NOMA-MEC系统中基于改进遗传算法的协作式计算卸载与资源管理

为平衡网络负载与充分利用网络资源，针对超密集异构的多用户和多任务边缘计算网络，在用户时延约束下，该文构造了协作式计算任务卸载与无线资源管理的联合优化问题以最小化系统能耗。问题建模时，为应对基站超密集部署导致的严重干扰问题，该文采用了频带划分机制，并引入了非正交多址技术(NOMA)以提升上行频谱利用率。鉴于该目标优化问题具备非线性混合整数的形式，根据多样性引导变异的自适应遗传算法(AGADGM)，设计出了协作式计算卸载与资源分配算法。仿真结果表明，在严格满足时延约束条件下，该算法能获取较其他算法更低的系统能耗。

Cooperative Computation Offloading and Resource Management Based on Improved Genetic Algorithm in NOMA-MEC Systems

To balance the network loads and utilize fully the network resources, joint cooperative computation offloading and wireless resource management is considered for ultra-dense heterogeneous edge computing networks with multiple users and multiple tasks, which minimizes the system energy consumption under the constraints of users’ delay. During the problem modeling, a frequency spectrum partitioning mechanism is introduced to tackle serious network interference caused by ultra-dense deployment of base stations, and Non-Orthogonal Multiple Access (NOMA) technology is introduced to improve the uplink frequency spectrum efficiency. Considering that the optimization problem is a nonlinear mixed-integer form, according to Adaptive Genetic Algorithm with Diversity-Guided Mutation (AGADGM), an effective algorithm used for cooperative computation offloading and resource allocation is designed. The simulation results show that proposed algorithm could achieve lower system energy consumption than other existing algorithms under strict constraints of users’ delay.