Abstract
Physical Unclonable Functions (PUFs) exploit the manufacturing process variations inherent in silicon-based chips to generate unique secret keys. Although PUFs are supposed to be unclonable or unbreakable, researchers have found that they are vulnerable to machine learning (ML) attacks. In this article, we analyze the vulnerability of different FPGA-based Ring Oscillator PUFs (ROPUFs) to machine learning attacks. The challenge-response pairs (CRPs) data obtained from different ROPUFs is trained using different machine learning algorithms. From the study, it is found that the Artificial Neural Network (ANN) models can be used to train the ROPUFs with a training accuracy of 99.9% and a prediction accuracy of 62% when 5,000 CRPs are used for a \(\) challenge-response ROPUF. In this article, we assume a realistic situation where a small set of the CRP dataset (approximately 15% maximum) is unscrupulously obtained by the hacker. A prediction accuracy of 62% makes the PUF vulnerable to machine learning attacks. Therefore, a secondary goal of this article is the design of a ROPUF capable of thwarting machine learning modeling attacks. The modified XOR-inverter ROPUF drastically reduces the prediction accuracy from 62% to 13.1%, thus making it increasingly difficult for hackers to attack the ROPUF.
- Ujjwal Guin, Ke Huang, Daniel DiMase, John M. Carulli, Mohammad Tehranipoor, and Yiorgos Makris. 2014. Counterfeit integrated circuits: A rising threat in the global semiconductor supply chain. Proc, IEEE 102, 8 (2014), 1207–1228.Google ScholarCross Ref
- Peter Samarin and Kerstin Lemke-Rust. 2017. Detection of counterfeit ICs using public identification sequences. In Proceedings of the IEEE International Symposium on Hardware Oriented Security and Trust (HOST’17). IEEE, 163–163.Google ScholarCross Ref
- Swarup Bhunia, Michael S. Hsiao, Mainak Banga, and Seetharam Narasimhan. 2014. Hardware Trojan attacks: Threat analysis and countermeasures. Proc. IEEE 102, 8 (2014), 1229–1247.Google ScholarCross Ref
- Mohmmad Tehranipoor, Hassan Salmani, and Xuehui Zhang. 2014. Integrated Circuit Authentication. Springer, Cham. Google ScholarDigital Library
- Farinaz Koushanfar. 2012. Hardware metering: A survey. In Introduction to Hardware Security and Trust. Springer, New York, NY, 103–122Google Scholar
- G. Edward Suh and Srinivas Devadas. 2007. Physical unclonable functions for device authentication and secret key generation. In Proceedings of the 44th ACM/IEEE Design Automation Conference. IEEE, 9–14 Google ScholarDigital Library
- Muhtadi Choudhury, Nitin Pundir, Mohammed Niamat, and Muslim Mustapa. 2017. Analysis of a novel stage configurable ROPUF design. In Proceedings of the IEEE 60th International Midwest Symposium on Circuits and Systems (MWSCAS’17). IEEE, 942–945.Google ScholarCross Ref
- Muslim Mustapa, Mohammed Y. Niamat, Atul Prasad Deb Nath, and Mansoor Alam. 2016. Hardware-oriented authentication for advanced metering infrastructure. IEEE Trans. Smart Grid 9, 2 (2016), 1261–1270.Google ScholarCross Ref
- Abhranil Maiti and Patrick Schaumont. 2009. Improving the quality of a physical unclonable function using configurable ring oscillators. In Proceedings of the International Conference on Field Programmable Logic and Applications. IEEE, 703–707.Google ScholarCross Ref
- Fathi Amsaad, Tamzidul Hoque, and Mohammed Niamat. 2015. Analyzing the performance of a configurable ROPUF design controlled by programmable XOR gates. In Proceedings of the IEEE 58th International Midwest Symposium on Circuits and Systems (MWSCAS’15). IEEE, 1–4.Google ScholarCross Ref
- Muhtadi Choudhury, Nitin Pundir, Mohammed Niamat, and Muslim Mustapa. 2017. Analysis of a novel stage configurable ROPUF design. In Proceedings of the IEEE 60th International Midwest Symposium on Circuits and Systems (MWSCAS’17). IEEE, 942–945.Google ScholarCross Ref
- Ulrich Rührmair, Jan Sölter, Frank Sehnke, Xiaolin Xu, Ahmed Mahmoud, Vera Stoyanova, Gideon Dror, Jürgen Schmidhuber, Wayne Burleson, and Srinivas Devadas. 2013. PUF modeling attacks on simulated and silicon data. IEEE Trans. Info. Forens. Secur. 8, 11 (2013), 1876–1891. Google ScholarDigital Library
- Sharad Kumar and Mohammed Niamat. 2018. Machine learning-based modeling attacks on a configurable PUF. In Proceedings of the IEEE National Aerospace and Electronics Conference (NAECON’18). IEEE, 169–173.Google ScholarCross Ref
- Fatemeh Ganji, Shahin Tajik, and Jean-Pierre Seifert. 2015. Let me prove it to you: RO PUFs are provably learnable. In Proceedings of the Annual International Conference on Information Security and Cryptology (ICISC’15). Springer, Cham, 345–358.Google Scholar
- Noor Ahmad Hazari, Ahmed Oun, and Mohammed Niamat. 2019. Analysis and machine learning vulnerability assessment of XOR-inverter-based ring oscillator PUF design. In Proceedings of the IEEE 62nd International Midwest Symposium on Circuits and Systems (MWSCAS’19). IEEE, 590–593.Google ScholarCross Ref
- Dominik Merli, Dieter Schuster, Frederic Stumpf, and Georg Sigl. 2011. Side-channel analysis of PUFs and fuzzy extractors. In Proceedings of the International Conference on Trust and Trustworthy Computing. Springer, Berlin, 33–47. Google ScholarDigital Library
- Ganji Fatemeh, Shahin Tajik, and Jean-Pierre Seifert. 2016. PAC learning of arbiter PUFs. J. Cryptogr. Eng. 6, 3 (2016), 249–258.Google ScholarCross Ref
- Indrasish Saha, Ratan Rahul Jeldi, and Rajat Subhra Chakraborty. 2013. Model building attacks on physically unclonable functions using genetic programming. In Proceedings of the IEEE International Symposium on Hardware-Oriented Security and Trust (HOST’13). IEEE, 41–44.Google ScholarCross Ref
- Qingli Guo, Jing Ye, Yue Gong, Yu Hu, and Xiaowei Li. 2016. Efficient attack on non-linear current mirror PUF with genetic algorithm. In Proceedings of the IEEE 25th Asian Test Symposium (ATS’16). IEEE, 49–54.Google ScholarCross Ref
- Mahmoud Khalafalla and Catherine Gebotys. 2019. PUFs deep attacks: Enhanced modeling attacks using deep learning techniques to break the security of double arbiter PUFs. In Proceedings of the Design, Automation & Test in Europe Conference & Exhibition (DATE’19). IEEE, 204–209.Google ScholarCross Ref
- Jeroen Delvaux. 2019. Machine-learning attacks on PolyPUFs, OB-PUFs, RPUFs, LHS-PUFs, and PUF–FSMs. IEEE Trans. Info. Forensics Secur. 14, 8 (2019), 2043–2058. Google ScholarDigital Library
- Junye Shi, Yang Lu, and Jiliang Zhang. 2020. Approximation attacks on strong PUFs. IEEE Trans. Comput.-Aided Design Integr. Circ. Syst. 39, 10 (2020), 2138--2151. DOI:10.1109/TCAD.2019.2962115Google ScholarCross Ref
- Sahoo Durga, Phuong Nguyen, Debdeep Mukhopadhyay, Rajat Chakraborty. 2015. A case of lightweight puf constructions: Cryptanalysis and machine learning attacks. IEEE Trans. Comput.-Aided Design Integr. Circ. Syst. 34, 1334–1343. 10.1109/TCAD.2015.2448677.Google ScholarDigital Library
- Hongxiang Gu and Miodrag Potkonjak. 2018. Securing interconnected PUF network with reconfigurability. In Proceedings of the IEEE International Symposium on Hardware Oriented Security and Trust (HOST’18). IEEE, 231–234.Google ScholarCross Ref
- Masoud Rostami, Mehrdad Majzoobi, Farinaz Koushanfar, Dan S. Wallach, and Srinivas Devadas. 2014. Robust and reverse-engineering resilient PUF authentication and key-exchange by substring matching. IEEE Trans. Emerg. Topics Comput. 2, 1 (2014), 37–49.Google ScholarCross Ref
- Yansong Gao, Gefei Li, Hua Ma, Said F. Al-Sarawi, Omid Kavehei, Derek Abbott, and Damith C. Ranasinghe. 2016. Obfuscated challenge-response: A secure lightweight authentication mechanism for PUF-based pervasive devices. In Proceedings of the IEEE International Conference on Pervasive Computing and Communication Workshops (PerCom Workshops’16). IEEE, 1–6.Google Scholar
- Jiliang Zhang, Xiao Tan, Yuanjing Zhang, Weizheng Wang, and Zheng Qin. 2018. Frequency offset-based ring oscillator physical unclonable function. IEEE Trans. Multi-Scale Comput. Syst. 4, 4 (2018), 711–721.Google ScholarCross Ref
- Nitin Pundir, Noor Ahmad Hazari, Fathi Amsaad, and Mohammed Niamat. 2017. A novel hybrid delay-based physical unclonable function immune to machine learning attacks. In Proceedings of the IEEE National Aerospace and Electronics Conference (NAECON’17). IEEE, 84–87.Google ScholarCross Ref
- Haile Yu, Qiang Xu, and Philip H. W. Leong. 2010. Fine-grained characterization of process variation in FPGAs. In Proceedings of the International Conference on Field-Programmable Technology. IEEE, 138–145.Google Scholar
- Yohei Hori, Takahiro Yoshida, Toshihiro Katashita, and Akashi Satoh. 2010. Quantitative and statistical performance evaluation of arbiter physical unclonable functions on FPGAs. In Proceedings of the International Conference on Reconfigurable Computing and Field-Programmable Gate Arrays. IEEE, 298–303. Google ScholarDigital Library
- Abhranil Maiti, Vikash Gunreddy, and Patrick Schaumont. 2013. A systematic method to evaluate and compare the performance of physical unclonable functions. In Embedded Systems Design with FPGAs. Springer, New York, NY, 245–267.Google Scholar
- Dankmar Böhning. 1992. Multinomial logistic regression algorithm. Ann. Inst. Stat. Math. 44, 1 (1992), 197–200.Google ScholarCross Ref
- Sotiris B. Kotsiantis. 2013. Decision trees: A recent overview. Artific. Intell. Rev. 39, 4 (2013), 261–283. Google ScholarDigital Library
- Qingyao Wu, Yunming Ye, Haijun Zhang, Michael K. Ng, and Shen-Shyang Ho. 2014. ForesTexter: An efficient random forest algorithm for imbalanced text categorization. Knowl.-based Syst. 67 (2014), 105–116. Google ScholarDigital Library
- Thierry Denoeux. 2008. A k-nearest neighbor classification rule based on Dempster-Shafer theory. In Classic Works of the Dempster-Shafer Theory of Belief Functions. Springer, Berlin, 737–760.Google Scholar
- Corinna Cortes and Vladimir Vapnik. 1995. Support-vector networks. Mach. Learn. 20, 3 (1995), 273–297. Google ScholarDigital Library
- Tim Hill, Leorey Marquez, Marcus O’Connor, and William Remus. 1994. Artificial neural network models for forecasting and decision making. Int. J. Forecast. 10, 1 (1994), 5–15.Google ScholarCross Ref
- Diederik P. Kingma and Jimmy Ba. 2014. Adam: A method for stochastic optimization. Retrieved from https://arXiv:1412.6980.Google Scholar
- Sebastian Ruder. 2016. An overview of gradient descent optimization algorithms. Retrieved from https://arXiv:1609.04747.Google Scholar
- Martín Abadi, Ashish Agarwal, Paul Barham, Eugene Brevdo, Zhifeng Chen, Craig Citro, and Greg S. Corrado, et al. 2016. Tensorflow: Large-scale machine learning on heterogeneous distributed systems. Retrieved from https://arXiv:1603.04467.Google Scholar
- M. I. Takanori Machida, Dai Yamamoto, and K. Sakiyama. 2015. A new arbiter PUF for enhancing unpredictability on FPGA. Sci. World J. Article 864812 (2015), 13 pages. https://doi.org/10.1155/2015/864812.Google Scholar
Index Terms
- Machine Learning Vulnerability Analysis of FPGA-based Ring Oscillator PUFs and Counter Measures
Recommendations
The Gap Between Promise and Reality: On the Insecurity of XOR Arbiter PUFs
Cryptographic Hardware and Embedded Systems -- CHES 2015AbstractIn this paper we demonstrate the first real-world cloning attack on a commercial PUF-based RFID tag. The examined commercial PUFs can be attacked by measuring only 4 protocol executions, which takes less than 200 ms. Using a RFID smartcard ...
Attack Is the Best Defense: A Multi-Mode Poisoning PUF Against Machine Learning Attacks
Advances in Knowledge Discovery and Data MiningAbstractResistance to modeling attacks is an important issue for Physical Unclonable Functions (PUFs). Deep learning, the state-of-the-art modeling attack, has recently been shown to be able to break many newly developed PUFs. Since then, many more ...
A Machine Learning Attack Resistant Dual-mode PUF
GLSVLSI '18: Proceedings of the 2018 on Great Lakes Symposium on VLSISilicon Physical Unclonable Function (PUF) is arguably the most promising hardware security primitive. In particular, PUFs that are capable of generating a large amount of challenge response pairs (CRPs) can be used in many security applications. ...
Comments