Files in this item

FilesDescriptionFormat

application/pdf

application/pdfRiccardo_Crepaldi.pdf (5MB)
(no description provided)PDF

Description

Title:Using vehicular networks for building distributed networks of services
Author(s):Crepaldi, Riccardo
Director of Research:Kravets, Robin H.
Doctoral Committee Chair(s):Kravets, Robin H.
Doctoral Committee Member(s):Abdelzaher, Tarek F.; Gupta, Indranil; Lee, Sung-Ju; Satyanarayanan, Mahadev
Department / Program:Computer Science
Discipline:Computer Science
Degree Granting Institution:University of Illinois at Urbana-Champaign
Degree:Ph.D.
Genre:Dissertation
Subject(s):Vehicular Networks
Vehicular Ad Hoc Network (VANET)
Opportunistic Internet
Clustering
Energy Management
Channel State Information (CSI)
Abstract:Current mobile networking support in urban environments is mostly achieved through wide-area cellular systems, resulting in expensive, limited access that must be shared among many users. While cellular access may be appropriate for smartphones, modern cars are built with abundant resources, compared to the small devices that mobile networks are generally identified with, which can generate data at rates that push the limits of the cellular infrastructure. These resource rich platforms also have the potential to enable city-wide distributed networks of services. Since most current research in vehicular networks focuses on user-to-user communication, new solutions are needed to support this large-scale, dynamic environment, which not only best supports service-based communication, but also introduces new service-based communication patterns. Networking challenges in vehicular networks stem from the dynamics of vehicles' movements, resulting in a lack of structure in the network and short duration of contact opportunities. However, by exploiting the specific properties of vehicular networks these challenges can be mitigated. In our research, we explore the effect of the specific characteristics of vehicular networks, in shaping the services and their demands. We leverage the natural phenomenon of clustering and the abundance of parked cars to design a network that supports a variety of services. We also take into account the challenge of energy management, especially important for parked vehicles, and that of quickly changing channel conditions that requires fast adaptation algorithms. Given that all of these solutions are targeted at real-world problems, the strength of our research lies in the fact that we designed and built a prototype device based on customized hardware for use in vehicular networks. This prototype provides the necessary capabilities to practically demonstrate the validity of the design of our solution and confirms our results in a real-world scenario.
Issue Date:2014-01-16
URI:http://hdl.handle.net/2142/46740
Rights Information:Copyright 2013 Riccardo Crepaldi
Date Available in IDEALS:2014-01-16
Date Deposited:2013-12


This item appears in the following Collection(s)

Item Statistics