期刊
IEEE INTERNET OF THINGS JOURNAL
卷 6, 期 1, 页码 388-398出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/JIOT.2018.2849324
关键词
Artificial neural networks (ANNs); authentication; deep neural network; device signatures; Internet-of-Things (IoT); machine learning (ML); physical unclonable function (PUF); radio frequency (RF); security
资金
- National Science Foundation SaTC, CNS [1719235]
- Semiconductor Research Corporation [2720.001]
- Division Of Computer and Network Systems
- Direct For Computer & Info Scie & Enginr [1719235] Funding Source: National Science Foundation
Traditional authentication in radio-frequency (RF) systems enable secure data communication within a network through techniques such as digital signatures and hash-based message authentication codes (HMAC), which suffer from key-recovery attacks. State-of-the-art Internet of Things networks such as Nest also use open authentication (OAuth 2.0) protocols that are vulnerable to cross-site-recovery forgery (CSRF), which shows that these techniques may not prevent an adversary from copying or modeling the secret IDs or encryption keys using invasive, side channel, learning or software attacks. Physical unclonable functions (PUFs), on the other hand, can exploit manufacturing process variations to uniquely identify silicon chips which makes a PUF-based system extremely robust and secure at low cost, as it is practically impossible to replicate the same silicon characteristics across dies. Taking inspiration from human communication, which utilizes inherent variations in the voice signatures to identify a certain speaker, we present RF-PUF: a deep neural network-based framework that allows real-time authentication of wireless nodes, using the effects of inherent process variation on RF properties of the wireless transmitters (Tx), detected through in-situ machine learning at the receiver (Rx) end. The proposed method utilizes the already-existing asymmetric RF communication framework and does not require any additional circuitry for PUF generation or feature extraction. The burden of device identification is completely shifted to the gateway Rx, similar to the operation of a human listener's brain. Simulation results involving the process variations in a standard 65-nm technology node, and features such as local oscillator offset and I-Q imbalance detected with a neural network having 50 neurons in the hidden layer indicate that the framework can distinguish up to 4800 Tx(s) with an accuracy of 99.9% [approximate to 99% for 10 000 Tx(s)] under varying channel conditions, and without the need for traditional preambles. The proposed scheme can be used as a stand-alone security feature, or as a part of traditional multifactor authentication.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据