Implementation of Oblivious Bloom Intersection in Private Set Intersection Protocol (PSI)

Today we are in the era of Big Data. The design of privacy preserving protocols in data processing is really challenging as the amount of data grows largely and complex. How to preserve privacy while meeting the requirements of speed and throughput have become critical criteria in the design. In this paper, we implement a practical use of Private Set Intersection (PSI) Protocol based on the new approach of oblivious Bloom intersection. The high scalability is achieved with parallel operations. We implemented the basic protocol and utilized Google Contact API to directly access the private contact information from two different Google accounts. The intersection of contact information could be found without disclosing any other private information from each account. We reported the result of the performance with respected to the number of contacts for different security levels. We only computed the intersections of two sets up to 25,000 contacts.