Kap-Joong Kim

Modern information and communications security has relied on public-key cryptosystems-such as RSA and ECC-based on computational complexity. With the accelerating global research and investment in quantum computing, however, warnings that these traditional schemes could be fundamentally broken by quantum algorithms such as Shor’s algorithm[1] are becoming increasingly realistic. As a result, quantum key distribution (QKD), which provides security independent of computational complexity by harnessing quantum-mechanical principles, has emerged as a promising alternative. QKD has advanced from laboratory experiments to inter-city trials, nationwide fiber-optic networks, and satellite links. Nevertheless, challenges in manufacturing, especially cost, size, and power consumption, continue to hinder the commercialization of QKD. To mitigate these issues, photonic integrated circuit (PIC) transmitters and receivers based on silicon, InP, SiN, and LiNbO_3 platforms have been developed[2,3,4]; to our knowledge, no commercial form-factor transmitter integrating both the light source and the modulator has been reported. In this work, we present the first QKD transmitter that hybrid-integrates a laser source, asymmetric delay interferometer, variable optical attenuator, and phase modulator into a CFP2 form-factor. We first measure and report the performance of each component, including the pulse characteristics depending on the system rate, the interferometer characteristics, the maximum optical attenuation, and the phase modulator’s half-wave voltage (V_pi). Furthermore, using the integrated transmitter, we implement a phase-encoding BB84 system, measure the interference visibility as a function of the phase modulator’s drive voltage and the sifted-key rate, then calculate the resulting quantum bit error rate (QBER) and secret-key rate (SKR).

Quantum Key Distribution (QKD) has been actively studied due to its unconditional security against eavesdropping, especially in the era of quantum computing. Among various QKD implementations, free-space QKD—which transmits and receives single photons through free-space channels—has gained significant attention due to its lower signal attenuation over long distances and its potential for establishing global quantum networks without the need for fiber-optic infrastructure. Recently, it has also emerged as a promising solution for secure communications on mobile platforms such as drones and autonomous vehicles. To integrate QKD systems into compact mobile platforms, it is essential to achieve system miniaturization and weight reduction, while also maintaining stable optical link in environments with motion and vibration. One effective approach is the use of gimbal-based beam tracking systems, as gimbals are well-established technologies that provide high stabilization performance with relatively low weight, making them ideal for mobile use. In this study, we investigate the performance of a lightweight gimbal-based beam tracking system in an outdoor environment. Lasers with visible-wavelength and CMOS cameras were used as beacon beams and detectors to implement the tracking system. To compensate for the limited angular resolution of the gimbal, beam expansion of the beacon beam and appropriate design of fine tracking is incorporated. The results demonstrate the feasibility of using gimbal-based tracking systems for free-space QKD applications on mobile platforms.

In polarization-based Quantum Key Distribution (QKD), imperfections in the states of polarization (SoPs)—such as limited orthogonality within a basis and insufficient mutual unbiasedness between bases—can significantly increase the quantum bit error rate (QBER), undermining the reliability and security of key exchange. Reducing QBER is therefore critical for enhancing protocol robustness and extending the operational range of practical QKD systems [1]. This issue is particularly pronounced in compact and lightweight implementations using Photonic Integrated Circuits (PICs), where the high level of integration restricts the flexibility to optimize or substitute individual polarization-controlling components. Although some degree of electrical tunability is possible, dependence on active feedback and real-time correction compromises the intrinsic advantages of simplicity, stability, and low power consumption. A passive approach requiring minimal intervention is thus especially desirable. We present an analytic method for determining the optimal unitary transformation that minimizes QBER under given non-ideal SoPs. When the polarization states are not mutually orthogonal, it is unclear how to align or transform them to ensure secure and efficient operation. Our method addresses this ambiguity by introducing QBER minimization as a physically meaningful and practical criterion for optimizing polarization alignment, as exemplified by scenarios such as six-state polarization encoding in reference-frame-independent QKD [2]. For the optimal transformation, we derive a closed-form expression, which can be implemented using static optical elements such as waveplates or integrated passive structures [3]. The proposed framework identifies the transformation direction that maximizes QKD performance. While initially designed for fixed SoPs, it is also expected to outperform iterative methods in dynamic environments by enabling fast, precomputed correction. As such, it is particularly useful in passive or resource-constrained polarization-based QKD systems, where active control is undesirable or impractical.

Photonic integrated circuits (PICs) are emerging as a key enabler for compact, rugged and mass-producible quantum key distribution (QKD) terminals, dramatically reducing size, weight, power and cost while facilitating co-packaging with classical transceivers. However, inevitable fabrication asymmetries and on-chip optical components such as modulator, attenuator, and wavelength division multiplexer (WDM) introduce polarization dependent loss (PDL) of up to several decibels. In polarization-based QKD that employs multiple polarization states, such intrinsic PDL can distort those states and thereby degrade overall system performance. The impact is particularly severe in polarization-based reference frame independent (RFI) QKD, where the six polarization states must remain mutually unbiased and orthogonal within each basis so that the secret key rate depends only on the quantum bit error rate (QBER) and the security parameter. PDL skews the Jones amplitudes, breaks these state relations, inflates QBER and can drive the secret key rate to low values at moderate channel loss. To counter this impairment, post-selection methods non-optically compensating PDL have been proposed. In this work, we introduce an optical PDL compensation which is a passive optical compensator consisting of a matched PDL element followed by a half-wave plate that swaps the orthogonal polarization components. The combined transfer matrix is effectively polarization-isotropic apart from a uniform attenuation so both the security parameter and the secret key rate are preserved, at the expense of additional insertion loss. Experiments with a free-space RFI QKD link confirm the scheme’s effectiveness: while uncompensated PDL quickly degrades secret key raete, the compensator restores state and sustains secret key rate across the entire operating range explored, validating the effectiveness of the compensator and demonstrating a practical path toward PDL tolerant, PIC-based QKD systems.

Since the first quantum key distribution (QKD) BB84 was proposed, various relevant QKD systems have been proposed. The systems mainly are implemented based on bulk-optics. The bulk-optics based implementation has large volume, heavy weight, high power consumption, and high cost, so that it is not easy to compatible to current communication system and is hard to be commercialized. In order to overcome the aforementioned, recently photonic integrated chip (PIC) based implementation of QKD systems have been proposed. Unfortunately, PIC based implementation has inevitable polarization depende loss (PDL), which can be induced by imperfect fabrication of waveguide, or plasma dispersion effect which can change absorption in case of a polarization modulator. Additionally, a QKD system currently cannot be implemented with only PIC so that it requires additional optical components that has PDL as well. PDL may cause the severe performance degradation of QKD, by destroying the mutually unbiasedness between the states in QKD and increasing quantum bit error rate (QBER). In this study, we analyze effect of PDL in QKD. First, we experimentally analyze state change of polarization qubit depending on PDL which can be emulated by a PDL emulator. Based on this, we theoretically calculate intrinsic quantum bit error rate (QBER) and the corresponding secure key rate of QKD.

In this study, we report the performance of our polarization-based BB84 protocol QKD system using a silica-based polarization encoding chip and a fiber-based WDM filter, which can easily combine the quantum channel of 785nm with the synchronization channel of 1550nm. Among the optical transmission windows in the atmosphere, 785nm was selected as the wavelength of the quantum channel due to its availability for high-performance silicon-based single-photon avalanche diodes and laser diodes. Additionally, 1550nm was selected for beam tracking and synchronization signal for its availability in commercial transceivers and optical amplifiers. The operation speed of the system was 100 MHz. The sifted key rate and quantum bit error rate (QBER) were measured in real-time by a field programmable gate array(FPGA). The system was installed in an indoor laboratory, and the QBER and sifted key rate were measured while increasing channel attenuation to test the system’s performance in lossy environments. We obtained superior QKD performance with a QBER of 0.62% and a sifted key rate of 1.61 Mbps at no attenuation, and 5.64% and 5.69 kbps at 25dB attenuation, showing potential application for outdoor operation over longer distances.

Research on quantum key distribution (QKD) using discrete variables is intensively progressing towards commercialization. The development of Photonic Integrated Chips is essential for this success. Specifically, in fiber-based QKD systems utilizing time-bin and phase-encoding, the characteristics of the asymmetric delay line interferometer play a critical role in the system's performance. This paper investigates the variations in the extinction ratio of interferometer based on the power loss ratio between optical signals traversing the short and long paths and the time delay error in the interferometer. Additionally, we analyze how the characteristics of the input optical waveform affect the extinction ratio of the interferometer.

In this work, we investigate the effect of the Kalman filter, an algorithm predicting future values of a system, for reducing pointing errors and improving the tracking performance of the coarse tracking system. We present the pointing error based on the angular velocity of the target when the Kalman filter is applied to the tracking algorithm. The tracking system is mounted on a fixed tripod, while the mobile platform moves around the system at a constant speed as a target. The effect of the Kalman filter on the performance of the tracking system and future work will be given.

In recent times, field trials of quantum key distribution (QKD) have been conducted using the existing optical fiber infrastructure. However, one significant challenge faced during these trials is ensuring the stability of QKD operation. The instability of QKD operation is caused by the two factors: random fluctuations in polarization of photon over time and time drift of the photon as it traverses the deployed optical fiber. These issues are unavoidable due to the inability to accurately estimate and control factors such as temperature, vibration, and stress in the deployed optical fiber. To address this instability, various solutions based on active or passive optics have been proposed. In this paper, we present an active optics-based simple polarization stabilizer utilizing an optical polarizer, an active polarization controller, and a single photon detector. For the fast operation, we utilize only 2 out of the 4 axes of the polarization controller for the stabilizer. The experimental results verify the stability of the stabilizer.