A secure and privacy-preserving incentive framework for vehicular cloud

Vehicular Cloud Computing (VCC) plays a critical role in data generation where a large number of vehicles collect various kinds of sensing resources with large-volume features. While the information that should be collected is essential for the success of VCC applications, how to stimulate the vehicle owners to provide their sensing resources in VCC is also crucial to its success. When vehicle owners choose to contribute their data for economically appealing compensation, they may be concerned about their privacy. This thesis proposes a promising secure and privacy-preserving incentive mechanism framework for VCC. An incentive to convince vehicle owners with excess on-board capabilities to join in VCC without the risk of privacy disclosure. The incentive mechanism employs game theory to model the interactions between the VCC server and vehicles. With the incentive mechanism, the VCC server, which represents the task announcement can select competent vehicles to collaborate for the announced task, and the vehicle owners can earn payments from their participation. The signcryption technique and homomorphic concept are exploited to achieve mutual authentication between the VCC server and vehicles, and prevent the privacy information of these vehicles from being disclosed. Moreover, we study the situation in which the VCC server announces a task that can be exploited by a malicious roadside unit to reveal vehicles' privacy. Therefore, we propose a novel secure and privacy-preserving scheme for enhancing security in VCC-based tasks announcement. The proposed scheme combines a multiple receiver signcryption technique with proxy re-encryption in order to protect message content that includes the private information of the vehicles from being disclosed during task announcement. In addition to the above schemes, while the VCC server is responsible for recruiting vehicles to collaborate for the announced task, it may not be fully trusted, and the disclosure of individual locations has serious privacy implications. Thus, by exploiting Lagrange interpolating polynomials, we design a privacy-preserving location matching mechanism, called LATE, to enable the VCC server determining whether the interested vehicle participant is in a geocast region of a spatial task or not without having any knowledge about the task's geocast regions and the vehicle's location