{"title":"RX-PUF:基于模拟无源RRAM交叉棒阵列的低功耗、密集、可靠和弹性物理不可克隆功能","authors":"M. Mahmoodi, H. Nili, D. Strukov","doi":"10.1109/VLSIT.2018.8510624","DOIUrl":null,"url":null,"abstract":"We propose a novel architecture (\"RX-PUF\") for physically unclonable functions (PUF) based on analog RRAM crossbar array circuits. RX-PUF takes advantage of unique RRAM properties, such as I-V nonlinearity, and its device-to-device (d2d) variations and tunability. As a proof of concept, we have prototyped a 600 kb challenge response pair (CRP) PUF using 250 nm half-pitch (F) 20×20 crossbar arrays with passively integrated devices. The RX-PUF prototype features excellent physical characteristics, e.g. ~1600 F2/bit density and up to 41 fJ/bit energy efficiency. Its functional performance, improved by utilizing hidden input, is also very promising. The measured bit error rate (BER) was 0.7% at RT and ≤ 5.3% at 100°C, even without using any error correction methods. The measured responses showed near-ideal uniformity (50.04%) and inter-HD (50.12%) and passed all relevant NIST randomness tests. The preliminary results showed also very high resilience of RX-PUF against machine learning (ML) attacks.","PeriodicalId":6561,"journal":{"name":"2018 IEEE Symposium on VLSI Technology","volume":"126 1","pages":"99-100"},"PeriodicalIF":0.0000,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"14","resultStr":"{\"title\":\"RX-PUF: Low Power, Dense, Reliable, and Resilient Physically Unclonable Functions Based on Analog Passive RRAM Crossbar Arrays\",\"authors\":\"M. Mahmoodi, H. Nili, D. Strukov\",\"doi\":\"10.1109/VLSIT.2018.8510624\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"We propose a novel architecture (\\\"RX-PUF\\\") for physically unclonable functions (PUF) based on analog RRAM crossbar array circuits. RX-PUF takes advantage of unique RRAM properties, such as I-V nonlinearity, and its device-to-device (d2d) variations and tunability. As a proof of concept, we have prototyped a 600 kb challenge response pair (CRP) PUF using 250 nm half-pitch (F) 20×20 crossbar arrays with passively integrated devices. The RX-PUF prototype features excellent physical characteristics, e.g. ~1600 F2/bit density and up to 41 fJ/bit energy efficiency. Its functional performance, improved by utilizing hidden input, is also very promising. The measured bit error rate (BER) was 0.7% at RT and ≤ 5.3% at 100°C, even without using any error correction methods. The measured responses showed near-ideal uniformity (50.04%) and inter-HD (50.12%) and passed all relevant NIST randomness tests. The preliminary results showed also very high resilience of RX-PUF against machine learning (ML) attacks.\",\"PeriodicalId\":6561,\"journal\":{\"name\":\"2018 IEEE Symposium on VLSI Technology\",\"volume\":\"126 1\",\"pages\":\"99-100\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2018-06-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"14\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2018 IEEE Symposium on VLSI Technology\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/VLSIT.2018.8510624\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2018 IEEE Symposium on VLSI Technology","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/VLSIT.2018.8510624","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
RX-PUF: Low Power, Dense, Reliable, and Resilient Physically Unclonable Functions Based on Analog Passive RRAM Crossbar Arrays
We propose a novel architecture ("RX-PUF") for physically unclonable functions (PUF) based on analog RRAM crossbar array circuits. RX-PUF takes advantage of unique RRAM properties, such as I-V nonlinearity, and its device-to-device (d2d) variations and tunability. As a proof of concept, we have prototyped a 600 kb challenge response pair (CRP) PUF using 250 nm half-pitch (F) 20×20 crossbar arrays with passively integrated devices. The RX-PUF prototype features excellent physical characteristics, e.g. ~1600 F2/bit density and up to 41 fJ/bit energy efficiency. Its functional performance, improved by utilizing hidden input, is also very promising. The measured bit error rate (BER) was 0.7% at RT and ≤ 5.3% at 100°C, even without using any error correction methods. The measured responses showed near-ideal uniformity (50.04%) and inter-HD (50.12%) and passed all relevant NIST randomness tests. The preliminary results showed also very high resilience of RX-PUF against machine learning (ML) attacks.