Team ECE Receives Best Paper Award at ISLPED 2017

A paper presented by ECE graduate student Muqing Liu, and co-authored by an all-ECE team (Muqing Liu, Chen Zhou, Qianying Tang, Prof. Keshab Parhi, and Prof. Chris Kim) has been selected for the best paper award at the 2017 International Symposium on Low Power Electronics and Design (ISLPED), a key conference on low power circuits and systems.

Titled “A Data Remanence based Approach to Generate 100% Stable Keys from an SRAM Physical Unclonable Function,” the paper’s authors have been recognized with the best paper award two years in a row, which is rather unusual for this conference. The paper proposes a method that is more effective in finding the most stable cells in a large SRAM array.

SRAM is an attractive circuit for generating encryption keys because the start-up value of an SRAM cell is unique, random, and unclonable. However, the primary challenge for SRAM based key generation, however, is poor stability when the circuit is subject to random noise, temperature and voltage changes, and device aging. Earlier papers have suggested the use of temporal majority voting (TMV) and bit masking to identify and store the location of unstable or marginally stable SRAM cells. However, TMV requires a long test time and significant hardware resources. In addition, the number of repetitive power-ups required to find the most stable cells is prohibitively high.

The authors developed a one-time highly efficient data remanence based test method that can identify the most stable bits in an SRAM array for reliable key generation. The approach only requires two tests to identify all the stable ‘0’s and ‘1’s, which is more efficient than the popular temporal majority voting (TMV) technique. Experimental studies have shown that the 256-bit key generated from a 512 kbit SRAM using the proposed data remanence method is 100% stable under different temperatures, power ramp up times, and device aging.

The work was jointly funded by the National Science Foundation and the Semiconductor Research Corporation.