Deep Learning based Model Building Attacks on Arbiter PUF Compositions

Robustness to modeling attacks is an important requirement for PUF circuits. Several reported Arbiter PUF compositions have resisted modeling attacks. and often require huge computational resources for successful modeling. In this paper we present deep feedforward neural network based modeling attack on 64-bit and 128-bit Arbiter PUF (APUF), and several other PUFs composed of Arbiter PUFs, namely, XOR APUF, Lightweight Secure PUF (LSPUF), Multiplexer PUF (MPUF) and its variants (cMPUF and rMPUF), and the recently proposed Interpose PUF (IPUF, up to the (4,4)-IPUF configuration). The technique requires no auxiliary information (e.g. side-channel information or reliability information), while employing deep neural networks of relatively low structural complexity to achieve very high modeling accuracy at low computational overhead (compared to previously proposed approaches), and is reasonably robust to error-inflicted training dataset.

[1]  G. Schwarz.  Estimating the Dimension of a Model , 1978 .

[2]  Debdeep Mukhopadhyay,et al.  A Case of Lightweight PUF Constructions: Cryptanalysis and Machine Learning Attacks , 2015, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems.

[3]  G. Edward Suh,et al.  Physical Unclonable Functions for Device Authentication and Secret Key Generation , 2007, 2007 44th ACM/IEEE Design Automation Conference.

[4]  François Chollet,et al.  Keras: The Python Deep Learning library , 2018 .

[5]  Kazukuni Kobara,et al.  Evaluation of Physical Unclonable Functions for 28-nm Process Field-Programmable Gate Arrays , 2014, J. Inf. Process..

[6]  Fatemeh Ganji.  PAC Learning of XOR Arbiter PUFs , 2018 .

[7]  Debdeep Mukhopadhyay,et al.  Security Analysis of Arbiter PUF and Its Lightweight Compositions Under Predictability Test , 2017, IACR Cryptol. ePrint Arch..

[8]  G. Edward Suh,et al.  Extracting secret keys from integrated circuits , 2005, IEEE Transactions on Very Large Scale Integration (VLSI) Systems.

[9]  Georg T. Becker,et al.  The Gap Between Promise and Reality: On the Insecurity of XOR Arbiter PUFs , 2015, CHES.

[10]  Masaya Yoshikawa,et al.  Deep learning attack for physical unclonable function , 2016, 2016 IEEE 5th Global Conference on Consumer Electronics.

[11]  Ulrich Rührmair,et al.  The Interpose PUF: Secure PUF Design against State-of-the-art Machine Learning Attacks , 2019, IACR Cryptol. ePrint Arch..

[12]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[13]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[14]  Mitsugu Iwamoto,et al.  Deep-Learning-Based Security Evaluation on Authentication Systems Using Arbiter PUF and Its Variants , 2016, IWSEC.

[15]  Srinivas Devadas,et al.  Silicon physical random functions , 2002, CCS '02.

[16]  Patrick Schaumont,et al.  An Analysis of Delay Based PUF Implementations on FPGA , 2010, ARC.

[17]  Kenji Kawaguchi,et al.  Deep Learning without Poor Local Minima , 2016, NIPS.

[18]  Jean-Pierre Seifert,et al.  Strong Machine Learning Attack Against PUFs with No Mathematical Model , 2016, CHES.

[19]  Mauro Conti,et al.  Trust and Trustworthy Computing , 2015, Lecture Notes in Computer Science.

[20]  Jean-Pierre Seifert,et al.  Laser Fault Attack on Physically Unclonable Functions , 2015, 2015 Workshop on Fault Diagnosis and Tolerance in Cryptography (FDTC).

[21]  Jan Sölter,et al.  Efficient Power and Timing Side Channels for Physical Unclonable Functions , 2014, CHES.

[22]  Hao Shen,et al.  Towards a Mathematical Understanding of the Difficulty in Learning with Feedforward Neural Networks , 2016, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[23]  Jean-Pierre Seifert,et al.  Why Attackers Win: On the Learnability of XOR Arbiter PUFs , 2015, TRUST.

[24]  Georg T. Becker,et al.  On the Scaling of Machine Learning Attacks on PUFs with Application to Noise Bifurcation , 2015, RFIDSec.

[25]  Jean-Pierre Seifert,et al.  Let Me Prove It to You: RO PUFs Are Provably Learnable , 2015, ICISC.

[26]  Jean-Pierre Seifert,et al.  Cloning Physically Unclonable Functions , 2013, 2013 IEEE International Symposium on Hardware-Oriented Security and Trust (HOST).

[27]  Kurt Hornik,et al.  Approximation capabilities of multilayer feedforward networks , 1991, Neural Networks.

[28]  Ulrich Rührmair,et al.  PUFs in Security Protocols: Attack Models and Security Evaluations , 2013, 2013 IEEE Symposium on Security and Privacy.

[29]  Srinivas Devadas,et al.  Modeling attacks on physical unclonable functions , 2010, CCS '10.

[30]  Patrick Schaumont,et al.  Offline Hardware/Software Authentication for Reconfigurable Platforms , 2006, CHES.

[31]  John F. Kolen,et al.  Backpropagation is Sensitive to Initial Conditions , 1990, Complex Syst..

[32]  Geoffrey E. Hinton,et al.  Rectified Linear Units Improve Restricted Boltzmann Machines , 2010, ICML.

[33]  Debdeep Mukhopadhyay,et al.  A Multiplexer-Based Arbiter PUF Composition with Enhanced Reliability and Security , 2018, IEEE Transactions on Computers.

[34]  Debdeep Mukhopadhyay,et al.  Composite PUF: A new design paradigm for Physically Unclonable Functions on FPGA , 2014, 2014 IEEE International Symposium on Hardware-Oriented Security and Trust (HOST).

[35]  Miodrag Potkonjak,et al.  Lightweight secure PUFs , 2008, ICCAD 2008.

[36]  Yoshua Bengio,et al.  Understanding the difficulty of training deep feedforward neural networks , 2010, AISTATS.

[37]  Yuan Yu,et al.  TensorFlow: A system for large-scale machine learning , 2016, OSDI.

[38]  Stefan Katzenbeisser,et al.  PUFs: Myth, Fact or Busted? A Security Evaluation of Physically Unclonable Functions (PUFs) Cast in Silicon , 2012, CHES.

[39]  Yoshua Bengio,et al.  Deep Sparse Rectifier Neural Networks , 2011, AISTATS.

[40]  Yu Zhuang,et al.  A Machine Learning-Based Security Vulnerability Study on XOR PUFs for Resource-Constraint Internet of Things , 2018, 2018 IEEE International Congress on Internet of Things (ICIOT).