Most of my current research is very applied.

My main research area is digital forensics, focusing on novel tools and techniques to make digital forensics investigation easier, faster, and more productive. Vassil Roussev and I are developing a distributed computing framework for digital forensics which runs on commodity compute clusters and provides astonishing performance gains for large targets. The framework not only accelerates current generation tasks such as keyword searches, image thumbnailing, and file carving, but also provides sufficient computing resources to enable a new generation of forensic analysis, including better image handling, evidence correlation, and steganography detection. The framework is called DELV (Digital Environment for Large Scale Investigation). A DFRWS paper that describes the preliminary version is in the publications area. Some of our papers on next-generation digital forensics also reference this system.

Some of our latest work is on using Graphics Processing Units (GPUs), specifically, the NVIDIA G80, to increase performance of certain digital forensics techniques. A paper on this was presented at the 7th Annual Digital Forensics Research Workshop (DFRWS 2007) in Boston. This paper received the highest review ratings in the conference. We are also working hard on live forensics tools for Linux.

Another ongoing project is Scalpel, a very fast file carving application for Linux (and Win32, if you must). You can always find the latest version of Scalpel here: http://www.digitalforensicssolutions.com/Scalpel. A paper describing Scalpel is available via my publications page. See the link at the bottom. Scalpel also forms the basis of our new "next-generation" rules-based carving framework. We've made some progress in "next generation" carving already, by introducing the idea of in-place carving. This work appeared in the volume Advances in Digital Forensics III, published by Springer in 2007.

In 2002, I wrote a book on service discovery, Service and Device Discovery: Protocols and Programming, for McGraw-Hill. The book covers Jini, UPnP, SLP, and Bluetooth SDP. I also completed a book on mobile computing with Frank Adelstein, Sandeep Gupta, and Loren Schwiebert. You can check it out on Amazon.com here: Fundamentals of Mobile and Pervasive Computing.

I'm currently working with Loren Schwiebert, of Wayne State University, on service discovery protocols for wireless sensor networks. A paper is coming that describes our initial efforts.

Abdul Altalhi, a recent Ph.D. graduate, worked on improvements to Dynamic Source Routing (DSR), including support for diverse route metrics. Another Ph.D. student, Lawrence KLos, who will graduate soon, is working on reliable multicast in ad hoc networks.

In the past, I worked with Loren Schwiebert of Wayne State and Frank Adelstein of ATC-NY on efficient multicast protocols for dynamic multicast groups. Part of this work involved development of Bessie, a network topology generation and analysis tool.   We also developed a simulation environment in csim to explore the performance of our algorithms.  Bessie is used to create random network topologies and then to visualize and verify the integrity of the multicast trees generated by the simulator.  Our concentration is now on mobile networks. Papers describing Bessie and the multicast algorithms are available via the papers link at the bottom of this page.

In the distant past, my research focus was on developing recovery techniques for distributed shared memory (DSM) and message-passing systems. In the DSM area, I introduced logging as a means of reducing checkpointing overhead in 1993. There followed an incredible number of papers improving on this work, none of which were mine. Later, some preliminary work on recovery mechanisms for Java was done in collaboration with Shengru Tu.

Along the way, a lot of random stuff.

General interests are computer security, operating systems, networking, and computer graphics.