Physical Pseudo Random Function in Radio Frequency Sources for Security

Di Qiu, Dave De Lorenzo, Sherman Lo, Dan Boneh, Per Enge

Abstract:	There is tremendous market potential for location-based services (LBS), enabled by the rapid growth in the numbers of personal navigation devices and GPS-enabled mobile handsets. One of the major difficulties for LBS applications is accurate and reliable indoor positioning caused by the difficulty of acquiring and tracking GPS satellite signals in the absence of a clear unobstructed path between the satellite and the user. An alternative approach is to use terrestrial signal sources, such as cellular transmitters, TV or FM broadcast, or Wi-Fi access points. However, it is difficult to achieve accuracies comparable to outdoor GPS, since either the time resolution is inferior (for time-of-arrival methods) or the signal propagation characteristics are unknown or poorly modeled (for received signal strength methods). Multiple RF signals can be used together to compute a location tag, which is a form of physical pseudo random function. There is no need to synchronize the different systems either on the transmitter or the receiver ends. The usable parameters for location tag generation include signal strength, time-of-arrival, and RF signal phase information. Increased parameter diversity, greater RF signal variety, and more transmitter numbers all increase the information entropy in the derived tags and improve position estimation robustness. The tradeoff of using many location parameters is the reduction in tag reproducibility as RF signals are contaminated by noises and biases. The analytical results then are applied to real data sets. The use of location tags and a fuzzy extractor is investigated in an office building, a parking structure, and a soccer field on the campus of Stanford University. We show that under ideal conditions, the position tag based approached can achieve an accuracy of 3~6 meters in an indoor environment. Furthermore, the usage of location tags is not restricted to positioning; many other applications include equipment tracking, object identification, and data access control for security services, etc.
Published in:	Proceedings of the 2009 International Technical Meeting of The Institute of Navigation January 26 - 28, 2009 Disney's Paradise Pier Hotel Anaheim, CA
Pages:	84 - 92
Cite this article:	Qiu, Di, De Lorenzo, Dave, Lo, Sherman, Boneh, Dan, Enge, Per, "Physical Pseudo Random Function in Radio Frequency Sources for Security," Proceedings of the 2009 International Technical Meeting of The Institute of Navigation, Anaheim, CA, January 2009, pp. 84-92.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In