Overview

Guarding Sensitive Data with Trigeneous Secure Computations

The use of personal information is essential for homeland security efforts, but it is also a balancing act securing this information and protecting individual privacy concerns that exist in the sharing and use of sensitive information such as surveillance images/videos, biometrics, and other individual-identifying data that is collected and generated from multiple sources across the Homeland Security Enterprise (HSE). In addition, there are technical and cost limitations with privacy-enhancing technologies that have made it challenging to implement in operating environments when employing PETs including field personnel’s ability to access and verify data quickly.

Solution 

This project combines three types of privacy-enhancing technologies (PETs) to protect private data while also making the data more quickly accessible to those in the field that need to use it for decision-making. This approach uses two methods of software cryptography called fully homomorphic encryption (FHE) and secure aggregation (SA) and pairs them with special hardware or trusted execution Environment (TEE). This method allows for secure querying of sensitive data sets from multiple sources and keeps the data encrypted and secure.

Impact 

The goal of this project is to develop highly efficient privacy-preserving solutions so that organizations can share and search data more quickly and cost-effectively from multiple sources while ensuring security and protecting individual privacy.

Research Leadership Team 

Principal Investigator: Taeho Jung, University of Notre Dame
Co-PI: Adam Czajka, University of Notre Dame
Co-PI: David Cousins, Duality