Efficient Computation Offloading in Mobile Edge Computing Based on Dynamic Programming

Computation offloading in mobile edge computing (MEC) systems is hopeful to save battery energy, improve service quality and enhance the user experience. This paper studies how to offload tasks in a multi-user MEC system more efficiently. The offloading decision problem is described as a NP-hard problem whose objective minimizes the users’ energy cost under the constraints of edge cloud capacity and latency tolerance. By discretizing the decision variable, the original problem is reduced to a two-dimensional knapsack problem. Finally, a computation offloading based on dynamic programming (COBDP) algorithm is developed to find the optimal strategy. According to the simulation results, the COBDP algorithm can achieve better performance in terms of users’ energy cost.