卫星边缘计算突破地理限制,实现全球无缝覆盖,赋予偏远地区服务计算能力,针对低轨道(Low Earth Orbit,LEO)卫星边缘计算场景计算能力和通信时间有限的特性,在单LEO卫星边缘节点多地面用户的系统模型下,提出一种基于博弈论计算卸载联合...卫星边缘计算突破地理限制,实现全球无缝覆盖,赋予偏远地区服务计算能力,针对低轨道(Low Earth Orbit,LEO)卫星边缘计算场景计算能力和通信时间有限的特性,在单LEO卫星边缘节点多地面用户的系统模型下,提出一种基于博弈论计算卸载联合资源分配(Game Theory with Computing Offloading and Resource Allocation,GT⁃CORA)的策略。由于该问题表示为一个混合整数非线性规划问题,将该问题拆解为资源分配和计算卸载两个子问题,通过二分法和拉格朗日乘数法实现最优资源分配,通过博弈论和势博弈证明了纳什均衡的存在性并解决计算卸载问题。仿真结果表明了该策略的有效性,该策略相较全部本地卸载、全部卫星卸载以及贪婪卸载策略,在保持较高的任务成功率下,平均计算开销明显降低,可以满足地面用户任务需求。展开更多
Mobile edge computing (MEC) is a novel technique that can reduce mobiles' com- putational burden by tasks offioading, which emerges as a promising paradigm to provide computing capabilities in close proximity to mo...Mobile edge computing (MEC) is a novel technique that can reduce mobiles' com- putational burden by tasks offioading, which emerges as a promising paradigm to provide computing capabilities in close proximity to mobile users. In this paper, we will study the scenario where multiple mobiles upload tasks to a MEC server in a sing cell, and allocating the limited server resources and wireless chan- nels between mobiles becomes a challenge. We formulate the optimization problem for the energy saved on mobiles with the tasks being dividable, and utilize a greedy choice to solve the problem. A Select Maximum Saved Energy First (SMSEF) algorithm is proposed to realize the solving process. We examined the saved energy at different number of nodes and channels, and the results show that the proposed scheme can effectively help mobiles to save energy in the MEC system.展开更多
文摘卫星边缘计算突破地理限制,实现全球无缝覆盖,赋予偏远地区服务计算能力,针对低轨道(Low Earth Orbit,LEO)卫星边缘计算场景计算能力和通信时间有限的特性,在单LEO卫星边缘节点多地面用户的系统模型下,提出一种基于博弈论计算卸载联合资源分配(Game Theory with Computing Offloading and Resource Allocation,GT⁃CORA)的策略。由于该问题表示为一个混合整数非线性规划问题,将该问题拆解为资源分配和计算卸载两个子问题,通过二分法和拉格朗日乘数法实现最优资源分配,通过博弈论和势博弈证明了纳什均衡的存在性并解决计算卸载问题。仿真结果表明了该策略的有效性,该策略相较全部本地卸载、全部卫星卸载以及贪婪卸载策略,在保持较高的任务成功率下,平均计算开销明显降低,可以满足地面用户任务需求。
基金supported by NSFC(No. 61571055)fund of SKL of MMW (No. K201815)Important National Science & Technology Specific Projects(2017ZX03001028)
文摘Mobile edge computing (MEC) is a novel technique that can reduce mobiles' com- putational burden by tasks offioading, which emerges as a promising paradigm to provide computing capabilities in close proximity to mobile users. In this paper, we will study the scenario where multiple mobiles upload tasks to a MEC server in a sing cell, and allocating the limited server resources and wireless chan- nels between mobiles becomes a challenge. We formulate the optimization problem for the energy saved on mobiles with the tasks being dividable, and utilize a greedy choice to solve the problem. A Select Maximum Saved Energy First (SMSEF) algorithm is proposed to realize the solving process. We examined the saved energy at different number of nodes and channels, and the results show that the proposed scheme can effectively help mobiles to save energy in the MEC system.