Quantified Edge Server Placement With Quantum Encoding in Internet of Vehicles
Jie Zhang, Jiawei Lu, Xuan Yan, Xiaolong Xu, Lianyong Qi, Wanchun Dou
Abstract
In the edge computing-supported Internet of Vehicles (IoV), the edge servers (ESs) are distributed near the road side units (RSUs) to process the transmitted data for various IoV services in real time. Generally, due to the budget constraints, the ES scale is limited. If the ESs are not appropriately placed, it may cause the unbalanced load distribution of the ESs. Therefore, how to place a constant number of ESs with the goal of avoiding the risk of overload and improving the quality of services remains a challenge. To tackle this challenge, a quantified edge server placement strategy, named QESP, is fully investigated, to improve the coverage rate, the workload balance and reduce the average waiting time of services in the IoV. Technically, binary encoding and quantum encoding are fully leveraged to model the ES locations. Then, the niched pareto genetic algorithm II (NPGA-II) and the quantum rotation gate are adopted to obtain the appropriate solutions for the ES placement problem. Furthermore, an assessment function is devised to find the optimal scheme among the solutions obtained. Finally, the validity of QESP is evaluated by using real-world vehicular big data.