Research - Faculty
Professor and Associate Chair for Undergraduate Studies Charles Boncelet
Message Authentication in Noisy Environments
- Image Compression
- Lossless Data Compression
- Message Authentication in Noisy Environments
- Steganalysis of Digital Images and Videos
Lisa Marvel, Army Research Laboratory
Office: Evans 102
Phone: 302- 831-8008
When one receives a message, how can one be sure the message comes from the purported sender, and how can one be sure the message has not been deliberately altered in transit? This is the realm of message authentication. Conventional solutions use a secret key and compute a signature of the message. This signature is attached to the message and both are transmitted. The receiver then computes a signature (using the same secret key) and compares the computed and received signatures. If they match the message is accepted as authentic; if they do not, the message is rejected and often retransmitted.
Unfortunately, conventional methods are ill suited for noisy wireless channels that may introduce a small number of random bit errors. We have introduced a series of "noise tolerant message authentication codes" that can identify these bit errors and allow the user to decide whether or not to accept the message.
These new message authentication codes also have some important cryptographic properties, principally they may not be vulnerable to recent attacks on conventional codes.
Our current work is on quantifying the performance of these new codes, on extending them to various application scenarios, and proving cryptographic properties.
"The CRC-NTMAC for Noisy Message Authentication," Yu Liu and C. G. Boncelet Jr., IEEE Trans. on Information Forensics and Security, December 2006.
"The NTMAC for Authentication of Noisy Messages," C. G. Boncelet Jr., IEEE Trans. on Information Forensics and Security, March 2006.
"The BCH-NTMAC for Noisy Message Authentication," Y. Liu and C. G. Boncelet Jr., Proceedings of the 2006 CISS, Mar. 2006.
"Efficient Noise-Tolerant Message Authentication Codes Using Direct Sequence Spread Spectrum Technique," S. Xiao and C. G. Boncelet Jr., Proceedings of the 2006 CISS, Mar. 2006.