Accession Number : AD1041821

Title :   Automatic Feature Selection and Improved Classification in SICADA Counterfeit Electronics Detection

Descriptive Note : Conference Paper

Corporate Author : Defense Microelectronics Activity (DMEA) McClellan United States

Personal Author(s) : Milechin,Lauren ; Koziel,Eric ; Vai,Michael ; Bergevin,Keith ; Comer,Phil

Full Text :

Report Date : 20 Mar 2017

Pagination or Media Count : 4

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.

Descriptors :   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

Subject Categories : Electrical and Electronic Equipment

Distribution Statement : APPROVED FOR PUBLIC RELEASE