Since its inception only six years ago, the field of fully homomorphic encryption has advanced at an incredible pace. In only a few years, numerous optimizations were introduced to bridge the efficiency bottleneck that prevents practical deployments. Informally, fully homomorphic encryption allows the efficient evaluation of arbitrary functions directly on encrypted data while somewhat homomorphic encryption (SWHE) is more limited and permits the evaluation of only limited depth circuits.
In this talk we first briefly outline some of the advances in the area SWHE. In particular we present several noise management and new key size reduction techniques along with a number of optimizations. The SWHE construction is based on a variation of the scheme introduced by Lopez-Alt, Tromer and Vaikuntanathan. In what follows, we very briefly discuss a few implementation results that include CPU and GPU libraries, and domain specific ASIC/FPGA designs. We put our implementations into practice for the blind sorting of lists of integers. We present benchmarks for various sorting algorithms that includes two new algorithms optimized for shallow circuit evaluation.
Berk Sunar received his BSc degree in Electrical and Electronics Engineering from Middle East Technical University in 1995 and his Ph.D. degree in Electrical and Computer Engineering (ECE) from Oregon State University in December 1998. After briefly working at rTrust Inc. in Pasadena, CA as a Security Architect, in Fall 2000 Sunar joined Worcester Polytechnic Institute as a Professor. Sunar held visiting positions at the Bochum Ruhr University in Germany and Bilkent University, Turkey in 2007.
Sunar is currently leading the Vernam Applied Cryptography Group (vernam.wpi.edu) a research cluster specializing in the area of security. His research interests include applied cryptography and hardware security. Sunar published numerous papers in select peer-reviewed journals and conference proceedings. Sunar received the National Science Foundation CAREER award in 2002. In 2007 he won the IBM Research Pat Goldberg Memorial Best Paper award for his work on Hardware Trojan Detection.