http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
박서현(Seo Hyeon Park),백창재(Changjae Beak),이정원(Jungwon Lee),선우경(Wookyoung Sun) 대한전자공학회 2022 대한전자공학회 학술대회 Vol.2022 No.11
As IoT technology has developed, security becomes important. However, there are weaknesses in software security, therefore hardware-based Physical Unclonable Function (PUF) has been proposed. PUF, which generates challenge-response pairs, has a disadvantage it is vulnerable to machine learning attacks. Therefore, we propose a structure with a feedback process by mimicking a multilayer neural network to prepare for machine learning attacks and reuse the state distribution. As a result, the prediction accuracy of machine learning attacks using CRP decreased by 3.1%, 6%, and 3.1%, respectively, in both LR, SVM, and MLP compared to MC-XAT.