DRL-based Resource Allocation Optimization for Computation Offloading in Mobile Edge Computing

Abstract

Mobile edge computing (MEC) provides a new development direction for emerging computing-intensive applications because it can improve computing performance and lower the threshold for users to use such applications. However, designing an effective computation offloading strategy to determine which tasks should be uninstalled to an edge server is still a crucial challenge. To this end, we propose a computation offload scheme based on dynamic resource allocation to optimize computing performance and energy consumption in MEC systems. We further formulate the resource allocation as a partially observable Markov decision process, which is solved by a policy gradient deep reinforcement learning method. Compared with other existing solutions, simulation results show that our proposal reduces the computational latency and energy consumption.