Energy-efficient workload offloading and power control in vehicular edge computing

In this paper, an energy-efficient vehicular edge computing (VEC) framework is proposed for in-vehicle user equipments (UEs) with limited battery capacity. Firstly, the energy consumption minimization problem is formulated as a joint workload offloading and power control problem, with the explicit consideration of energy consumption and delay models. Queuing theory is applied to derive the stochastic traffic models at UEs and VEC nodes. Then, the original NP-hard problem is transformed to a convex global consensus problem, which can be decomposed into several parallel subproblems and solved subsequently. Next, an alternating direction method of multipliers (ADMM)-based energy-efficient resource allocation algorithm is developed, whose outer loop representing iterations of nonlinear fractional programming, while inner loop representing iterations of primal and dual variable updates. Finally, the relationships between energy consumption and key parameters such as workload offloading portion and transmission power are validated through numerical results.