User Energy Efficiency Fairness Algorithms for Task Offload in Cloud Edge Networks

Abstract

With the emergence of several new services such as driverless vehicles and virtual reality, mobile communication networks face problems such as heavy load and insufficient computing resources. The development of cloud, edge, and mobile edge network computing provides a good solution to this problem. This paper proposes the development of a user energy efficiency fairness task unloading algorithm for cloud-side networks. First, a cloud-side network cooperation model is constructed. The model ensures the efficient use of user energy and addresses the task offloading decision and resource allocation optimization problem jointly. Using the generalized fraction theory, the optimization problem is transformed into an equivalent convex problem by introducing relaxation as well as auxiliary variables. Next, the centralized energy efficiency fairness (CEEF) and alternating direction method of multiplier (ADMM)-based energy efficiency fairness algorithms are implemented to obtain an optimal solution for the optimization problem. Finally, through experimental simulation, the convergence of the CEEF- and ADMM-based energy efficiency fairness algorithms is verified. Compared with noncooperative algorithms, the performance of our proposed method increased by 30.76%. The proposed algorithm has been verified to ensure the fairness of user energy efficiency.