SRAMPUF
Digitized Anything
Consortium: TalTech – Estonia /
Funded by: Internal Experiment
The increasing use of Internet of Things (IoT) devices has posed a significant challenge for cybersecurity. It is imperative to ensure that IoT devices are thoroughly assessed for their provenance to prevent potential security threats. Physically unclonable functions (PUFs) are considered to be ideal for reliable key-generation, as they generate strong hardware-generated signatures, which are far more secure than software-generated ones. PUFs can create unique signatures by leveraging inherent manufacturing variations.
TalTech has successfully designed a chip from TSMC, the world's largest silicon foundry, to validate the robustness of SRAM-based PUFs. To validate the design, 50 chip samples were packaged, and all packaged samples were confirmed to be fully functional. The total number of datapoints collected is 131M. After collecting the data from the chip, the technical outcome is analyzing the statistical data from the chips. Reliability, entropy, uniqueness, randomness, and bias pattern were considered during the analysis.
The aim of the technology transfer from Tallinn University of Technology to Intrinsic ID was to find appropriate Physical Unclonable Functions (PUFs) for the chips that are used in Internet of Things (IoT) devices. To achieve this, TalTech conducted a study to investigate the design-time decisions that ensured the successful implementation of SRAM-based PUFs. The study had specific objectives, which are outlined below:
- Design-time decisions could affect the robustness of the SRAM-based PUF, therefore TalTech evaluated how various memory-specific characteristics such as aspect ratio, size, density, and memory type influence the generation and stability of signatures.
- The study focused on finding the most effective way to obtain dependable signatures from the memories present in these devices while minimizing the overheads associated with specialized signature generator circuitry.
- Cloud service providers (CSPs) like Amazon AWS, Microsoft Azure, Google Cloud, and Alibaba Cloud provide IoT platforms that enable the integration of new edge IoT devices with their cloud services.
- For the most secure end-to-end encryption, it is essential to have strong and unique keys that form a robust root-of-trust.
- SRAM-based PUFs are simple and scalable but, the design time decisions affect the robustness of PUF which is crucial factor to consider.
- Although SRAM-based PUFs are highly reliable, evaluating their robustness could help in designing error correction algorithms.
TalTech team successfully created a chip using 65nm technology and put it to the test. TalTech team was able to assess the robustness of the SRAM-based PUF. This project was a breakthrough in chip design and PUF's reliability, as it merged the different technological aspects to produce an innovative solution which are given below:
- IoT technologies
- ASIC design
- SRAM-based PUFs
- Trustworthy hardware
- VLSI Circuits
- Cryptography
SMART4ALL has received funding from the European Union’s Horizon 2020 research and innovation programme under Grant Agreement No 872614
SMART4ALL is a four-year Innovation Action project funded under Horizon 2020 framework under call DT-ICT-01-2019: Smart Anything Everywhere – Area 2: Customized low energy computing powering CPS and the IoT.