as compared with the scheme in the literature where only the multicarrier technique is employed.
作者
罗至威
黄栩蔚
叶炜
黄高飞
LUO Zhi-wei;HUANG Xu-wei;YE Wei;HUANG Gao-fei(School of Electronics and Communication Engineering,Guangzhou University,Guangzhou 510006,China)
出处
《广州大学学报（自然科学版）》
CAS
2024年第1期84-96,共13页
Journal of Guangzhou University:Natural Science Edition
基金
国家自然科学基金资助项目(61902098,61872084)
大学生创新训练资助项目(202111078006)。
关键词
移动边缘计算
协作通信
非正交多址接入
多载波传输
凸优化
mobile edge computing
cooperative communication
non-orthogonal multiple access
multicarrier transmission
convex optimization
