Wang Yijun, Li Yanyan, Jiang Wenqi, Guo Ying
School of Automation, Central South University, Changsha 410083, China.
School of Computer Science, Beijing University of Posts and Telecommunications, Beijing 100876, China.
Entropy (Basel). 2025 Jun 13;27(6):631. doi: 10.3390/e27060631.
A homodyne detector, which is also a common element in current telecommunication, is a core component of continuous-variable quantum key distribution (CV-QKD) since it is considered the simplest setup for the distinguishing of coherent states with minimum error. However, the theoretical security of CV-QKD is based on the assumption that the responses of the homodyne detector are always linear with respect to the input, which is impossible in practice. In the real world, a homodyne detector has a finite linear domain, so the linearity assumption is broken when the input is too large. Regarding this security vulnerability, the eavesdropper Eve can perform the so-called homodyne detector-blinding attack by saturating the homodyne detector and then stealing key information without being detected by the legitimate users. In this paper, we propose a countermeasure for the homodyne detector-blinding attack by using an adjustable optical attenuator with a feedback structure. Specifically, we estimate the suitable attenuation value in the data processing of CV-QKD and feed it back to the adjustable optical attenuator before the detector in real time. Numerical simulation shows that the proposed countermeasure can effectively defend against homodyne detector-blinding attacks and ensure the security of the Gaussian-modulated coherent state protocol with finite-size effect.
零差探测器也是当前电信领域的常见元件，它是连续变量量子密钥分发（CV-QKD）的核心组件，因为它被认为是区分相干态且误差最小的最简单装置。然而，CV-QKD的理论安全性基于这样的假设，即零差探测器的响应始终与输入呈线性关系，而这在实际中是不可能的。在现实世界中，零差探测器具有有限的线性域，所以当输入过大时，线性假设就会被打破。针对这种安全漏洞，窃听者伊芙可以通过使零差探测器饱和，然后窃取密钥信息而不被合法用户察觉，来实施所谓的零差探测器致盲攻击。在本文中，我们提出了一种通过使用具有反馈结构的可调光衰减器来应对零差探测器致盲攻击的对策。具体而言，我们在CV-QKD的数据处理过程中估计合适的衰减值，并实时将其反馈到探测器前的可调光衰减器。数值模拟表明，所提出的对策能够有效抵御零差探测器致盲攻击，并确保具有有限尺寸效应的高斯调制相干态协议的安全性。
