Stacked hardware controller enhances Edge AI security
Original title: Enhancing Edge AI System Security with a Stacked Hardware Security Controller for Robust and Energy-Efficient Anomaly Detectio
A stacked hardware security controller enables low-overhead anomaly detection in Edge AI systems using performance counters.
Edge AI systems encounter security threats from side-channel leakage and adversarial manipulation. Conventional software tools fail due to resource limits. The proposed mechanism provides continuous detection with tamper-resistant isolation and negligible overhead, resilient to compromised OS.
- Hardware performance counters detect anomalous patterns
- Tamper-resistant root-of-trust isolation
- Negligible runtime overhead
- Resilient against compromised operating systems
- Observable deviations in sensor and thermal readings
From the original article
1 min read at arxiv.orgIn recent years, the growing adoption of Edge AI systems has been accompanied by a surge in security threats. Edge AI deployments face an rapidly evolving spectrum of security threats, from side-channel information leakage to ad- versarial model manipulation and physical device tampering.
By Carl Benda · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.