Both the radio resources in wireless networks and the computational resources in cloud have big impact on the performance of the mobile cloud computing system. In this paper, we study the joint radio and computational resource allocation in a mobile edge cloud system with a heterogeneous radio access network and a close-by edge cloud. The objective of the proposed resource allocation scheme is to maximize the system utility as well as satisfy the diverse quality requirements for the delay-sensitive and computation-intensive applications of mobile users. The requirements for economic cost reduction and energy conservation are considered in the proposed scheme to achieve the balance between the user-centric and network-centric resource allocation. The proposed scheme takes advantage of both ant colony optimization (ACO) and genetic algorithm (GA) to explore and exploit the search space to obtain the near optimal solution at the lower computational complexity. ACO is applied for generating the initial population, and GA operations such as mapping, crossover, and repair are proposed to improve the search ability and avoid premature convergence through the search of solution in a broader search space. Simulation results show that our proposed scheme outperforms the existing schemes in terms of convergence performance and the accuracy of final results. Moreover, the results demonstrate that it can not only achieve significant system utility improvement, but also achieve higher resource utilization as well as remarkably lower average latency.

无线网络中的无线电资源和云中的计算资源都对移动云计算系统的性能产生很大影响。在本文中，我们研究了具有异构无线电接入网络和邻近边缘云的移动边缘云系统中的联合无线电和计算资源分配。提出的资源分配方案的目的是最大化系统实用性，并满足移动用户对延迟敏感和计算密集型应用的各种质量要求。在提出的方案中考虑了降低经济成本和节约能源的要求，以实现以用户为中心和以网络为中心的资源分配之间的平衡。提出的方案利用蚁群优化（ACO）和遗传算法（GA）的优势来探索和利用搜索空间，以较低的计算复杂度获得接近最优的解决方案。应用ACO生成初始种群，并提出了遗传算法操作（如映射，交叉和修复），以提高搜索能力并通过在更宽的搜索空间中搜索解决方案来避免过早收敛。仿真结果表明，本文提出的方案在收敛性能和最终结果的准确性方面均优于现有方案。此外，结果表明，它不仅可以显着提高系统实用性，而且还可以提高资源利用率并显着降低平均延迟。