Safer Illinois and RokWall: Privacy Preserving University Health Apps for COVID-19

Vikram Sharma Mailthody, James Wei, Nicholas Chen, Mohammad Behnia, Ruihao Yao, Qihao Wang, Vedant Agrawal, Churan He, Lijian Wang, Leihao Chen, Amit Agarwal, Edward Richter, Wen-Mei Hwu, Christopher W. Fletcher, Jinjun Xiong, Andrew Miller, Sanjay PatelarXiv 2021 · arXiv

Introduction

During the COVID-19 pandemic, digital contact tracing emerged as a critical tool for public health. However, developers faced a fundamental "privacy-utility trade-off." Decentralized systems, such as the Google/Apple Exposure Notification (GAEN) protocol, offer robust individual privacy but provide limited aggregate data for public health officials. Conversely, centralized systems enable powerful analytics like hotspot detection but raise significant surveillance concerns. Researchers at the University of Illinois at Urbana-Champaign (UIUC) addressed this by deploying Safer Illinois, a decentralized app, and proposing RokWall, a centralized architecture designed to preserve privacy through hardware-level encryption.

Bridging the Privacy-Utility Gap

The UIUC team argues that universities serve as ideal "trusted intermediaries" for such technology. Unlike private corporations, universities lack a profit motive to monetize data and often maintain a high level of trust within their community. This trust enabled the rapid deployment of Safer Illinois to a population of approximately 60,000 students and staff.

System Design: Safer Illinois and RokWall

Safer Illinois operates on a decentralized model using Bluetooth Low Energy (BLE) to exchange anonymous "Rolling Proximity Identifiers." This ensures that contact tracing occurs locally on user devices. To complement this, the researchers proposed RokWall, which utilizes Trusted Execution Environments (TEEs)—such as Intel SGX or AWS Nitro Enclaves—to perform centralized analytics without exposing raw user data to the service provider.

RokWall architecture utilizing Trusted Execution Environments (TEEs) for secure data processing.

In the RokWall model, data is uploaded to an encrypted database. Analysis occurs within a "secure enclave," a protected portion of the CPU where data is decrypted, processed (e.g., to generate a heatmap), and re-encrypted before leaving the enclave. This architecture employs Remote Attestation, allowing users to cryptographically verify the integrity of the code, and Monotonic Counters to prevent rollback attacks that might attempt to leak data by replaying old states.

Implementation Challenges and iOS Backgrounding

A significant hurdle in deploying Safer Illinois was the restrictive nature of mobile operating systems, particularly iOS. Apple’s iOS severely limits BLE advertising when an app is in the background. To maintain functionality, the team implemented a "local notification" workaround that briefly wakes the device screen, allowing the app to broadcast its presence for a 10-second window. While effective for contact tracing, this workaround had measurable impacts on device performance.

Real-World Performance and Adoption

The deployment at UIUC during the Fall 2020 semester saw remarkable success. Despite the app being optional, 82.5% of the campus population used Safer Illinois at least once, with 53% of those users voluntarily opting into digital contact tracing. The system successfully supported a high-throughput saliva-testing program capable of processing 10,000 tests per day.

However, the technical workarounds required for background operation resulted in significant battery drain, particularly on iOS devices.

Battery drain comparison between Android and iOS devices with the Safer Illinois app active.

As shown in the table above, the battery consumption on an iPhone X was nearly four times higher than on a Google Pixel 3 when the app was active. This highlights the ongoing tension between mobile OS security policies and the functional requirements of public health tools.

Conclusion and Future Work

The UIUC experience demonstrates that large-scale, privacy-preserving health apps are viable within trusted ecosystems. While Safer Illinois proved the efficacy of decentralized tracing, RokWall offers a path forward for secure centralized analytics. Future work remains to address the "digital divide" for users without smartphones and to refine TEE-based systems as cloud-based secure hardware becomes more widely available. Ultimately, these systems provide a blueprint for managing future public health crises while respecting individual digital rights.

Summary updated

Abstract

COVID-19 has fundamentally disrupted the way we live. Government bodies, universities, and companies worldwide are rapidly developing technologies to combat the COVID-19 pandemic and safely reopen society. Essential analytics tools such as contact tracing, super-spreader event detection, and exposure mapping require collecting and analyzing sensitive user information. The increasing use of such powerful data-driven applications necessitates a secure, privacy-preserving infrastructure for computation on personal data. In this paper, we analyze two such computing infrastructures under development at the University of Illinois at Urbana-Champaign to track and mitigate the spread of COVID-19. First, we present Safer Illinois, a system for decentralized health analytics supporting two applications currently deployed with widespread adoption: digital contact tracing and COVID-19 status cards. Second, we introduce the RokWall architecture for privacy-preserving centralized data analytics on sensitive user data. We discuss the architecture of these systems, design choices, threat models considered, and the challenges we experienced in developing production-ready systems for sensitive data analysis.
BibTeX
@misc{mailthody2021safer,
  author = {Vikram Sharma Mailthody and James Wei and Nicholas Chen and Mohammad Behnia and Ruihao Yao and Qihao Wang and Vedant Agrawal and Churan He and Lijian Wang and Leihao Chen and Amit Agarwal and Edward Richter and Wen-Mei Hwu and Christopher W. Fletcher and Jinjun Xiong and Andrew Miller and Sanjay Patel},
  title = {{Safer Illinois and RokWall: Privacy Preserving University Health Apps for COVID-19}},
  year = {2021},
  eprint = {2101.07897},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2101.07897},
}
Figure