Content Caching and Delivery in Heterogeneous Vehicular Networks

Connected and automated vehicles (CAVs), which enable information exchange and content delivery in real time, are expected to revolutionize current transportation systems for better driving safety, traffic efficiency, and environmental sustainability. However, the emerging CAV applications such as content delivery pose stringent requirements on latency, throughput, reliability, and global connectivity. The current wireless networks face significant challenges to satisfy the requirements due to scarce radio spectrum resources, inflexibility to dynamic traffic demands, and geographic-constrained fixed infrastructure deployment. To empower multifarious CAV content delivery, heterogeneous vehicular networks (HetVNets), which integrate the terrestrial networks with aerial networks formed by unmanned aerial vehicles (UAVs) and space networks constituting of low Earth orbit (LEO) satellites, can guarantee reliable, flexible, cost-effective, and globally seamless service provisioning. In addition, edge caching is a promising solution to facilitate content delivery by caching popular files in the HetVNet access points (APs) to relieve the backhaul traffic with a lower delivery delay. The main technical issues are: 1) to fully reveal the potential of HetVNets for content delivery performance enhancement, content caching scheme design in HetVNets should jointly consider network characteristics, vehicle mobility patterns, content popularity, and APs’ caching capacities; 2) to fully exploit the controllable mobility and agility of UAVs to support dynamic vehicular content demands, the caching scheme and trajectory design for UAVs should be jointly optimized, which has not been well addressed due to their intricate inter-coupling relationships; and 3) for caching-based content delivery in HetVNets, a cooperative content delivery scheme should be designed to enable the cooperation among different network segments with ingenious utilization of heterogeneous network resources. In this thesis, we design the content caching and delivery schemes in the caching-enabled HetVNet to address the three technical issues. First, we study the content caching in HetVNets with fixed terrestrial APs including cellular base stations (CBSs), Wi-Fi roadside units (RSUs), and TV white space (TVWS) stations. To characterize the intermittent network connection caused by limited network coverage and high vehicle mobility, we establish an on-off model with service interruptions to describe the vehicular content delivery process. Content coding then is leveraged to resist the impact of unstable network connections and enhance caching efficiency. By jointly considering file characteristics and network conditions, the content placement is formulated as an integer linear programming (ILP) problem. Adopting the idea of the student admission model, the ILP problem is then transformed into a many-to-one matching problem between content files and HetVNet APs and solved by our proposed stable-matching-based caching scheme. Simulation results demonstrate that the proposed scheme can achieve near-optimal performances in terms of delivery delay and offloading ratio with a low complexity. Second, UAV-aided caching is considered to assist vehicular content delivery in aerial-ground vehicular networks (AGVN) and a joint caching and trajectory optimization (JCTO) problem is investigated to jointly optimize content caching, content delivery, and UAV trajectory. To enable real-time decision-making in highly dynamic vehicular networks, we propose a deep supervised learning scheme to solve the JCTO problem. Specifically, we first devise a clustering-based two-layered (CBTL) algorithm to solve the JCTO problem offline. With a given content caching policy, we design a time-based graph decomposition method to jointly optimize content delivery and UAV trajectory, with which we then leverage the particle swarm optimization algorithm to optimize the content caching. We then design a deep supervised learning architecture of the convolutional neural network (CNN) to make online decisions. With the CNN-based model, a function mapping the input network information to output decisions can be intelligently learnt to make timely inferences. Extensive trace-driven experiments are conducted to demonstrate the efficiency of CBTL in solving the JCTO problem and the superior learning performance with the CNN-based model. Third, we investigate caching-assisted cooperative content delivery in space-air-ground integrated vehicular networks (SAGVNs), where vehicular content requests can be cooperatively served by multiple APs in space, aerial, and terrestrial networks. In specific, a joint optimization problem of vehicle-to-AP association, bandwidth allocation, and content delivery ratio, referred to as the ABC problem, is formulated to minimize the overall content delivery delay while satisfying vehicular quality-of-service (QoS) requirements. To address the tightly-coupled optimization variables, we propose a load- and mobility-aware ABC (LMA-ABC) scheme to solve the joint optimization problem as follows. We first decompose the ABC problem to optimize the content delivery ratio. Then the impact of bandwidth allocation on the achievable delay performance is analyzed, and an effect of diminishing delay performance gain is revealed. Based on the analysis results, the LMA-ABC scheme is designed with the consideration of user fairness, load balancing, and vehicle mobility. Simulation results demonstrate that the proposed LMA-ABC scheme can significantly reduce the cooperative content delivery delay compared to the benchmark schemes. In summary, we have investigated the content caching in terrestrial networks with fixed APs, joint caching and trajectory optimization in the AGVN, and caching-assisted cooperative content delivery in the SAGVN. The proposed schemes and theoretical results should provide useful guidelines for future research in the caching scheme design and efficient utilization of network resources in caching-enabled heterogeneous wireless networks