Caching Placement Minimised Service Time for Video Streaming over Roadside Units in VANETs

Abstract

Video streaming services (VSSs) are considerably requested by on-the-road users (RUs) in vehicular ad-hoc networks (VANETs). Gaining high quality of experience (QoE) of VSSs is more challenging due to dynamic characteristics of the RUs and complicated parameters of the videos. In this paper, we exploit the available storage and processing capacity of roadside units (RSUs) to improve the QoE of VSSs in VANETs. To do so, a caching placement (CAP) optimisation problem is formulated for optimal caching radius (measured in hops) of each video. The optimal solution is found in accordance with the patterns of not only the arrival and service at the RSUs but also the access rate of the videos, so as to minimise the service time for video streaming over RSUs in VANETs. Simulation results are presented to demonstrate the advantages of the CAP solution compared to other conventional schemes.