[Paper Review] Air-Ground Integrated Vehicular Network Slicing with Content Pushing and Caching

Shan Zhang, Wei Quan+4 more|arXiv (Cornell University)|Jun 11, 2018

Opportunistic and Delay-Tolerant Networks38 references3 citations

TL;DR

This paper proposes an air-ground integrated vehicular network (AGIVEN) with network slicing that leverages high-altitude platforms (HAPs) for proactive content pushing and roadside units (RSUs) for on-demand unicast services. By jointly optimizing HAP broadcast rates, vehicle cache sizes, and RSU transmission rates, the framework reduces RSU traffic by 40% while maintaining QoS for map and popular content services.

ABSTRACT

In this paper, an Air-Ground Integrated VEhicular Network (AGIVEN) architecture is proposed, where the aerial High Altitude Platforms (HAPs) proactively push contents to vehicles through large-area broadcast while the ground roadside units (RSUs) provide high-rate unicast services on demand. To efficiently manage the multi-dimensional heterogeneous resources, a service-oriented network slicing approach is introduced, where the AGIVEN is virtually divided into multiple slices and each slice supports a specific application with guaranteed quality of service (QoS). Specifically, the fundamental problem of multi-resource provisioning in AGIVEN slicing is investigated, by taking into account typical vehicular applications of location-based map and popularity-based content services. For the location-based map service, the capability of HAP-vehicle proactive pushing is derived with respect to the HAP broadcast rate and vehicle cache size, wherein a saddle point exists indicating the optimal communication-cache resource trading. For the popular contents of common interests, the average on-board content hit ratio is obtained, with HAPs pushing newly generated contents to keep on-board cache fresh. Then, the minimal RSU transmission rate is derived to meet the average delay requirements of each slice. The obtained analytical results reveal the service-dependent resource provisioning and trading relationships among RSU transmission rate, HAP broadcast rate, and vehicle cache size, which provides guidelines for multi-resource network slicing in practice. Simulation results demonstrate that the proposed AGIVEN network slicing approach matches the multi-resources across slices, whereby the RSU transmission rate can be saved by 40% while maintaining the same QoS.

Motivation & Objective

• To address the challenge of managing multi-dimensional, heterogeneous resources in air-ground integrated vehicular networks (AGIVENs) with diverse QoS requirements.
• To enable efficient network slicing that supports distinct vehicular services—such as HD map delivery and popular content distribution—through virtualized, service-specific slices.
• To quantify the trade-off between HAP broadcast rate, vehicle cache size, and RSU transmission rate for optimal resource provisioning.
• To derive analytical models for content hit ratios and delay constraints in proactive caching scenarios to guide system design.

Proposed method

• Proposes an AGIVEN architecture integrating HAPs for wide-area broadcast and RSUs for high-rate unicast, enabling proactive content delivery and on-board caching.
• Introduces a service-oriented network slicing framework that partitions the physical network into three slices: MaNa (HD map), FoCI (popular content), and ODT (on-demand traffic).
• Derives the accomplishment ratio for the MaNa slice as a function of HAP broadcast rate and vehicle cache size, revealing a saddle point for optimal resource trade-off.
• Models the on-board content hit ratio for FoCI services using a popularity-based caching model, showing that HAP pushing improves cache freshness.
• Develops a closed-form expression for the minimal RSU transmission rate required to meet average delay constraints, derived from a stochastic model of vehicle mobility and request dynamics.
• Uses the Gamma distribution and incomplete gamma function to model the expected number of map segments cached, enabling analytical evaluation of caching performance.

Experimental results

Research questions

• RQ1 How can HAP broadcast and vehicle caching be jointly optimized to minimize RSU load while ensuring low-latency map delivery in vehicular networks?
• RQ2 What is the optimal trade-off between HAP broadcast rate and vehicle cache size for achieving maximum map service completion ratio?
• RQ3 How does proactive content pushing from HAPs affect the on-board hit ratio for popular content, and what is the required HAP broadcast rate to maintain freshness?
• RQ4 What is the minimal RSU transmission rate required to meet the average delay requirements of each network slice under mobility and content popularity constraints?

Key findings

• The accomplishment ratio for the MaNa slice increases with HAP broadcast rate, but the rate of increase first rises and then falls, indicating a saddle point for optimal HAP broadcast rate.
• The minimal RSU transmission rate required to meet average delay constraints is derived in closed form, showing a non-linear dependence on vehicle mobility and request patterns.
• Simulation results confirm a 40% reduction in RSU transmission rate when using proactive HAP content pushing and edge caching, while maintaining the same QoS across slices.
• The content hit ratio for popular content decreases with increasing content popularity skew (ρ), and the model proves this monotonic decrease analytically for ρ < 1 and ρ > 1.
• The optimal HAP broadcast rate exhibits an inversely linear relationship with vehicle cache size, indicating that larger caches reduce the need for high broadcast rates.
• The analytical model for the expected number of cached map segments is validated to be strictly increasing with respect to the HAP broadcast rate, confirming the effectiveness of proactive pushing.