Fine - tuned implementation of an efficient secure profile matching protocol

In this report we study and try to minimize the total execution time of a Profile Matching protocol, a secure computations protocol based on an operation called conditional gate. We focus on reducing the complexities and study the interplay between them, mainly the round, communication, computational, space and random complexities. We propose several optimizations and study their effect , both the theoretical limits and their effect on a real world implementation. The involved protocols were implemented in an application that performed an Approximate matcher protocol. In addition to this report, parts of this project are also the " Profile Matcher " application and the " Profile Matcher API Reference Manual ". What we show in this report is that the expected high round complexity can be reduced for this protocol and give ways to utilize the idle time caused by the need for interaction. Furthermore we show that the effects of scheduling of operations with high round complexity can be impressive compared to a straightforward implementation.