A Lockdown Technique to Prevent Machine Learning on PUFs for Lightweight Authentication
We present a lightweight PUF-based authentication approach that is practical in settings where a server authenticates a device, and for use cases where the number of authentications is limited over a device's lifetime. Our scheme uses a server-managed challenge/response pair (CRP) lockdown prot...
| Main Authors: | , , , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | en_US |
| Published: |
Institute of Electrical and Electronics Engineers (IEEE)
2017
|
| Online Access: | http://hdl.handle.net/1721.1/110985 https://orcid.org/0000-0001-8253-7714 |
| _version_ | 1826215141259083776 |
|---|---|
| author | Hiller, Matthias Delvaux, Jeroen Sowell, Richard Verbauwhede, Ingrid Yu, Meng-Day Devadas, Srinivas |
| author2 | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory |
| author_facet | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Hiller, Matthias Delvaux, Jeroen Sowell, Richard Verbauwhede, Ingrid Yu, Meng-Day Devadas, Srinivas |
| author_sort | Hiller, Matthias |
| collection | MIT |
| description | We present a lightweight PUF-based authentication approach that is practical in settings where a server authenticates a device, and for use cases where the number of authentications is limited over a device's lifetime. Our scheme uses a server-managed challenge/response pair (CRP) lockdown protocol: unlike prior approaches, an adaptive chosen-challenge adversary with machine learning capabilities cannot obtain new CRPs without the server's implicit permission. The adversary is faced with the problem of deriving a PUF model with a limited amount of machine learning training data. Our system-level approach allows a so-called strong PUF to be used for lightweight authentication in a manner that is heuristically secure against today's best machine learning methods through a worst-case CRP exposure algorithmic validation. We also present a degenerate instantiation using a weak PUF that is secure against computationally unrestricted adversaries, which includes any learning adversary, for practical device lifetimes and read-out rates. We validate our approach using silicon PUF data, and demonstrate the feasibility of supporting 10, 1,000, and 1M authentications, including practical configurations that are not learnable with polynomial resources, e.g., the number of CRPs and the attack runtime, using recent results based on the probably-approximately-correct (PAC) complexity-theoretic framework. |
| first_indexed | 2024-09-23T16:17:13Z |
| format | Article |
| id | mit-1721.1/110985 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T16:17:13Z |
| publishDate | 2017 |
| publisher | Institute of Electrical and Electronics Engineers (IEEE) |
| record_format | dspace |
| spelling | mit-1721.1/1109852022-10-02T07:35:32Z A Lockdown Technique to Prevent Machine Learning on PUFs for Lightweight Authentication Hiller, Matthias Delvaux, Jeroen Sowell, Richard Verbauwhede, Ingrid Yu, Meng-Day Devadas, Srinivas Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Yu, Meng-Day Devadas, Srinivas We present a lightweight PUF-based authentication approach that is practical in settings where a server authenticates a device, and for use cases where the number of authentications is limited over a device's lifetime. Our scheme uses a server-managed challenge/response pair (CRP) lockdown protocol: unlike prior approaches, an adaptive chosen-challenge adversary with machine learning capabilities cannot obtain new CRPs without the server's implicit permission. The adversary is faced with the problem of deriving a PUF model with a limited amount of machine learning training data. Our system-level approach allows a so-called strong PUF to be used for lightweight authentication in a manner that is heuristically secure against today's best machine learning methods through a worst-case CRP exposure algorithmic validation. We also present a degenerate instantiation using a weak PUF that is secure against computationally unrestricted adversaries, which includes any learning adversary, for practical device lifetimes and read-out rates. We validate our approach using silicon PUF data, and demonstrate the feasibility of supporting 10, 1,000, and 1M authentications, including practical configurations that are not learnable with polynomial resources, e.g., the number of CRPs and the attack runtime, using recent results based on the probably-approximately-correct (PAC) complexity-theoretic framework. 2017-08-18T17:43:49Z 2017-08-18T17:43:49Z 2016-04 Article http://purl.org/eprint/type/JournalArticle 2332-7766 http://hdl.handle.net/1721.1/110985 Yu, Meng-Day, et al. “A Lockdown Technique to Prevent Machine Learning on PUFs for Lightweight Authentication.” IEEE Transactions on Multi-Scale Computing Systems 2, 3 (July 2016): 146–159 © 2016 Institute of Electrical and Electronics Engineers (IEEE) https://orcid.org/0000-0001-8253-7714 en_US http://dx.doi.org/10.1109/TMSCS.2016.2553027 IEEE Transactions on Multi-Scale Computing Systems Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf Institute of Electrical and Electronics Engineers (IEEE) MIT Web Domain |
| spellingShingle | Hiller, Matthias Delvaux, Jeroen Sowell, Richard Verbauwhede, Ingrid Yu, Meng-Day Devadas, Srinivas A Lockdown Technique to Prevent Machine Learning on PUFs for Lightweight Authentication |
| title | A Lockdown Technique to Prevent Machine Learning on PUFs for Lightweight Authentication |
| title_full | A Lockdown Technique to Prevent Machine Learning on PUFs for Lightweight Authentication |
| title_fullStr | A Lockdown Technique to Prevent Machine Learning on PUFs for Lightweight Authentication |
| title_full_unstemmed | A Lockdown Technique to Prevent Machine Learning on PUFs for Lightweight Authentication |
| title_short | A Lockdown Technique to Prevent Machine Learning on PUFs for Lightweight Authentication |
| title_sort | lockdown technique to prevent machine learning on pufs for lightweight authentication |
| url | http://hdl.handle.net/1721.1/110985 https://orcid.org/0000-0001-8253-7714 |
| work_keys_str_mv | AT hillermatthias alockdowntechniquetopreventmachinelearningonpufsforlightweightauthentication AT delvauxjeroen alockdowntechniquetopreventmachinelearningonpufsforlightweightauthentication AT sowellrichard alockdowntechniquetopreventmachinelearningonpufsforlightweightauthentication AT verbauwhedeingrid alockdowntechniquetopreventmachinelearningonpufsforlightweightauthentication AT yumengday alockdowntechniquetopreventmachinelearningonpufsforlightweightauthentication AT devadassrinivas alockdowntechniquetopreventmachinelearningonpufsforlightweightauthentication AT hillermatthias lockdowntechniquetopreventmachinelearningonpufsforlightweightauthentication AT delvauxjeroen lockdowntechniquetopreventmachinelearningonpufsforlightweightauthentication AT sowellrichard lockdowntechniquetopreventmachinelearningonpufsforlightweightauthentication AT verbauwhedeingrid lockdowntechniquetopreventmachinelearningonpufsforlightweightauthentication AT yumengday lockdowntechniquetopreventmachinelearningonpufsforlightweightauthentication AT devadassrinivas lockdowntechniquetopreventmachinelearningonpufsforlightweightauthentication |