Cost-Aware Computation Offloading and Resource Allocation in Ultra-Dense Multi-Cell, Multi-User and Multi-Task MEC Networks

When many users offload computation tasks to edge servers in mobile edge computing (MEC) networks, how to economically utilize edge nodes with limited computation resources to ensure users' quality of experiences (QoEs) is a vital issue, especially for an ultra-dense multi-cell, multi-user, and multi-task framework. To minimize the task processing cost of all users, we try jointly optimizing computational offloading decisions, task offloading proportion, and allocation of communication and computation resources in this paper. Herein, the task processing cost is mainly caused by local energy consumption, wireless communication, and edge computing. Although the proposed optimization problem is non-convex, we design an efficient algorithm to find its optimum solution. Specifically, by taking advantage of both the improved artificial fish swarm algorithm (IAFSA) and the improved particle swarm optimization (IPSO), a hierarchical algorithm used for computation offloading (HACO) is designed. Then, the convergence, complexity, and parallel implementation of such an algorithm are analyzed. Finally, compared with other existing algorithms through simulation experiments, it is verified that the designed algorithm may achieve better performance in reducing the task processing cost under strict delay constraints.