安全与密码学2019lightcommands.com编辑精选
针对语音可控系统的激光音频注入攻击(Light Commands)
Laser-Based Audio Injection Attacks on Voice-Controllable Systems
进阶 · 约 55 分钟读完前置知识:信号调制基础、MEMS 麦克风原理、语音助手系统
导读
这篇论文提出 LightCommands 攻击:攻击者把音频信号调制到激光上,照射语音助手设备的麦克风,利用麦克风对光信号的响应还原出声音并注入命令。作者在 Siri、Portal、Google Assistant、Alexa 等商用系统上验证成功,最远超过 100 米,且能穿透透明玻璃窗。论文还指出这类攻击可进一步波及门锁、汽车等第三方硬件。
HN 关注点:这种无需接触、远距离穿窗的物理攻击,以及语音助手生态和门锁/汽车等下游安全风险,并讨论攻击成本与防护是否可行。
后来怎么样
成为语音助手物理层安全与 MEMS 麦克风光注入研究的代表性工作,后续研究沿此方向继续分析光/声注入攻击及防护。
HN 评论精选
这几条评论里:支持 1 质疑 1 补充 1
质疑攻击成本被低估,除激光笔外还需昂贵的激光驱动器和音频放大器。
这个攻击很酷,但我不确定你从哪得出“14 美元”这个数字。激光笔是三支 18 美元,但你还需要单独的激光驱动器,比如作者用的那种 300 美元的驱动器,可能还需要音频放大器。攻击视频在 https://lightcommands.com/。
sp332查看英文原文
This is a cool attack, but I'm not sure where you got "$14" from. The laser pointers were $18 for a pack of 3, but you need a separate laser driver like the $300 one the authors used, and probably an audio amplifier as well. There are some videos of the attack at https://lightcommands.com/
补充论文结论:远距离穿窗注入语音命令,并可波及门锁和汽车等下游设备。
给懒得读论文的人:论文结论部分说,LightCommands 用光把音频信号注入语音控制系统;攻击者传输经过音频调制的光,光在麦克风内被转换回原始音频信号。作者在 Siri、Portal、Google Assistant、Alexa 等商用系统上演示成功,最远距离超过 100 米,并能穿透透明玻璃窗。论文还指出语音控制系统安全缺陷会导致门锁、汽车等第三方硬件的进一步失陷,并建议从物理原理上理解该攻击以发展新攻击和对策,例如用光对运动传感器等其他声学注入攻击,或用激光加热向传感器注入假信号。
smolsky查看英文原文
For the ones too lazy to RTFA: VIII. CONCLUSIONS AND FUTURE WORK In this paper we presented LightCommands , which is an attack that uses light to inject commands into voice-controllable systems from large distances. To mount the attack, an attacker transmits light modulated with an audio signal, which is converted back to the original audio signal within a microphone. We demonstrated LightCommands on many commercially available voice-controllable systems that use Siri, Portal, Google Assistant, and Alexa, obtaining successful command injections at a maximum distance of more than 100 meters while penetrating clear glass windows. Next, we highlight deficiencies in the security of voice-controllable systems, which leads to additional compromises of third-party hardware such as locks and cars. Better understanding of the physics behind the attack will benefit both new attacks and countermeasures. In particular, we can possibly use the same principle to mount other acoustic injection attacks (e.g., on motion sensors) using light. In addition, heating by laser can also be an effective way of injecting false signals to sensors.
支持物理遮挡对策可被提高激光功率或直接烧穿屏障绕过。
我很喜欢他们对对策的看法:“可以用物理屏障挡住直射光,减少到达麦克风振膜的光量……但我们指出这种物理屏障只在某程度上有效,因为攻击者总能提高激光功率来补偿遮挡造成的衰减。最后,如果无法补偿,攻击者总能用激光烧穿屏障,自己造出一条光路。”当然,你可以试着给麦克风加个盖子来防御,但我们已经有一台激光了,直接烧穿就行。
w-m查看英文原文
I really liked their take on the countermeasures: "It is possible to reduce the amount of light reaching the microphone’s diaphragm using a barrier that physically blocks straight light beams [...] However, we note that such physical barriers are only effective to a certain point, as an attacker can always increase the laser power in an attempt to compensate for the cover-induced attenuation. Finally, in case such compensation is not possible, the attacker can always use the laser to burn through barriers, creating his own light path." Sure, you can try to defend by putting a lid on the microphone. But we already have a laser here, we'll just burn through that.
相关资源
从这篇论文所有 HN 讨论帖的全部评论里整理出来的链接。
历次讨论
- 2019Hacker News 讨论帖218 分 · 107 评
同领域还值得读
如何入侵一台已关机的电脑:在 Intel ME 中运行未签名代码
How to hack a turned-off computer, or running unsigned code in Intel ME
这篇 Black Hat 2017 白皮书讨论 Intel Management Engine(ME)的安全问题:如何在 ME 中运行未签名代码,以及攻破处于关机状态的电脑的相关…
安全与密码学2017 · 543 分 · 107 评专业 · 约 40 分钟编辑精选比特币:一种点对点的电子现金系统(2008)
Bitcoin: A Peer-to-Peer Electronic Cash System (2008)
中本聪 2008 年发表的比特币白皮书,提出了无需可信第三方的点对点电子现金系统。
安全与密码学2017 · 536 分 · 394 评入门 · 约 45 分钟编辑精选GPRS 加密算法 GEA-1 的密码分析表明存在故意弱点
Cryptanalysis of GPRS Encryption Algorithms GEA-1 suggest intentional weakness
首次公开了对 GPRS 加密算法 GEA-1 和 GEA-2 的密码分析攻击。
安全与密码学2021 · 487 分 · 107 评专业 · 约 90 分钟编辑精选Signal 消息协议的形式化安全分析
A Formal Security Analysis of the Signal Messaging Protocol
用形式化方法分析 Signal 消息协议(TextSecure)的安全性:作者在多阶段密钥交换模型下,借助自动化证明工具对协议的密钥协商与 Double Ratchet 密钥更新…
安全与密码学2016 · 353 分 · 230 评专业 · 约 120 分钟编辑精选A2:模拟恶意硬件
A2: Analog Malicious Hardware
论文研究如何在芯片设计阶段植入模拟电路形式的恶意硬件后门,使其难以被常规检测发现。
安全与密码学2016 · 315 分 · 94 评专业 · 约 75 分钟编辑精选