Connectivity-Aware Routing in Vehicular Ad Hoc Networks

Vehicular ad hoc networks (VANETs) is a promising emerging technology that enables a wide range of appealing applications in road safety, traffic management, and passengers and driver comfort. The deployment of VANETs to enable vehicular Internet-based services and mobile data offloading is also envisioned to be a promising solution for the great demand of mobile Internet access. However, developing reliable and efficient routing protocols is one of the key challenges in VANETs due to the high vehicle mobility and frequent network topology changes. In this thesis, we highlight the routing challenges in VANETs with a focus on position-based routing (PBR), as a well-recognized routing paradigm in the vehicular environment. As the current PBR protocols do not support VANET users with connectivity information, our goal is to design an efficient routing protocol for VANETs that dynamically finds long life paths, with reduced delivery delay, and supports vehicles with instant information about connectivity to the infrastructure. The focus of this thesis will be on predicting vehicular mobility to estimate inter-vehicle link duration in order to support routing protocols with proactive connectivity information for a better routing performance. Via three stages to meet our goal, we propose three novel routing protocols to estimate both broad and comprehensive connectivities in VANETs: iCAR, iCAR-II, and D-CAR. iCAR supports VANET users with instant broad connectivity information to surrounding road intersections, iCAR-II uses cellular network channels for comprehensive connectivity awareness to Roadside Units (RSUs), and finally D-CAR supports users with instant comprehensive connectivity information without the assistance of other networks. Detailed analysis and simulation based evaluations of our proposed protocols demonstrate the validity of using VANETs for Internet-based services and mobile data offloading in addition to the significant improvement of VANETs performance in terms of packet delivery ratio and end-to-end delay.