PROJECT TITLE :

Optimized Search-and-Compute Circuits and Their Application to Query Evaluation on Encrypted Data

ABSTRACT:

Personal query processing on encrypted databases permits users to get knowledge from encrypted databases in such a approach that the users’ sensitive data can be shielded from exposure. Given an encrypted database, users usually submit queries kind of like the subsequent examples: one) How several employees in a corporation make over U.S. $100000? a pair of) What's the common age of factory workers tormented by leukemia? Answering the questions requires one to go looking and then compute over the relevant encrypted information sets in sequence. In this paper, we are curious about efficiently processing queries that need both operations to be performed on absolutely encrypted databases. One immediate resolution is to use many special-purpose encryption schemes simultaneously; however, this approach is associated with a high computational value for maintaining multiple encryption contexts. Another resolution is to use a privacy homomorphic scheme. However, no secure solutions are developed that satisfy the efficiency requirements. During this paper, we have a tendency to construct a unified framework to efficiently and privately process queries with search and compute operations. For this purpose, the first half of our work involves devising several underlying circuits as primitives for queries on encrypted knowledge. Second, we have a tendency to apply two optimization techniques to improve the potency of those circuit primitives. One technique involves exploiting single-instruction-multiple-information (SIMD) techniques to accelerate the basic circuit operations. Not like general SIMD approaches, our SIMD implementation can be applied even to a single basic operation. The opposite technique is to use a large integer ring (e.g., $mathbb Z_2^t$ ) as a message space rather than a binary field. Even for an integer of $k$ bits with $k>- $ , addition will be performed using degree 1 circuits with lazy carry operations. Finally, we have a tendency to present various experiments performed by varying the thought-about parameters, such as the query sort and the number of tuples.