CL-ADMM: A Cooperative-Learning-Based Optimization Framework for Resource Management in MEC

We consider the problem of intelligent and efficient resource management framework in mobile-edge computing (MEC), which can reduce delay energy consumption, features distributed optimization congestion avoidance. In this article, we present a cooperative learning for MEC from an alternating direction method multipliers (ADMMs) perspective, named CL-ADMM framework. First, task requires both user personal data corresponding program that processes it, to efficiently cache group, novel popularity estimation scheme is proposed, based on semi-Markov process model. Then, greedy caching mechanism established, effectively consumption. Second, address group congestion, dynamic migration improved Q-learning alleviate congestion. Third, minimize consumption allocation formulate it as with large number variables, then exploit ADMM-based solve problem, complexity new set auxiliary these subproblems are all convex problems be solved by using primal-dual approach, guarantees its convergence. Finally, prove convergence Lyapunov theory. The numerical results demonstrate effectiveness reducing MEC.