Task Processing Energy Minimization Based on Hybrid Mode in D2D-Enabled Fog Network

The low processing power and computation capability of users' devices have become a significant challenge to perform emerging tasks. This paper studies the energy consumption problem of users in the D2D-enabled Fog network. We assume users can decide to perform the tasks via local mode, D2D mode, Fog mode, or use these modes at the same time as a hybrid mode. The optimization problem is formulated as an energy minimization problem by considering resource allocation, offloading decision, and task partitioning in a partial task offloading scenario. Since the optimization problem is NP-hard, it is decomposition into the resource allocation, offloading decision, and task-partitioning sub-problems, respectively. To solve this, we propose Lagrange multipliers and bisection methods with task partitioning and offloading decision strategies. The simulation results illustrate the performance of our proposed algorithm.