Discontinuous Computation Offloading for Energy-Efficient Mobile Edge Computing

We propose a novel strategy for energy-efficient dynamic computation offloading, in the context of edge-computing-aided beyond 5G networks. The goal is to minimize energy consumption overall system, comprising multiple User Equipment (UE), an access point (AP), and edge server (ES), under constraints on end-to-end service delay packet error rate performance over wireless interface. To reduce consumption, we exploit low-power sleep operation modes users, AP ES, shifting computing paradigm from always on available architecture, capable guaranteeing on-demand target quality with minimum consumption. this aim, online algorithm optimal orchestration radio computational resources called xmlns:xlink="http://www.w3.org/1999/xlink">Discontinuous Computation Offloading (DisCO) . In such framework, translate into queueing delays, including both communication phases offloading service. DisCO hinges Lyapunov stochastic optimization, does not require any prior knowledge statistics traffic or channels, satisfies long-term imposed by users. Several numerical results illustrate advantages proposed method.