[Paper Review] BeeTrace: A Unified Platform for Secure Contact Tracing that Breaks Data Silos

Xiaoyuan Liu, Ni Trieu+2 more|arXiv (Cornell University)|Jul 5, 2020

COVID-19 Digital Contact Tracing10 references4 citations

TL;DR

BeeTrace is a privacy-preserving, unified contact tracing platform that breaks data silos by integrating multi-source data—including business records and personal device data—using secure multi-party computation (MPC) and standardized data formats. It enables efficient, private identification of high-risk locations through oblivious aggregation and heavy hitters computation, achieving strong privacy guarantees while improving outbreak detection coverage.

ABSTRACT

Contact tracing is an important method to control the spread of an infectious disease such as COVID-19. However, existing contact tracing methods alone cannot provide sufficient coverage and do not successfully address privacy concerns of the participating entities. Current solutions do not utilize the huge volume of data stored in business databases and individual digital devices. This information is typically stored in data silos and cannot be used due to regulations in place. To successfully unlock the potential of contact tracing, we need to consider both data utilization from multiple sources and the privacy of the participating parties. To this end, we propose BeeTrace, a unified platform that breaks data silos and deploys state-of-the-art cryptographic protocols to guarantee privacy goals.

Motivation & Objective

• To address the limitations of current contact tracing systems that rely only on GPS and Bluetooth, missing indirect contacts and failing to utilize business or device data.
• To enable secure data sharing across disparate data sources—such as businesses and individuals—without compromising user privacy.
• To standardize data formats and develop a unified platform that supports decentralized, privacy-preserving contact tracing across multiple stakeholders.
• To provide cryptographic solutions that protect sensitive data while enabling accurate identification of high-risk locations and exposed individuals.
• To demonstrate the feasibility of a scalable, privacy-preserving data collaboration model applicable beyond contact tracing.

Proposed method

• Introduces a unified data format with granularity adaptation to standardize contact and location data across diverse sources, including businesses and personal devices.
• Employs secure multi-party computation (MPC) protocols to enable private aggregation of visit records without exposing individual data.
• Uses oblivious aggregation to compute the total number of visits to each place across users while preserving privacy of individual visit records.
• Applies the heavy hitters problem solution to identify high-risk locations with frequent visits from diagnosed or undiscovered carriers.
• Deploys cryptographic techniques such as secure key-value aggregation and differential privacy-adjacent protocols to ensure confidentiality and integrity.
• Designs a decentralized architecture where a trusted aggregator (MA) computes risk heatmaps without learning individual user data.

Experimental results

Research questions

• RQ1 How can contact tracing systems effectively detect indirect contacts through shared locations, beyond direct person-to-person proximity?
• RQ2 What cryptographic protocols can enable secure aggregation of location data from multiple untrusted sources while preserving user privacy?
• RQ3 How can data silos in business and personal devices be broken down to improve contact tracing coverage without violating privacy regulations?
• RQ4 What are the trade-offs between efficiency, scalability, and security in privacy-preserving contact tracing systems?
• RQ5 Can a standardized, unified data format enable interoperability and large-scale deployment of secure contact tracing platforms?

Key findings

• BeeTrace enables the detection of high-risk locations by aggregating visit records from multiple sources—including businesses and individuals—while preserving user privacy through secure multi-party computation.
• The platform achieves private identification of frequently visited places (heavy hitters) using a secure, scalable protocol with communication and time complexity suitable for large-scale deployment.
• Oblivious aggregation allows the untrusted aggregator to compute the total number of visits per location without learning individual visit patterns or user identities.
• The system supports both diagnosed and undiscovered carriers as tracing starting points, improving detection of asymptomatic or pre-symptomatic transmission chains.
• By standardizing data formats and enabling cross-source data integration, BeeTrace significantly enhances contact tracing coverage compared to systems relying only on GPS or Bluetooth.
• The platform's design is extensible beyond contact tracing, with potential applications in private contact discovery and credential checking.