Side-Channel Analysis for Intelligent and Connected Vehicle Security: A New Perspective
Side-Channel Analysis for Intelligent and Connected Vehicle Security: A New Perspective
复制标题
智能网联汽车安全的侧通道分析：新视角
DOI：
10.1109/mnet.001.1900214
复制
发表时间：
2020
期刊：
影响因子：
9.3
通讯作者：
Zhang Yanning
中科院分区：
文献类型：
--
作者：
Xun Yijie;Liu Jiajia;Zhang Yanning
随着ICV的不断拓展和深化，先进技术不断涌现，使ICV更加智能化地服务和保护驾驶员。虽然ICV的发展为用户带来了极大的便利和更好的驾驶体验，但同时也为恶意攻击者带来了众多的入侵入口。有许多恶意入侵者只专注于如何利用一些攻击技术来实现恶意入侵，例如通过SCA窃取用户隐私。然而，技术是一把双刃剑。我们可以利用现有的攻击技术做一些有意义的事情，这对社会发展是有益的。为此，本文研究了ICV中的安全威胁，特别是SCA的应用。在此基础上，提出了基于SCA的驾驶员识别和非法驾驶员检测方案，并深入分析了各种方案的优点和局限性。实验结果表明，使用至少10分钟的训练数据和超过1分钟的测试数据，身份认证的准确率可以达到100%。根据最近的研究进展，讨论了未来的研究机会，以激励更多的研究人员关注ICV的安全性。
With the continuous expansion and deepening of ICVs, advanced technologies continue to emerge, making ICVs more intelligent to serve and protect drivers. Although the development of ICVs has brought great convenience and better driving experiences for users, at the same time it also brings many intrusion portals for malicious attackers. There are many hostile invaders who only focus on how to achieve malicious invasions by exploiting some attack technologies, such as stealing users’ privacy via SCA. However, technology is a double-edged sword. We can utilize existing attack technologies to do something meaningful, which is beneficial to social development. To this end, the security threats are investigated in ICVs, especially the application of SCA. After that, we propose the schemes of driver identification and illegal driver detection by using SCA, and provide an insight into the advantages and limitations of each scheme. The experimental results showed that identity authentication accuracy can reach 100 percent by using at least 10-minute training data and more than 1-minute testing data. Based on recent progress, future research opportunities are discussed to motivate more researchers to focus on the safety of ICVs.