Soft Response Generation and Thresholding Strategies for Linear and Feed-Forward MUX PUFs

Size: px

Start display at page:

Download "Soft Response Generation and Thresholding Strategies for Linear and Feed-Forward MUX PUFs"

Dorthy Stewart
5 years ago
Views:

1 Soft Response Generation and Thresholding Strategies for Linear and Feed-Forward MUX PUFs Chen Zhou, SarojSatapathy, YingjieLao, KeshabK. Parhiand Chris H. Kim Department of ECE University of Minnesota International Symposium on Low Power Electronics and Design

2 Outline Physical Unclonable Function (PUF) 32nm PUF Chip Measurements Soft Response Thresholding Strategies Linear PUF vs. Feed-forward PUF Conclusion 2

3 Physical Unclonable Function (PUF) Fingerprint of chip PUF Inputs Challenge # Challenge #2 Numerous input choices Outputs Response # Response #2 Challenge #n Response #n Unique and random: Based on inherent process variation Secure: Large # of challenge-response pairs (CRPs) Unique and random responses Challenges Chip # Chip #2 Responses # Responses #2 3

4 Typical Authentication Process Server User Database Store CRPs of all PUFs Public ID Chip Private PUF Server-user based authentication Challenge-response pairs tested and stored before usage 4

5 Typical Authentication Process Public chip ID is first sent to the server 5

6 Typical Authentication Process Server User Database Retrieve CRP subset ID PUF Chip Challenge 209A (Hex) 4B (Hex) Response (Hex) Server retrieves CRP subset table for the given chip ID 6

7 Typical Authentication Process Server User Database Retrieve CRP subset Challenges ID PUF Chip Challenge 209A (Hex) 4B (Hex) Response (Hex) Challenges are sent to the user 7

8 Typical Authentication Process Server User Database ID PUF Chip Challenge 209A (Hex) 4B (Hex) Response 0 Response (Hex) User generates responses using PUF circuit 8

9 Typical Authentication Process Server User Database ID Chip PUF Challenge Response Response 209A (Hex) 4B (Hex) 0 Compare Responses (Hex) User responses are sent to server for comparison 9

10 Typical Authentication Process Server User Database ID Chip PUF Challenge Response Response 209A (Hex) 0 Compare 0 4B (Hex) match mismatch 9283 (Hex) Approved Denied Decision Approved if responses match; denied if mismatch Final step: decision sent to user 0

11 Hamming Distance (HD) Calculation Hamming distance can be used as matching criteria Intra-chip HD: Same chip, noise effects, close to 0% Inter-chip HD: Different chip, process variation effects, close to 50%

12 Outline Physical Unclonable Function (PUF) 32nm PUF Chip Measurements Soft Response Thresholding Strategies Linear PUF vs. Feed-forward PUF Conclusion 2

13 Motivation of This Work Ideal case (all CRPs) Intra-chip Intra-chip Actual case (all CRPs) PDF No overlap Inter-chip PDF Overlap Inter-chip Hamming distance Hamming distance.0 Stable CRPs have less intra-chip variation Measure soft response (=probability of response being or 0 ) to find stable CRPs Actual case (only stable CRPs) PDF 0.0 Intra-chip No overlap 0.5 Hamming distance Inter-chip.0 3

14 Contributions of This Work Implemented soft response collection circuits in a 32nm test chip Generated MUX PUF soft response distribution based on 3.3 Gb test data Proposed soft response thresholding strategies to select stable challenge-response pairs Implemented and characterized feed-forward MUX PUF 4

15 Proposed Soft Response Measurement Circuit Soft response = response probability information >GHz sampling circuits facilitate efficient soft response measurements 5

16 Linear MUX PUF Delay Stages PUF stage (c='') 0 PUF stage (c='0') Parallel or crossed signal paths configured by challenge bits Delay difference determined by inherent process variation 6

17 Arbiter Circuit 32 MUX stages Arbiter Response Onchip VCO c c2 c32 Δ S Q SR Latch R On-chip counter On-chip counter M N Arbiter generates response bit based on delay difference 7

18 32nm PUF Test Chip 8

19 Soft Response Measurements Soft response is a function of the actual delay difference Above distribution generated using 3.3 Gb of PUF response data 9

20 Outline Physical Unclonable Function (PUF) 32nm PUF Chip Measurements Soft Response Thresholding Strategies Linear PUF vs. Feed-forward PUF Conclusion 20

21 Soft Response Thresholding Strategy Symmetric thresholds used to define stable and unstable CRPs Unstable CRPs not used for authentication 2

22 Impact of Soft Response (SR) Threshold Left: HD distributions overlap when threshold=0.5 Right: No overlap when threshold=0 and (i.e. only stable responses are used) 22

23 Fixed Threshold Scheme No stable to stable 0 flips when threshold > 0.8 Stable to unstable flips always exist, necessitating more tests to find stable CRPs 23

24 Relaxed Threshold Scheme Stringent threshold during enrollment phase and relaxed threshold during authentication Results in fewer unstable and 0 unstable flips 24

25 Outline Physical Unclonable Function (PUF) 32nm PUF Chip Measurements Soft Response Thresholding Strategies Linear PUF vs. Feed-forward PUF Conclusion 25

26 Linear MUX PUF Vulnerability 32 MUX stages Arbiter Δ S Q SR Latch R Response = c c2 c32 2 )(2 ) (2 (2 ) (2 ) (2 ) = = =( +) 2 Linear PUFs are susceptible to modelling attack That is, attacker can predict correct response with very high probability using past CRP data 26

27 Feed-forward MUX PUF for Improved Security Use intermediate response for some challenge bits Non-linear relationship between delay and response harder for attacker to predict correct response No experimental data reported on feed-forward PUF Feed-forward MUX PUF ref.: J. W. Lee, et al., VLSI Circuits Symposium,

28 32nm Test Chip Data: Linear vs. Feedforward MUX PUF % of stable CRPs decreases from 94.6% to 9.02% due to instability of internal challenge bit 28

29 Conclusion Soft response measurement circuit demonstrated in a 32nm test chip On-chip VCO and counters enable fast measurement Different thresholding strategies evaluated Enables robust authentication across wider voltage and temperature range Feed-forward MUX measured for the first time % of stable CRPs decreases slightly due to instability of internal challenge bit Acknowledgements National Science Foundation and Semiconductor Research Corporation for funding 29

A Physical Unclonable Function based on Capacitor Mismatch in a Charge-Redistribution SAR-ADC

A Physical Unclonable Function based on Capacitor Mismatch in a Charge-Redistribution SAR-ADC Qianying Tang, Won Ho Choi, Luke Everson, Keshab K. Parhi and Chris H. Kim University of Minnesota Department