Integrated Circuit Wear out Prediction and Recycling Detection using Radio Frequency Distinct Native Attribute Features
Abstract
Radio Frequency Distinct Native Attribute (RF-DNA) has shown promise for detecting differences in Integrated Circuits(IC) using features extracted from a devices Unintentional Radio Emissions (URE). This ability of RF-DNA relies upon process variation imparted to a semiconductor device during manufacturing. However, internal components in modern ICs electronically age and wear out over their operational lifetime. RF-DNA techniques are adopted from prior work and applied to MSP430 URE to address the following research goals: 1) Does device wear-out impact RF-DNA device discriminability?, 2) Can device age be continuously estimated by monitoring changes in RF-DNA features?, and 3) Can device age state (e.g., new vs. used) be reliably estimated? Conclusions include: 1) device wear-out does impact RF-DNA, with up to a 16 change in discriminability over the range of accelerated ages considered, 2) continuous(hour-by-hour) age estimation was most challenging and generally not supported, and 3) binary new vs. used age estimation was successful with 78.7 to 99.9 average discriminability for all device-age combinations considered.
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 22, 2016
- Accession Number
- AD1031986
Entities
People
- Randall D Deppensmith
Organizations
- Air Force Institute of Technology