Authors: Norrathep Rattanavipanon, Ivan de Oliviera Nunes

Published on: April 10, 2024

Impact Score: 7.2

Arxiv code: Arxiv:2404.06721

Summary

• What is new: The introduction of Proofs of Stateful Execution (PoSX) combined with the use of ARM TrustZoneM for IoT/embedded devices in the context of Federated Learning (FL) and Local Differential Privacy (LDP).
• Why this is important: The vulnerability of Federated Learning and Local Differential Privacy mechanisms to poisoning attacks due to compromised edge devices.
• What the research proposes: A system-level approach, SLAPP, that leverages ARM TrustZoneM and PoSX to securely verify the integrity of data collected and processed by IoT devices in FL/LDP settings.
• Results: SLAPP demonstrates robust security against poisoning attacks with minimal performance overhead, validated through real-world prototypes.

Technical Details

Technological frameworks used: SLAPP (System-Level Approach for Poisoning Prevention)

Models used: Proofs of Stateful Execution (PoSX)

Data used: ARM TrustZoneM security extensions

Potential Impact

IoT security solutions, privacy-preserving data analytics platforms, companies specializing in Federated Learning and Local Differential Privacy

Want to implement this idea in a business?

We have generated a startup concept here: SecureFleet.