In the mobile edge computing (MEC) networks, the heterogeneous demands can be met through device-to-device (D2D) technology due to the idle local resources from the neighbors. Although the overload of mobile edge computing servers can be relieved by sharing idle local resources among neighbors, the active impact of the cooperation in the close neighbors is ignored in the most existing works. In this paper, we study the mutual cooperation between the users with heterogeneous demands such as the computation tasks or the desired contents, and then propose an innovative D2D-assisted heterogeneous users cooperation framework where two users with different demands can share their idle local resources. The optimization is to minimize the aggregate delay of all users in MEC networks. First, the problem is formulated as a heterogeneous demands based matching game involving D2D matching and edge matching. Next, the multi-round cooperation matching (MRCM) algorithm is proposed to find the stable match of the proposed game. Finally, the simulation results present the validity and effectiveness of our proposed game as well as the proposed algorithm.

中文翻译：

---

针对异构需求的协作任务卸载和内容交付：一种匹配博弈论方法

在移动边缘计算（MEC）网络中，由于邻居的本地资源空闲，异构需求可以通过设备到设备（D2D）技术来满足。虽然移动边缘计算服务器的过载可以通过在邻居之间共享空闲的本地资源来缓解，但在大多数现有工作中都忽略了近邻合作的积极影响。在本文中，我们研究了具有异构需求（例如计算任务或所需内容）的用户之间的相互协作，然后提出了一种创新的 D2D 辅助异构用户协作框架，其中两个具有不同需求的用户可以共享其空闲的本地资源。优化是最小化 MEC 网络中所有用户的总延迟。第一的，该问题被表述为一个基于异构需求的匹配游戏，涉及 D2D 匹配和边缘匹配。接下来，提出了多轮合作匹配（MRCM）算法来找到所提出游戏的稳定匹配。最后，仿真结果展示了我们提出的游戏以及提出的算法的有效性和有效性。