Quantum key distribution (QKD) enables private communications with information-theoretic security. To guarantee the practical security of QKD, it is essential that QKD systems are implemented in accordance with theoretical requirements and are robust with regard to side-channel attacks. Here we study a prominent attack on QKD transmitters known as the laser-seeding attack (LSA). It consists in injecting photons into the laser of the transmitter in an attempt to modify the outgoing light in some way that is beneficial to the eavesdropper. In this work we measure the response of a QKD transmitter to the LSA as a function of the optical power injected, allowing us to quantify the level of optical attenuation required to mitigate the attack. Further, we use a laser-rate-equation model to numerically simulate the effects of the LSA on a gain-switched laser. With this model we are able to reproduce previous experimental results, as well as generate new insight into the LSA by examining the effects of the LSA when the QKD transmitter is operated with different laser current driving parameters.
Quantum key distribution (QKD) enables private communications with information-theoretic security. To guarantee the practical security of QKD, it is essential that QKD systems are implemented in accordance with theoretical requirements and are robust with regard to side-channel attacks. Here we study a prominent attack on QKD transmitters known as the laser-seeding attack (LSA). It consists in injecting photons into the laser of the transmitter in an attempt to modify the outgoing light in some way that is beneficial to the eavesdropper. In this work we measure the response of a QKD transmitter to the LSA as a function of the optical power injected, allowing us to quantify the level of optical attenuation required to mitigate the attack. Further, we use a laser-rate-equation model to numerically simulate the effects of the LSA on a gain-switched laser. With this model we are able to reproduce previous experimental results, as well as generate new insight into the LSA by examining the effects of the LSA when the QKD transmitter is operated with different laser current driving parameters.
A well-protected and characterized source in a quantum key distribution system is needed for its security.Unfortunately, the source is vulnerable to light-injection attacks, such as Trojan-horse, laser-seeding, and laser-damage attacks, in which an eavesdropper actively injects bright light to hack the source unit.The hacking laser could be a high-power one that can modify properties of components via the laserdamage attack and also further help the Trojan-horse and other light-injection attacks.Here we propose a countermeasure against the light-injection attacks, consisting of an additional sacrificial component placed at the exit of the source.This component should either withstand high-power incoming light while attenuating it to a safe level that cannot modify the rest of the source, or get destroyed into a permanent high-attenuation state that breaks up the line.We demonstrate experimentally that off-the-shelf fiberoptic isolators and circulators have these desired properties, at least under attack by a continuous-wave high-power laser.
Quantum Key Distribution (QKD) promises information-theoretic security based on the fundamental laws of quantum mechanics. However, practical implementations deviate from idealized models and are therefore susceptible to side-channel attacks that exploit physical imperfections rather than theoretical vulnerabilities. Over the past two decades, a wide range of attacks have been demonstrated against QKD systems, including those targeting detectors, sources, timing information, power consumption, and electromagnetic emissions. This paper addresses a selection of a few side-channel attacks reported in the literature, classifying them according to their underlying physical mechanisms and attack strategies. We analyze their impact on the security of different QKD protocols and platforms, and highlight how experimental demonstrations have shaped the community's understanding of implementation security. In addition, we present and evaluate mitigation techniques proposed to date, such as device-independent and measurement-device-independent QKD, hardware countermeasures, monitoring strategies, and so on. We also identify open challenges and promising directions for enhancing the robustness of future quantum communication systems.
Unidimensional-modulation continuous-variable quantum key distribution has received considerable attention due to its simplified transmitter architecture, reduced random-number consumption, and low deployment cost. However, its security against source-side light-injection attacks remains unexplored. In this paper, the effects of such attacks on unidimensional-modulation continuous-variable quantum key distribution systems are systematically investigated. A theoretical model of quantum-state preparation at the transmitter under light-injection attacks is established, parameter estimation bias formulas are derived, the resulting degradation in the secret key rate is quantified, and the theoretical results are validated through numerical simulations. In addition, the security implications of light-injection attacks in the finite-size regime are examined. The results demonstrate that light-injection attacks cause overestimation of the channel transmittance and underestimation of excess noise. Under finite-size conditions, these attacks reduce the minimum data-block length required to achieve a positive secret key rate, thereby creating a false indication of security even when no secure key can actually be generated. These findings demonstrate that light-injection attacks constitute a serious threat to unidimensional-modulation continuous-variable quantum key distribution systems and provide an important theoretical foundation for future secure practical deployment.
In the transmitter of a quantum key distribution (QKD) system, a wavelength-division multiplexer (WDM) is typically used to combine quantum and synchronization signals and is directly connected to the quantum channel. As a result, it becomes the first optical component exposed to laser-injection attacks. Therefore, understanding the behavior of WDMs under such attacks is essential for assessing the practical security of QKD systems. In this work, we systematically investigate the characteristics of WDMs under high-power laser illumination. Our experimental results show that certain WDM samples exhibit pronounced changes in their spectral features once the injected laser power surpasses a specific threshold. Taking the Trojan-horse attack as an illustrative example, we further perform a theoretical analysis of the resulting spectral side channel and show that it can reduce the maximum secure transmission distance to below 66.9% of its original value. By combining experimental observations with theoretical modeling, this study advances the understanding of the influence of WDMs on the practical security of QKD systems.
Quantum key distribution (QKD) is a technology that allows two users to communicate with theoretically perfect security using standard optical fibres. This is possible by transmitting the key on single photons, meaning a measurement by an eavesdropper disturbs the system in a way observable to the legitimate parties. The technology has advanced since the first protocol proposed in 1984, to the stage where there are now many protocols that can be experimentally implemented. These protocols have allowed secure keys to be generated over distances greater than 400 km and with secure key rates over 10 Mbit/s. In a metropolitan QKD network, it would be desirable for as many users to be connected as possible. Unfortunately, each protocol comes with different requirements on the transmitter and receiver. Even within a single protocol, different clock rates can require individualised transmitter and receiver hardware. This prohibits users from communicating with all receivers, unless they have complex transmitters with hardware for many protocols. This thesis develops a transmitter for practical QKD that is able to adapt to a number of different protocols with no changes to the hardware. The transmitter works using optical injection locking, where a pulse preparation laser adopts the phase of a phase preparation laser. Controlling the phase and intensity of the pulses in this way removes the side channel ordinarily present with direct modulation, in that the phase, intensity and frequency simultaneously change in response to an applied current. The cavity-enhanced electro-optic effect allows for the first demonstration of sub-volt half-wave phase modulation at high clock rates. The transmitter successfully demonstrates phase encoding, intensity encoding and on-demand phase randomisation. This allows for the experimental realisation and direct comparison of different QKD protocols, including one that has never before been implemented due to experimental complexity. A stable intensity modulator is also developed, based on a Sagnac interferometer. This removes a side channel in QKD systems and integrates well with the directly-modulated quantum transmitter. This development also means that the transmitter can implement all current two-party QKD protocols based on weak coherent pulses. The transmitter has the potential to become the standard transmitter for future quantum communication networks due to its stability, versatility and power efficiency. The design could also be demonstrated on a photonic chip, making it compact enough to fit in small transmitter units.
For effective quantum communication, the security of the photon source is particularly important in the era of measurement-device-independent quantum key distribution (MDI-QKD) and twin-field QKD (TF-QKD). In practice, the security of the source can still be cracked by an adversary. This study experimentally demonstrates that a practical source based on a semiconductor laser diode is vulnerable to a laser-seeding attack, in which light injected from the communication line into the laser yields increased intensities of the prepared states. Theory shows that the unnoticed intensity increase compromises the security of the prepare-and-measure decoy-state BB84 and MDI-QKD protocols.