ID 原文 译文
3743 通过综合考虑异构处理器调度、计算资源分配以及带宽资源分配，将通信及计算资源受限下的系统效用最大化问题建模为混合整数非线性问题。 Under the limited computation and communication resources,a system-wide utility maximization problem by jointly considering heterogeneous processor scheduling, computation andbandwidth resource allocation was formulated as a mixed-integer nonlinear programming problem.
3744 为了快速求解该问题，将所提模型进一步解耦为业务驱动的异构处理器调度问题和资源联合分配问题，并提出了基于拉格朗日对偶理论的联合优化算法。 To solve the problem efficiently, the formulated problem was transformed into two sub-problems, namely application-driven heterogeneousprocessor scheduling and joint resource allocation, and a Lagrange-dual based algorithm was proposed.
3745 仿真结果表明，所提算法可以有效提升 BMEC 系统的系统效用。 Simulation results show that the proposed scheme can effectively improve the system-wide utility of the BMEC system.
3746 针对无人机辅助的边缘计算系统，提出了一种两阶段交替优化算法以最大化系统能量效率，该优化问题复杂且非凸。 For the unmanned aerial vehicle (UAV)-assisted edge computing system, a two-stage alternative algorithmwas proposed to solve the formulated complex non-convex problem.
3747 为此，首先利用 Dinkelbach 方法将建模的非线性分式规划问题转换为一类等价的参数寻优问题。 Firstly, the formulated non-linear fractional pro-gramming problem was reformulated to the equivalent parametric problem by using Dinkelbach method.
3748 其次，将其拆分成 2 个子问题进行交替优化。利用拉格朗日对偶法，给出了中央处理单元频率和数据比特量的闭式解。 Secondly,two sub-problems were further considered based on it. By employing the Lagrange duality method, the closed-formsolutions for the central processing unit frequencies and the number of data bits were derived.
3749 最后，所获得的解揭示了源节点选择卸载与共享自身数据和无人机中继选择转发计算结果的必要条件，以及实现更高能量效率的方法。 Finally, based on thesolutions obtained, the conditions that the source node prefers to offload/share its data and the relay chooses to for-ward the computation results, as well as the approaches to achieve high energy efficiency were revealed.
3750 仿真结果表明，与传统算法相比，所提算法在能量效率方面可获得最高近 20 倍的性能提升。 Numericalresults demonstrate that the proposed design can achieve a performance improvement of up to 20 times over the con-ventional schemes.
3751 为了提升多用户计算任务卸载时的系统效用，提出了一种基于边云联合计算的多用户任务卸载方案。 A computation offloading scheme based on edge-cloud computing was proposed to improve the system utilityof multiuser computation offloading.
3752 该方案在提升系统效用的同时，考虑了边云资源的协同优化问题。 This scheme improved the system utility while considering the optimization ofedge-cloud resources.