Towards a content-centric approach to crowd-sensing in vehicular clouds

• Published in: Journal of systems architecture Vol. 59; no. 10; pp. 976 - 984
• Main Authors: Talebifard, Peyman, Leung, Victor C.M.
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 01.11.2013; Elsevier Sequoia S.A
• Subjects: Cloud computing; Coding; Connectivity; Content-centric networking; Crowd-sensing; Data dissemination; Decision making; Delay; Information dissemination; IP (Internet Protocol); Messages; Network coding; Network topologies; Networks; Semantics; Studies; Telematics; Vehicular cloud; Vehicular networks; Network coding; Content-centric networking; Vehicular networks; Data dissemination; Crowd-sensing; Vehicular cloud
• DOI: 10.1016/j.sysarc.2013.07.013
• ISSN: 1383-7621

Vehicular mobile cloud computing is a new research direction that will address the interactions of vehicles and other mobile devices that participate in collaborative sensing, processing and dissemination of information. Due to intermittent connectivity that can cause substantial message delivery delays or loss of messages, conventional internet protocols may not be suitable to support reliable and efficient information dissemination over vehicular networks. In this paper we address the problem of information dissemination in vehicular clouds using the new paradigm of semantic based networking of information. We show that a more intelligent and context-aware networking of information is advantageous in developing more scalable and reliable methods of routing and dissemination of information. In particular we look at the network topology beyond physical connectivity but in terms of content connectivity and how the participating nodes can be clustered based on their context and interests in some relevant information. We propose a novel selective network coding method to enhance the reliability and efficiency of information dissemination, which uses node clusters in a graph model of the information connectivity as the basis of the decision making. Analysis and numerical results show improved reliability and a lower complexity compared with the epidemic network coding based dissemination approach.