Quantum-secured networking constitutes a layering architecture comprising a user network and a quantum key distribution network (QKDN). In the wavelength division multiplexing-based quantum-secured optical networking, wavelength and quantum key are scarce resources in the user network and QKDN, respectively; hence, an efficient resource dimensioning algorithm is required for practical quantum-secured services. Therefore, we investigated a resource dimensioning problem for the standardization-compatible quantum-secured networking model, where a data routing path in the user network and a quantum key relay path in the QKDN are interdependent. As benchmarking solutions, we first developed integer linear programming (ILP) formulations for wavelength and quantum key resource optimal quantum-secured networking. To obtain the tractable solution, an iterative heuristic algorithm was proposed, in which a path of a quantum-secured service was calculated by considering the statuses of both wavelength and quantum key resources. Simulation study on practical backbone topologies revealed that the proposed heuristic algorithm reduced manifold order-of-magnitude in the algorithm computation time at the costs of acceptable overheads in the wavelength and quantum key resources, from those of the ILP optimizations.
Quantum-secured networking constitutes a layering architecture comprising a user network and a quantum key distribution network (QKDN). In the wavelength division multiplexing-based quantum-secured optical networking, wavelength and quantum key are scarce resources in the user network and QKDN, respectively; hence, an efficient resource dimensioning algorithm is required for practical quantum-secured services. Therefore, we investigated a resource dimensioning problem for the standardization-compatible quantum-secured networking model, where a data routing path in the user network and a quantum key relay path in the QKDN are interdependent. As benchmarking solutions, we first developed integer linear programming (ILP) formulations for wavelength and quantum key resource optimal quantum-secured networking. To obtain the tractable solution, an iterative heuristic algorithm was proposed, in which a path of a quantum-secured service was calculated by considering the statuses of both wavelength and quantum key resources. Simulation study on practical backbone topologies revealed that the proposed heuristic algorithm reduced manifold order-of-magnitude in the algorithm computation time at the costs of acceptable overheads in the wavelength and quantum key resources, from those of the ILP optimizations.
Quantum key distribution (QKD) uses the laws of quantum mechanics to distribute secret keys with information-theoretic security. To realize QKD for secure communications among multiple distant users, fiber-based QKD networks can be deployed with today's available technologies. Incorporating QKD into existing wavelength-division multiplexing (WDM) network infrastructure provides a practical way to reduce the difficulty and cost of QKD networking. Nonetheless, the costs of QKD nodes and QKD links are still considered as the major barriers to long-distance QKD network deployment in practice. In this work, we study how to minimize the cost of deploying QKD-over-WDM backbone networks. We introduce a general architecture of QKD-over-WDM backbone networks. A novel cost-oriented model is defined to elaborate the cost-minimized problem, in which various QKD network components including QKD transceivers, QKD auxiliary equipment for QKD backbone nodes and trusted repeater nodes (TRNs), and QKD links are considered. The physical-layer parameters such as secret-key rate, physical distance, and the layout of TRNs are also incorporated in our defined model. We develop an integer linear programming (ILP) model and a novel cost-efficient QKD networking (CEQN) heuristic algorithm to address the cost-minimized problem. The extensive simulations indicate that our proposed approaches are effective in reducing the cost of deploying QKD-over-WDM backbone networks, where the designed CEQN algorithm is efficient to demonstrate similar to the ILP model. Finally, some open issues in future work are discussed, given that this work only presents an early highlevel analysis as opposed to a definitive detailed model.
Quantum key distribution (QKD)-based optical networks utilize quantum mechanics to secure communications by establishing secure, albeit expensive, quantum channels. A critical criterion for these QKD networks is to ensure network availability, which requires each service to adhere to a minimum unavailability constraint to maintain maximum network availability. This paper tackles the challenge of providing dedicated path protection for a select set of key services within QKD optical networks, adhering to specific unavailability constraints to ensure the gap between working and protection paths' unavailability remains within the allowable limit for each services. Considering the constraints of the limited timeslot resources and unavailability for the dedicated-path protection, we develop integer linear programming (ILP) models, to minimize the overall timeslot consumption in QKD optical networks. Additionally, we propose two novel algorithms for timeslot and availability management as minimum timeslot consumption (MTC) algorithm, and maximum availability (MA) algorithm to reduce the timeslot consumption. Simulations indicate that the ILP model based solution achieves the lowest timeslot consumption, however, it requires considerably more processing time compared to the proposed MTC and MA algorithms. Moreover, the MTC and MA algorithms outperform conventional dedicated-path protection (CDP) and fixed routing (FR) algorithms in reducing total timeslot consumption and improving average unavailability.
The maturity and developments of quantum key distribution (QKD) enable its application for encrypting dense wavelength-division multiplexing networks (DWDMNs), ensuring security in the presence of scalable quantum computers. The potential infrastructure reuse and costly deployment of a QKD network (QKDN) require jointly optimized deployment and utilization. We develop and solve a multi-layer problem formulation that fulfills the required keys of the DWDMN demands with the cost-minimized QKDN and maximized reuse of the present infrastructure. Our approach proves the feasibility of currently available QKD devices to encrypt nation-wide DWDMN with multi-period network planning. Furthermore, we compare static and dynamic routing approaches to quantify the advantage of software-defined networking for key management networks (KMNs). Depending on the network load, our multi-layer optimization shows potential savings between 9% and 45% of QKD devices compared to the baseline. Additionally optimizing the operation with dynamic routing increases the performance further, but the gain is negligible for encrypting DWDMNs.
Data transmission with optical fiber is vulnerable to eavesdropping. Moreover, conventional key distribution technology suffers from increasing computational power and upgraded attack algorithms. To address these issues, quantum key distribution (QKD), a quantum technology that secures secret information (such as a cryptographic key) exchange between two parties, can be used to guarantee secure data transmission. Integrating QKD into existing wavelength division multiplexing optical networks has been verified through a series of experiments, which contribute to ensuring network security and saving fiber resources. This paper addresses the resource assignment problem in QKD-enabled optical networks. First, a QKD-enabled optical network architecture is introduced. A small fraction of wavelength channels are segmented into multiple time slots with optical time division multiplexing technology to construct quantum key channels (QKChs) and measuring-basis channels, and then the remaining wavelengths can construct traditional data channels. Second, a static routing, wavelength, and time-slot assignment (RWTA) strategy is proposed and verified by the integer linear programming formulation and a heuristic algorithm. In the RWTA, QKChs are assigned for service requests according to the security levels specified by relevant key-updating periods. Thus, the secret keys for data encryption can update periodically to enhance security. Simulation results indicate that there is a trade-off between security (i.e., security levels and security-level types) and resource utilization.
Time-sharing of Quantum Key Distribution (QKD) transceivers with the help of optical switches and a central Software-Defined Networking (SDN) controller is a promising technique to better amortize the large investments required to build a Quantum Key Distribution Network (QKDN). In this work, we investigate the implications of introducing Time-Division Multiplexing (TDM) in trusted-relay QKDNs at the wide-area network scale in terms of performance and cost-saving. To this end, we developed both a Mixed Integer Linear Programming (qTDM-MILP) model and a Heuristic Algorithm (qTDM-HA) to solve the allocation of QKD transceivers and network resources for a novel switched QKDN operating scheme: qTDM-QKDN. Our heuristic method provides a close-to-optimal resource planning for the offline problem that computes the minimum number of QKD transceivers and optical switch ports at each node, as well as the number of quantum channels on each link required to satisfy a target set of end-to-end secret-keyrate demands. Moreover, both the model and the heuristic provide the time fractions that each QKD transceiver needs to peer with each neighbor QKD transceiver. We compared our proposed model and heuristic algorithm for cost minimization with non-time sharing QKD transceivers (nTDM) as baseline. The results show that qTDM can achieve substantial cost-savings in the range of 10%-40% compared to nTDM. Furthermore, this work sheds light on the selection of the value for the working cycle T and its influence on network performance.
With the rapid development of optical networks and the increasing amount of data transmission, information security has become a critical concern. Quantum key distribution can theoretically provide security for data encryption key distribution, and its integration into optical networks can enhance overall security. To reduce deployment costs, Wavelength Division Multiplexing is used to combine quantum and classical signals. However, limited wavelength resource necessitate the use of Time Division Multiplexing. How to efficiently allocate the multidimensional resources, such as wavelength, time slots, and encryption keys, has become a key issue in the design of secure optical networks. This paper proposes a Security-Level-based routing, wavelength, and time slot assignment strategy for optical networks. To evaluate the effectiveness of the proposed strategy, we introduce a new metric called Average Key Consumption. This metric allows us to assess key usage efficiency and network performance. We analyze the impact of different Average Key Consumption values on network resource allocation. Simulation results demonstrate that the Security-Level-based routing, wavelength, and time slot assignment strategy outperforms other strategies in handling high security demands.