Practical private DNA string searching and matching through efficient oblivious automata evaluation

Abstract

In [18] it was shown that the ability to perform oblivious automata evaluation was useful for performing DNA searching and matching. By oblivious automata evaluation we mean that one participant has a finite state machine and the other participant has a sequence, and at the end of the protocol the sequence owner learns whether the machine accepts the sequence. A protocol was given in [18], but it required O(n) rounds (where n is the number of characters in the sequence) and O(mn) modular exponentiations (where m is the number of states in the automata). Both of these factors limit the applicability of this approach. In this paper we propose a new protocol that requires only O(1) rounds and reduces the number of modular exponentiations to O(n) without revealing any additional information. We have implemented both schemes and have shown experimentally that our scheme is two to three orders of magnitude faster than the previous scheme. © 2009 Springer Berlin Heidelberg.