Automatic Feature Selection and Improved Classification in SICADA Counterfeit Electronics Detection

reportActive / Technical Report | Accession Number: AD1041821 | Open PDF

Abstract:

Counterfeiters seeking financial gain can introduce misrepresented or recycled microelectronic components to both government and commercial supply chains. This reduces system reliability and trust, and currently has few comprehensive and practical solutions. The SICADA methodology was developed to detect such counterfeit microelectronics by collecting power side channel data and applying machine learning to identify counterfeits. This methodology has been extended to include a two-step automated feature selection process and now uses a one-class SVM classifier. We describe this methodology and show results for empirical data collected from several types of Microchip dsPIC33F microcontrollers.

Author(s):
Milechin, Lauren ; Koziel, Eric ; Vai, Michael ; Bergevin, Keith ; Comer, Phil
Author Organization(s):
Defense Microelectronics Activity (DMEA) McClellan United States
Descriptive Note:
Conference Paper
Supplementary Note:
42nd Annual Gomactech Conference (GOMACTech 17) Government Microcircuit Applications and Critical Technology Conference, 20 Mar 2017, 23 Mar 2017, See also AD1034309 - 42nd Annual Gomactech Conference (GOMACTech 17) Government Microcircuit Applications And Critical Technology Conference. Technologies For Secure Spectrum Access From Dc To Light. Held At The Grand Sierra Resort, Reno, Nevada, March 20-23, 2017.
Pagination:
0004

Security Markings

DOCUMENT & CONTEXTUAL SUMMARY

Distribution:
Approved For Public Release
Distribution Statement:
Approved For Public Release;

RECORD

Collection: TR
Identifying Numbers
Contract/Grant Number(s):
FA8721-05-C-0002, FA8702-15-D-0001
 Subject Terms
Joint Capability Areas:
JCA_5_Command and Control; JCA_5.4_Decide; JCA_1.2_Force Preparation; JCA_1_Force Support; JCA_5.3_Planning; JCA_1.2.1_Training; JCA_1.2.5_Lessons Learned
Modernization Areas:
AI and Machine Learning; Autonomy
Communities of Interest:
Energy and Power Technologies
Descriptor(s):
test and evaluation, feature selection, change detection, machine learning, discriminate analysis, random variables, reliability, measurement, pulse generators, calibration, coefficients, classification, life cycle management, supply chain management
Keyword(s):
counterfeit detection, device authentication, supply chain security, hardware identity
Report Date:
2017 Mar 20
Creation Date:
2017 Nov 28