Mobile edge computing (MEC) is a promising paradigm, which brings computation resources in proximity to mobile devices and allows the tasks of mobile devices to be offloaded to MEC servers with low latency. The joint problem of cooperative computation task offloading and resource allocation is a challenging issue. The joint problem of cooperative computation task offloading scheme and resource assignment in MEC is investigated in this paper, where the vertical cooperation among mobile devices, mobile edge server nodes and mobile cloud server nodes is considered, and the horizontal computation cooperation between edge nodes is considered as well. A computation offloading decision, cooperative selection, power allocation and CPU cycle frequency assignment problem is formulated. The objective is to minimize the latency while guaranteeing the constraint of transmission power, energy consumption and CPU cycle frequency. The formulated latency optimization problem is a nonconvex mixed-integer problem in general, which has binary variables and continuous variables. In order to solve the formulated problem. A joint iterative algorithm based on the Lagrangian dual decomposition, ShengJin Formula method, and monotonic optimization method is proposed. The CPU cycle frequence allocation is handled by the ShengJin Formula method due to the cubic equation of one variable about the CPU frequence allocation. The transmission power assignment is handled by the monotonic optimization method. In the algorithm convergence with different number of tasks, the proposed algorithm can quickly and effectively reach the convergence state and getting the minimum task execution delay. Numerical results demonstrate that the proposed algorithm outperforms the Full MEC, Full Local and Full Cloud three schemes in terms of execution latency.

移动边缘计算（MEC）是一种很有前途的范例，它将计算资源带到移动设备附近，并允许以低延迟将移动设备的任务卸载到MEC服务器上。协作计算任务卸载和资源分配的共同问题是一个具有挑战性的问题。研究了MEC中协同计算任务分流方案与资源分配的联合问题，其中考虑了移动设备，移动边缘服务器节点与移动云服务器节点之间的垂直合作，并考虑了边缘节点之间的水平计算合作也一样 提出了计算分流决策，协同选择，功率分配和CPU周期频率分配问题。目的是在保证传输功率，能耗和CPU周期频率的约束的同时最小化等待时间。公式化的延迟优化问题通常是一个非凸混合整数问题，具有二进制变量和连续变量。为了解决制定的问题。提出了一种基于拉格朗日对偶分解，圣金公式方法和单调优化方法的联合迭代算法。由于有关CPU频率分配的一个变量的三次方程式，因此通过圣金公式方法处理了CPU周期频率分配。传输功率分配通过单调优化方法处理。在不同任务数量的算法收敛中，该算法可以快速有效地达到收敛状态，并获得最小的任务执行延迟。数值结果表明，本文提出的算法在执行时延方面优于Full MEC，Full Local和Full Cloud三个方案。