Leveraging Layout-based Effects for Locking Analog ICs
September 05, 2022 ยท Declared Dead ยท ๐ ASHES@CCS
"No code URL or promise found in abstract"
Evidence collected by the PWNC Scanner
Authors
Muayad J. Aljafar, Florence Azais, Marie-Lise Flottes, Samuel Pagliarini
arXiv ID
2209.01856
Category
cs.CR: Cryptography & Security
Citations
7
Venue
ASHES@CCS
Last Checked
3 months ago
Abstract
While various obfuscation methods exist in the digital domain, techniques for protecting Intellectual Property (IP) in the analog domain are mostly overlooked. Understandably, analog components have a small footprint as most of the surface of an Integrated Circuit (IC) is digital. Yet, since they are challenging to design and tune, they constitute a valuable IP that ought to be protected. This paper is the first to show a method to secure analog IP by exploiting layout-based effects that are typically seen as undesirable detractors in IC design. Specifically, we make use of the effects of Length of Oxide Diffusion and Well Proximity Effect on transistors for tuning the devices' critical parameters (e.g., gm and Vth). Such parameters are hidden behind key inputs, akin to the logic locking approach for digital ICs. The proposed technique is applied for locking an Operational Transconductance Amplifier. In order to showcase the robustness of the achieved obfuscation, the case studied circuit is simulated for a large number of key sets, i.e., >50K and >300K, and the results show a wide range of degradation in open-loop gain (up to 130dB), phase margin (up to 50 deg), 3dB bandwidth (approx. 2.5MHz), and power (approx. 1mW) of the locked circuit when incorrect keys are applied. Our results show the benefit of the technique and the incurred overheads. We also justify the non-effectiveness of reverse engineering efforts for attacking the proposed approach. More importantly, our technique employs only regular transistors and requires neither changes to the IC fabrication process nor any foundry-level coordination or trust.
Community Contributions
Found the code? Know the venue? Think something is wrong? Let us know!
๐ Similar Papers
In the same crypt โ Cryptography & Security
R.I.P.
๐ป
Ghosted
R.I.P.
๐ป
Ghosted
Membership Inference Attacks against Machine Learning Models
R.I.P.
๐ป
Ghosted
The Limitations of Deep Learning in Adversarial Settings
R.I.P.
๐ป
Ghosted
Practical Black-Box Attacks against Machine Learning
R.I.P.
๐ป
Ghosted
Distillation as a Defense to Adversarial Perturbations against Deep Neural Networks
R.I.P.
๐ป
Ghosted
Extracting Training Data from Large Language Models
Died the same way โ ๐ป Ghosted
R.I.P.
๐ป
Ghosted
Language Models are Few-Shot Learners
R.I.P.
๐ป
Ghosted
PyTorch: An Imperative Style, High-Performance Deep Learning Library
R.I.P.
๐ป
Ghosted
XGBoost: A Scalable Tree Boosting System
R.I.P.
๐ป
Ghosted