Multiple Precaching Vehicle Selection Scheme Based on Set Ranking in Intermittently Connected Vehicular Networks

Area of Science:

• Vehicular communication networks
• Wireless networking
• Mobile computing

Background:

• Vehicular networks rely on roadside units (RSUs) for content delivery, but deployment challenges create connectivity gaps (outage zones).
• These intermittently connected vehicular networks suffer from increased download delays and traffic overhead due to large content sizes and distances between RSUs.
• Existing precaching schemes for these networks often reduce content request success rates and fairness by prioritizing top-ranked vehicles.

Purpose of the Study:

• To propose a novel Multiple Pre-caching Vehicle Selection (MPVS) scheme for efficiently selecting a group of vehicles to precache and relay content in vehicular network outage zones.
• To address the limitations of previous schemes in terms of content request success rate and fairness in intermittently connected vehicular networks.

Main Methods:

• Developed numerical models to determine the necessity and amount of precaching, and to calculate available precaching capacities of individual vehicles.
• Designed MPVS to calculate all possible vehicle sets for precaching and rank them based on available precaching amounts.
• Identified the optimal content-precaching vehicle group by considering both set rankings and vehicle communication overheads.
• Established a content downloading process utilizing the selected vehicle group within outage zones.

Main Results:

• Simulation results demonstrated that the MPVS scheme significantly outperforms a representative previous scheme in various environmental and mobility models.
• The proposed MPVS scheme effectively manages content precaching and relaying in vehicular network outage zones.
• MPVS enhances the efficiency of content delivery in intermittently connected vehicular networks.

Conclusions:

• The MPVS scheme provides a superior solution for content delivery in intermittently connected vehicular networks compared to existing methods.
• MPVS improves the success rate and fairness of content requests by intelligently selecting a group of precaching vehicles.
• This approach offers a viable strategy for mitigating connectivity issues in challenging vehicular network deployments.