Date of Award

Spring 2010

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Computer Science

Committee Director

Stephan Olariu

Committee Member

Michele C. Weigle

Committee Member

Kurt J. Maly

Committee Member

Ravi Mukkamala

Committee Member

Jinhua Guo

Abstract

Location is fundamental information in Vehicular Ad-hoc Networks (VANETs). Almost all VANET applications rely on location information. Therefore it is of importance to ensure location information integrity, meaning that location information is original (from the generator), correct (not bogus or fabricated) and unmodified (value not changed). We present validation mechanisms to provide location integrity for VANETs. In the initial mechanism, we assume that all vehicles are equipped with a radar, a GPS receiver, and a transceiver. Since radar has a limited radar range and transceiver has a limited transmission range, we build network cells as a security unit as well as a communication unit. To ensure the intra-cell position information integrity, we propose an active validation mechanism (called active location integrity) that actively validates and enhances position security by enlisting the help of on-board radar to detect neighboring vehicles and to confirm their announced coordinates. Since radar is not currently installed in many vehicles, we weak the assumption by removing radar from the vehicle's equipments and propose the second mechanism (called passive location integrity) that maintains the mobility history records of vehicles, called the Map History. Based on a vehicle's Map History, we can predict a region where the vehicle will be present. The predicted region can be used to validate the announced position. In reality, vehicles are deployed with different combinations of equipment and some old vehicles may not have these devices. We address a validation mechanism (called general location integrity) which filtered and refined the location measurements obtained by the above active and passive location integrity methods. The three mechanisms above provide intra-cell position information integrity.

Since applications often involve position information of remote vehicles or entities which are beyond a cell (ranging to miles), we provide inter-cell position integrity as well. Vehicles request that neighbors or opposite-side vehicles check the announced position information of remote vehicles. Both the request and response messages will be propagated among cells. Because of the high mobility of vehicles, the routing path is fragile. To improve location availability, we propose a stable routing scheme which will select and maintain stable routing paths. Both selection and maintenance of routing paths are based on a proposed probability analysis of VANET links. In addition, plaintext location information, especially aggregated location information, is vulnerable to attack as an attacker could easily modify the location information and harm the location integrity. We propose both encryption/decryption and access control mechanisms to provide location information confidentiality. The aggregated position message is encrypted by a key which is a geographic location which specifies a decryption region. Vehicles have to be physically present in the specified decryption region to decrypt or access the aggregated position information. As we can ensure the position information confidentiality, integrity, and availability, we achieve position information security based on the security requirements outlined in the CIA model (confidentiality, integrity, and availability).

Comments

This record includes a pdf of the author's Dissertation Defense Presentation. Also available on Slideshare: http://www.slideshare.net/mweigle/providing-location-security-in-vehicular-networks

ISBN

9781124148113

yan-defense-2010-100831051807-phpapp02.pdf (3525 kB)
Dissertation Defense Presentation, April 26, 2010

Share

COinS