Efficient PUF Error Correction through Response Weighting

摘要: Reliability is an important performance metric of physical unclonable function (PUF) based authentication. This paper proposes a novel methodology that incorporates error rates into PUF response correction to improve the overall performance. The proposed method first obtains circuit parameters by using machine learning techniques, which are then used estimate corresponding rates. Then, we assign bits with different degrees error-tolerance, according their estimated Response weighting algorithm determine optimal weight assignment for each bit, formulated as integer optimization problem. Experimental results show can reduce not only false negative from 20.6% 8.3% under noisy environment, but also positive rate 58% PUF-based authentication 127-bit and 13-bit correction.