Date of Award

6-17-2010

Document Type

Thesis

Degree Name

Master of Science

Department

Department of Electrical and Computer Engineering

First Advisor

Yong C. Kim, PhD

Abstract

This research proposes two alternative methods for generating semantically equivalent circuit variants which leave the circuit's internal structure pseudo-randomly determined. Component fusion deterministically selects subcircuits using a component identification algorithm and replaces them using a deterministic algorithm that generates canonical logic forms. Component encryption seeks to alter the semantics of individual circuit components using an encoding function, but preserves the overall circuit semantics by decoding signal values later in the circuit. Experiments were conducted to examine the performance of component fusion and component encryption against representative trials of subcircuit selection-and-replacement and Boundary Blurring, two previously defined methods for circuit obfuscation. Overall, results support the conclusion that both component fusion and component encryption generate more secure variants than previous methods and that these variants are more efficient in terms of required circuit delay and the power and area required for their implementation.

AFIT Designator

AFIT-GCO-ENG-10-15

DTIC Accession Number

Not in DTIC

Download

Share

COinS