支持MEC的天地一体化网络下任务卸载和资源分配

TASK OFFLOADING AND RESOURCE ALLOCATION FOR THE MEC ENABLED INTEGRATED SATELLITE-TERRESTRIAL NETWORK

针对天地一体化网络(Integrated Satellite-Terrestrial Network,ISTN)高动态、时延敏感的特性,用户服务质量(Quality of Service,QoS)保障成为新的研究重点。为此,在天地一体化网络场景下使用移动边缘计算(Mobile Edge Computing,MEC)技术,在低轨道(LEO)卫星上部署MEC服务器,通过排队理论分析场景的时延成本,提出任务卸载和资源分配的联合优化方案。为最小化用户与LEO卫星的传输时延,基于凸优化理论给每个用户分配最优的带宽资源。基于整数规划在多项式时间内给出接近全局最优的任务卸载决策。仿真结果表明提出算法与低复杂度的启发式算法相比,减少了20%的总时延成本,验证了ISTN(Integrated Satellite-Terrestrial Network)下通过MEC技术保障地面用户QoS的可行性和有效性。

Aimed at the highly dynamic and delay-sensitive characteristics of the integrated satellite-terrestrial network(ISTN),the quality of service(QoS)guarantee of users become the new research focus.Therefore,mobile-edge computing(MEC)is used in the ISTN scenario and MEC server is deployed on the low-orbit(LEO)satellite.By analyzing the delay cost of the scenario through queuing theory,we propose a joint task offloading and resource allocation algorithm.In order to minimize the transmission delay between users and LEO satellite,each user was assigned the optimal bandwidth resource based on the convex optimization theory.Based on integer programming,an approximate global optimal task unloading decision was given in polynomial time.The simulation results show that the proposed algorithm reduces the total delay cost by 20%compared with the low-complexity heuristic algorithm.It also verifies the feasibility and effectiveness of using MEC technology to guarantee the QoS of ground users in ISTN.