The Past, Present, and Future of Physical Security Enclosures: From Battery-Backed Monitoring to PUF-Based Inherent Security and Beyond

Abstract

Withstanding physical attacks in a hostile environment is of utmost importance for nowadays electronics. However, due to the long and costly development of integrated circuits (ICs), IC-level countermeasures are typically only included in varying degree and not in every chip of a device. Therefore, multiple-chip modules requiring higher levels of security are additionally protected against tampering by a physical security enclosure, e.g., by an envelope that completely encloses the device. For decades, these physical boundaries on a device-level were monitored using battery-backed mechanisms to enable detection of an attempted physical intrusion even if the underlying system is powered off. However, the battery affects the system’s robustness, weight, prevents extended storage, and also leads to difficulties with the security mechanism while shipping the device. In this position paper, we present our assessment of various battery-backed tamper-respondent solutions and argue that while offering the intriguing benefit of instantaneous detection and response, the low-power nature of battery-backup contradicts a tamper-sensitive measurement, among other problems. We are therefore of the opinion that more effort should be spent towards enclosures that are based on tamper-evident physical unclonable functions (PUFs), as they are designated to provide a high level of security on the one hand and do not require a battery on the other hand. To further substantiate our argument, we summarize the work in this domain to also facilitate future research.