Applied Cryptography

Cryptography research at SPAR is concerned with the design and implementation of cryptographic protocols and primitives.  Our work focuses on efficiency and provable security, and includes digital signatures, hash functions, searchable encryption, identity-based encryption and elliptic curve cryptography.

Biometrics

Biometrics research as SPAR is primarily concerned with improving classic authentication techniques by exploiting dynamic biometric features such as keystroke dynamics, voice and handwriting.  Additionally, although biometric systems have recently received a great deal of interest, we believe that current practices for measuring the security of such systems are inadequate.  Thus, we seek to improve the security of biometric systems by developing more rigorous models for their evaluation.

Distributed Telescopes

We are investigating the use of multiple co-ordinated network telescopes to detect the spread of active Internet worms.  While individual telescopes have already been used for forensic analysis of Internet Worms such as Code Red and Slammer affecting million of computers worldwide, we believe that multiple small telescopes that are geographically distributed can have higher detection capability compared to a single large telescope and at the same time be easier to deploy.

Electronic Voting

Research in electronic voting at SPAR is conducted under the auspices of ACCURATE.  ACCURATE is a multi-institution voting research center funded by the National Science Foundation (NSF) under their CyberTrust program.  The goals of the center are to research ways in which technology can be used to improve voting systems and the voting process; to develop the science that will help inform the election community and the public about the tradeoffs among various voting technologies and procedures; to serve as a resource to the elections community, politicians, vendors and the public about issues related to public policy, technology, and law with respect to voting; and to publish and disseminate our research so that future systems can benefit from the center's work.

Network Simulation

Simnet is a discrete-event network simulator, designed specifically for analyzing network-security protocols.  The simulator was created for JHU's Network Security cource so its design and implementation is focused on simplicity of abstraction and extensibility.  Moreover, its modular architecture allows operators to dynamically customize running simulations.  Simnet is not only scalable and efficient, but provides a viable platform for prototyping and analyzing non-trivial security protocols.