Secure and Privacy-aware Beacon-based Communication in VANET

Abstract

Today's vehicles are no more ‘century old machines with wheels’, but rather becoming more intelligent and capable of making smart decisions in order to guarantee a safe and infotainment-rich driving experience to the drivers and passengers. The dream of such intelligent transportation system (ITS) is realized through a technology referred to as Vehicular Ad hoc NETwork (VANET). VANET is a noteworthy and specialized breed of Mobile Ad hoc NETwork (MANET) where mobility of the vehicular nodes is directed, and restricted by the road topologies. Nevertheless the ephemeral nature of vehicles and varying human behavior towards driving makes it more challenging than its parental MANET. The basic idea of VANET is to take the widely adopted and inexpensive wireless networking technology, with a few tweaks, and install it on the vehicles in order to communicate with each other. However if it was so simple just to migrate the wireless networking technology to the vehicles, research community including academia and consortia would never produce remarkable results after tapping their research resources and gigantic amount of brainstorming. On the other hand, despite the surge in VANET research, security and privacy issues have been the root cause, at least in part, of the impeded momentum in VANET deployment. Besides, from economical point of view, the investors are reluctant to put their huge investment at stake because of the upfront hardware and management cost of the infrastructure deployment. The complete deployment of VANET is still rapidly ahead and its success and adaptation in end-users (drivers), consumers, and governments will depend on viable design, security solutions, and consumers’ satisfaction. VANET is a hybrid architecture and a combination of both infrastructureless and infrastructure-based communication paradigms. Vehicle-to-Vehicle (V2) communication is the infrastructureless communication paradigm where vehicles communicate with other vehicles in the communication range directly. In Vehicle-to-Infrastructure (V2I) communication, roadside units (RSUs) serve as intermediate gateways between the vehicular nodes and the authorities and thusforth termed as infrastructure-based communication paradigm. VANET applications range from safety applications such as accident warning, black ice on the pavement, ambulance approaching, all the way to infotainment applications such as games, movie-on-demand, and internet-on-the-move, to name a few. This dissertation covers the security and privacy issues in VANET. More precisely we solve the privacy problems with conflicting requirements in beacons and other safety messages. First we focus on the privacy-aware efficient one-hop beaconing mechanism with stringent information delivery ratio requirement in VANET, and propose a cooperative solution to guarantee successful one-hop message delivery. The proposed solution normalizes the effect of obstacles that cause deteriorated quality of service (QoS) as a result of degraded information delivery even in case of single-hop communication. However strengthening the privacy aspect gives rise to rather more serious threat in the VANET environment, i.e. Sybil attack. In Sybil attack, the attacker fabricates and/or forges identities to disseminate false information to other nodes, and creates illusions for the benign vehicles. In the presence of Sybil attack, almost all kinds of attacks are possible in VANET, therefore we propose a lightweight pre-assembly analysis based Sybil attack detection scheme for beacons and an RSU-assisted token-based Sybil attack detection and mitigation mechanism in case of safety and/or warning messages. Nevertheless, in order to detect Sybil attack, RSUs need to link messages that are generated and sent by the same sender. In such scenario, the privacy of the users may be prone to abuse. Since privacy and Sybil attacks are two conflicting requirements, therefore we solve the conditional privacy preservation problem with a novel identity-exchange mechanism that increases the anonymity of nodes to preserve their conditional privacy. It is also worth noting that RSUs may become bottleneck in the Sybil attack detection mechanism due to either initial deployment stage of VANET or the sparse deployment of RSUs due to economical and optimal location problems. Therefore we solve the sparse coverage problem caused by the static RSUs with the help of an already established public buses-based mobile intermediary infrastructure (MII). Due to the spatiotemporal characteristics of public transport in urban scenarios, public buses with necessary hardware, are ideal candidates to provide coverage to the neighbor vehicles in VANET. We prove the feasibility of the proposed infrastructure with the help of extensive simulations of a use-case. Our simulation results show that public buses can guarantee 100% coverage for other vehicles in the urban scenarios. We argue that our proposed infrastructure can be a strong rationale for the initial incremental deployment of VANET and upon success and the level of coverage, can possibly turn out to be a reasonable partial or complete replacement for the static RSUs in the urban scenarios.