Advertisement

On the Learnability of Physically Unclonable Functions

  • Fatemeh Ganji

Part of the T-Labs Series in Telecommunication Services book series (TLABS)

Table of contents

  1. Front Matter
    Pages i-xxiv
  2. Fatemeh Ganji
    Pages 1-8
  3. Fatemeh Ganji
    Pages 9-20
  4. Fatemeh Ganji
    Pages 21-34
  5. Fatemeh Ganji
    Pages 35-47
  6. Fatemeh Ganji
    Pages 49-57
  7. Fatemeh Ganji
    Pages 59-72
  8. Fatemeh Ganji
    Pages 73-75
  9. Fatemeh Ganji
    Pages 77-79
  10. Back Matter
    Pages 81-86

About this book

Introduction

This book addresses the issue of Machine Learning (ML) attacks on Integrated Circuits through Physical Unclonable Functions (PUFs). It provides the mathematical proofs of the vulnerability of various PUF families, including Arbiter, XOR Arbiter, ring-oscillator, and bistable ring PUFs, to ML attacks. To achieve this goal, it develops a generic framework for the assessment of these PUFs based on two main approaches. First, with regard to the inherent physical characteristics, it establishes fit-for-purpose mathematical representations of the PUFs mentioned above, which adequately reflect the physical behavior of these primitives. To this end, notions and formalizations that are already familiar to the ML theory world are reintroduced in order to give a better understanding of why, how, and to what extent ML attacks against PUFs can be feasible in practice. Second, the book explores polynomial time ML algorithms, which can learn the PUFs under the appropriate representation. More importantly, in contrast to previous ML approaches, the framework presented here ensures not only the accuracy of the model mimicking the behavior of the PUF, but also the delivery of such a model.

 

Besides off-the-shelf ML algorithms, the book applies a set of algorithms hailing from the field of property testing, which can help to evaluate the security of PUFs. They serve as a “toolbox”, from which PUF designers and manufacturers can choose the indicators most relevant for their requirements. Last but not least, on the basis of learning theory concepts, the book explicitly states that the PUF families cannot be considered as an ultimate solution to the problem of insecure ICs. As such, it provides essential insights into both academic research on and the design and manufacturing of PUFs.​

Keywords

Physically Unclonable Function PUF PUF vulnerability ML attack XOR Arbiter ring-oscillator bistable ring PUFs Learnable PUFs physical security primitives PAC learning of PUFs PUF circuit design

Authors and affiliations

  • Fatemeh Ganji
    • 1
  1. 1.Institut für Softwaretechnik und Theoretische InformatikTechnischen Universität BerlinBerlinGermany

Bibliographic information

  • DOI https://doi.org/10.1007/978-3-319-76717-8
  • Copyright Information Springer International Publishing AG, part of Springer Nature 2018
  • Publisher Name Springer, Cham
  • eBook Packages Engineering
  • Print ISBN 978-3-319-76716-1
  • Online ISBN 978-3-319-76717-8
  • Series Print ISSN 2192-2810
  • Series Online ISSN 2192-2829
  • Buy this book on publisher's site