Master Thesis MSTR-3458

BibliographyRiaz, Zohaib: Optimized Position Update Protocols for Secure and Efficient Position Sharing.
University of Stuttgart, Faculty of Computer Science, Electrical Engineering, and Information Technology, Master Thesis No. 3458 (2013).
60 pages, english.
CR-SchemaC.2.0 (Computer-Communication Networks, General)
K.4.1 (Computers and Society Public Policy Issues)
Abstract

The large scale availability of sophisticated devices such as the mobile phones has opened a new paradigm of possibilities in terms of the services that can be provided to common masses. Most mobile phones nowadays come with integrated GPS receiver which can provide the user with their accurate location information. This has resulted in many common mobile applications which use this location data to provide services to users such as the point of interest finder, friend finder and navigation services etc. The location data from the GPS receiver is uploaded to Location Servers (LSs) by the user’s device. Location Based Services, LBSs, running on client device communicate with remote LSs to download a user’s location data and process it appropriately to offer client the desired service. If a malicious third party can get access to the location information on LSs, a user’s privacy can significantly be violated. This can result in serious concerns being raised over the security of the user. Many approaches have been proposed to sufficiently protect user privacy in location services. Most of them rely on the availability of a trusted third party (TTP) [1] at the LS’s end. However, the assumption of a TTP is justifiably unrealistic [2] as events have been recorded where trusted parties have given up private user information to unauthorized clients. To overcome this problem, a novel approach was proposed in [3] which assumes partially trusted LBSs and LSs. This approach breaks down the position information into a number of shares and distributes each share to a separate LS managed by a different operator. The location information in any one share is equal and can be incrementally compiled by a LBS based on the number of shares it can access. The user grants access rights to different LBSs according to their trust on them. The overall effect is to have a possibility of graceful degradation of privacy if some of the location shares are compromised. There is however a price paid in the form of higher communication load. When the mobile object sends an update, it has to send a share to each LS. Similarly, when a client needs a particular user’s location, it has to query a number of LSs according to its access rights. Consequently, a higher communication cost is incurred which is undesirable for implementation of these privacy schemes. This thesis studies the possibility of applying traditional as well as subjective approaches for reduction of communication overhead in Position Sharing (PS) algorithm. From among traditional approaches, we have focused on Dead Reckoning (DR) where LSs predict the user location based on last location update and additional information about user movement. This results in less location updates with the compromise of incurring a known amount of bounded inaccuracy in the predicted position calculated by the LSs. A subjective approach of Selective Update is chosen for analysis according to the structure of PS algorithm. In this approach, reduction in communication is achieved by updating only the minimum necessary shares for short movements made by the user. From the observed behavior, a third technique as a merger of the mentioned techniques is also inferred. Finally, the viability of these approaches as possible solutions for reducing communication overhead in PS algorithm is studied and their results are evaluated on a set of real world GPS traces.

Full text and
other links
PDF (3305076 Bytes)
Department(s)University of Stuttgart, Institute of Parallel and Distributed Systems, Distributed Systems
Superviser(s)Dürr, Frank
Entry dateMay 19, 2014
   Publ. Computer Science