Name : Susanne Wetzel

Institution : Stevens Institute of Technology,

Department of Computer Science,

Castle Point on Hudson, Hoboken, NJ 07030, USA

Susanne Wetzel is an Associate Professor at the Computer Science Department of the Stevens Institute of Technology. She first joined the faculty at Stevens as Assistant Professor in 2002. She received her Diplom in Computer Science from the University of Karlsruhe (Germany) and her Ph.D degree in Computer Science from Saarland University (Germany) in 1998. Subsequently, she worked at DaimlerChrysler Research (Stuttgart, Germany), Lucent Technologies Bell Laboratories (Murray Hill, USA) and RSA Laboratories (Stockholm, Sweden). Her research interests are in cryptography and algorithmic number theory. In the field of cryptography, her research is focused on wireless security, secret sharing, privacy, and biometrics, and her contributions range from analysis to protocol design. In algorithmic number theory, her research is centered on lattice theory, in particular on developing new algorithms and heuristics for lattice basis reduction. For more details, please refer to http://www.cs.stevens.edu/~swetzel.

Publications :

• Cordasco, J., Wetzel, S.: An Attacker Model for MANET Routing Security. Proceedings of the 2nd ACM Conference on Wireless Network Security (WiSec), 2009, pp. 87-94.
• Jakobsson, M., Yang, L., Wetzel, S.: Quantifying the Security of Preference-based Authentication. Proceedings of the ACM CCS 2008 Workshop on Digital Identity Management (DIM), 2008, pp. 61-69.

Title of Project: Privacy-preserving Policy Reconciliation

Enabling collaboration between organizations is a very challenging task. The main difficulty lies in determining the policy rules that should govern this process. That is, the participants must accept a set of rules that they will follow for the duration of the collaboration. Each party has to express their desired rules in a common format and follow a protocol with the other parties. The process will determine which rules control their future collaboration. This protocol is called a reconciliation protocol and its output is a policy that is consistent with the requirements of all the participants. If such a policy exists, the participants can continue their collaboration. Otherwise, they can decide not to collaborate or they can decide to modify their individual requirements and repeat the protocol.

Consider the following example: A company has posted a job opening. Multiple applications have been received and HR would like to schedule interviews with some of the applicants. To successfully schedule an interview, both HR and the respective applicant need to reach an agreement on the date and time for the interview. I.e., they need to reconcile their policies, which-in this case-regulate the dates and times when the individual parties could do the interview. Either party may want certain information not to be disclosed to the other party during the reconciliation process. For example, HR may not want to disclose its schedule to the applicants in order not to allow for them to infer the number of applicants interviewed for the opening.

However, state-of-the-art policy reconciliation mechanisms require that at least one of the parties discloses all the information to allow for the reconciliation process to work. It is in this context that we strive to integrate privacy mechanisms into the policy reconciliation processes. The goal of this project is to develop protocols that guarantee that parties participating in the reconciliation process learn nothing about the other party's policies other than what policies they have in common.

Joint work with Ulrike Meyer, Daniel Mayer, Sotiris Ioannidis, and Jonathan Voris.