Large scale characterization of SRAM on infineon XMC microcontrollers as PUF

SRAM-based physical unclonable functions (SRAM PUFs) derive a device dependent secret from the start-up pattern of their memory cells and have shown very promising results in previous publications. This work presents a dataset measured on 144 Infineon XMC4500 microcontrollers containing 160 KiB of SRAM sampled 101 times each in 2015 and 2016. Analyses are done using state-of-the-art metrics by Maiti et al., Hori et al., and by custom inspections. In extensive comparison to previous work, this work is found to score best in average Reliability and Bit-Alias, match with previous top results in average Uniformity and still mid-range in Uniqueness. This confirms previous results that general purpose SRAM on microcontrollers is adequate for most PUF applications. To support further research into SRAM PUFs and their post-processing, the full dataset originating from this work will be made publicly available on the internet.

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

[2]  Tony Tae-Hyoung Kim,et al.  Design of SRAM PUF with improved uniformity and reliability utilizing device aging effect , 2014, 2014 IEEE International Symposium on Circuits and Systems (ISCAS).

[3]  Stefan Katzenbeisser,et al.  Secure PRNG seeding on commercial off-the-shelf microcontrollers , 2013, TrustED '13.

[4]  Jorge Guajardo,et al.  FPGA Intrinsic PUFs and Their Use for IP Protection , 2007, CHES.

[5]  Matthias Hiller,et al.  Systematic Low Leakage Coding for Physical Unclonable Functions , 2015, AsiaCCS.

[6]  Ying Su,et al.  A Digital 1.6 pJ/bit Chip Identification Circuit Using Process Variations , 2008, IEEE Journal of Solid-State Circuits.

[7]  Said Hamdioui,et al.  Design dependent SRAM PUF robustness analysis , 2015, 2015 16th Latin-American Test Symposium (LATS).

[8]  Akashi Satoh,et al.  Quantitative and Statistical Performance Evaluation of Arbiter Physical Unclonable Functions on FPGAs , 2010, 2010 International Conference on Reconfigurable Computing and FPGAs.

[9]  Daniel J. Bernstein,et al.  Investigating SRAM PUFs in large CPUs and GPUs , 2015, SPACE.

[10]  Roel Maes,et al.  Physically Unclonable Functions , 2013, Springer Berlin Heidelberg.

[11]  Arenberg Doctoral,et al.  Physically Unclonable Functions: Constructions, Properties and Applications , 2012 .

[12]  Boris Skoric,et al.  Estimating the Secrecy-Rate of Physical Unclonable Functions with the Context-Tree Weighting Method , 2006, 2006 IEEE International Symposium on Information Theory.

[13]  Ahmad-Reza Sadeghi,et al.  On the Effectiveness of the Remanence Decay Side-Channel to Clone Memory-Based PUFs , 2013, CHES.

[14]  Patrick Schaumont,et al.  A Systematic Method to Evaluate and Compare the Performance of Physical Unclonable Functions , 2011, IACR Cryptol. ePrint Arch..

[15]  Daniel E. Holcomb,et al.  Initial SRAM State as a Fingerprint and Source of True Random Numbers for RFID Tags , 2007 .

[16]  Roel Maes.  Physically Unclonable Functions: Constructions, Properties and Applications (Fysisch onkloonbare functies: constructies, eigenschappen en toepassingen) , 2012 .

[17]  Patrick Schaumont,et al.  A large scale characterization of RO-PUF , 2010, 2010 IEEE International Symposium on Hardware-Oriented Security and Trust (HOST).

[18]  Antonino Mazzeo,et al.  Testing 90 nm microcontroller SRAM PUF quality , 2015, 2015 10th International Conference on Design & Technology of Integrated Systems in Nanoscale Era (DTIS).

[19]  Tolga Arul,et al.  Lightweight Anti-counterfeiting Solution for Low-End Commodity Hardware Using Inherent PUFs , 2014, TRUST.