Multi-Criteria Routing for Cryptographic Key Exchange in QKD Networks: Survey, Taxonomy, and a Conceptual Optimization Framework

The gist: This survey analyzes and classifies 26 routing strategies proposed between 2013 and 2024 for terrestrial, satellite, and hybrid QKD infrastructures.

Overview of Routing Approaches in Quantum Key Distribution Networks

This paper surveys 26 routing strategies proposed between 2013 and 2024 for terrestrial, satellite, and hybrid QKD infrastructures. It analyzes dynamic, key-aware routing algorithms which have been shown to reduce service rejection rates by 25–40% compared to static shortestpath methods by incorporating real-time key pool availability and link error rates. Multi-path strategies improve resilience against trusted-node compromise by distributing keys across disjoint routes, albeit with an increase in key consumption of 30–60%. SDN-based orchestration frameworks emerge as essential enablers of flexible, QoS-aware routing for hybrid networks integrating terrestrial fibers and intermittent satellite links. Critical research directions include reducing dependence on trusted nodes through quantum repeaters or post-quantum cryptography, developing standardized key management interfaces for interoperability, and adopting AI-driven predictive routing for adapting to fluctuating network states.

Terrestrial QKD Networks

In fiber-based QKD networks, routing must contend with high attenuation over long distances. Trusted nodes extend the reach by storing and relaying keys between shorter fiber segments. Algorithms such as those proposed in research [20], use dynamic path selection based on key buffer availability. The resource constraint makes it essential for routing protocols to monitor the current state of key buffers and adapt accordingly to avoid service disruptions due to key depletion.

Satellite QKD Networks

Satellite QKD introduces a unique set of constraints compared to terrestrial fiberbased networks. Unlike stable terrestrial links, satellite links are highly time-dependent, intermittent, and asymmetric. Low Earth Orbit (LEO) satellites exhibit periodic visibility windows with ground stations. A topology abstraction method was introduced to construct a static logical layer that aggregates all feasible satellite-ground pairings over a given pass interval, allowing the application of classical routing algorithms such as Dijkstra or ILP for determining optimal contact sequences for key relaying. Satellite QKD requires solving the contact scheduling problem—determining which users or links are to be served during each satellite pass.

Hybrid QKD Networks

Hybrid QKD networks combine terrestrial and satellite components. Routing in such systems involves selecting among heterogeneous paths with different trust levels and performance profiles. Heuristic approaches have been proposed to minimize total latency while ensuring sufficient key material is available for end-users. In general, hybrid routing algorithms must balance trade-offs between performance (throughput, delay), trust (number of intermediaries), and cost.

Routing Algorithms and Protocols

Early routing strategies in QKD networks adapted classical algorithms like Dijkstra’s shortest path, with edge weights reflecting QKD-specific metrics such as key rate or trust level. A dynamic graph approach was proposed in which only links with available keys are considered during route computation. Recent developments have focused on SDN-controlled dynamic routing approaches, wherein link weights are updated in real time based on current key buffer levels and predicted key lifetimes. Multi-path routing has been explored as a strategy to enhance robustness and confidentiality: one approach employed XOR-based schemes to distribute secret keys across disjoint paths.

SDN Orchestration and Control

In modern QKD networks, a centralized SDN controller (control layer) orchestrates both the quantum and classical resources. The controller uses this information to schedule key generation and path setup for user requests. The SDN control plane unifies classical and quantum routing: it computes routes that optimize key throughput or latency based on real-time key-buffer metrics, then programs both the classical switches and the QKD hardware accordingly.

Trends and Open Challenges

First, most routing strategies rely on trusted-node assumptions, which limit end-to-end security guarantees. Second, current algorithms often neglect fine-grained traffic engineering and load distribution between quantum and classical paths. Third, interoperability between heterogeneous QKD systems and vendor platforms is limited. Future research should aim to unify routing, key management, and service orchestration into a single adaptive framework capable of operating across diverse quantum network architectures.

Conclusion

Routing in such contexts requires integration of contact scheduling, security policies, and trust-model awareness. Multi-path strategies and SDN-enabled control were identified as essential techniques for achieving both robustness and scalability in large QKD networks. Practical QKD routing requires real-time awareness of quantum link status and key availability. Integration with classical control planes (via SDN) to coordinate quantum and classical network resources is necessary.

1. Improved AI system can perform dynamic, key-aware routing by incorporating real-time key pool availability and link error rates to reduce service rejection rates by 25–40% compared to static shortest-path methods.

2. AI systems can implement multi-path strategies to improve resilience against trusted-node compromise by distributing keys across disjoint routes, albeit with an increase in key consumption of 30–60%.

3. SDN-based orchestration frameworks can be utilized to enable flexible, QoS-aware routing for hybrid networks integrating terrestrial fibers and intermittent satellite links, allowing the system to balance trade-offs between latency, security, and cost based on real-time environmental conditions.

4. AI systems can utilize a centralized SDN controller that maintains a global view of the network to compute routes by penalizing links with low key availability or high latency, ensuring that links falling below a predefined key threshold can be excluded from the routing graph entirely.

5. Reinforcement learning-based methods can be developed to dynamically select routes based on observed network behavior and key utilization patterns, as suggested by the paper's mention of [9].

6. AI systems can employ topology abstraction methods to convert dynamic link states into quasi-static graphs for satellite networks, simplifying routing decisions over orbital cycles and enabling the application of classical algorithms like Dijkstra or ILP.

7. AI systems can implement predictive routing by using machine-learning models that forecast link availability (based on weather or satellite ephemeris) or key generation rates, and proactively reroute before failures occur.

8. The system can support QoS-aware routing by implementing utility-based optimization model[s] that prioritize key delivery for high-importance users in the context of limited key availability, ensuring high-priority sessions maintain uninterrupted access to key material.

9. The system can utilize SDN slicing to implement multi-tenant partitioning of QKD resources, ensuring that each user’s key demands are isolated and preventing interference across different user domains.

Quantum Key Distribution (QKD) networks enable unconditionally secure key exchange using quantum mechanical principles. However, routing cryptographic keys across multi-hop quantum networks introduces challenges unique to quantum communication. This survey analyzes and classifies 26 routing strategies proposed between 2013 and 2024 for terrestrial, satellite, and hybrid QKD infrastructures. Dynamic, key-aware routing algorithms have been shown to reduce service rejection rates by 25-40% compared to static shortest-path methods by incorporating real-time key pool availability and link error rates. Multi-path strategies improve resilience against trusted-node compromise by distributing keys across disjoint routes, albeit with an increase in key consumption of 30-60%. SDN-based orchestration frameworks emerge as essential enablers of flexible, QoS-aware routing for hybrid networks integrating terrestrial fibers and intermittent satellite links. Critical research directions include reducing dependence on trusted nodes through quantum repeaters or post-quantum cryptography, developing standardized key management interfaces for interoperability, and adopting AI-driven predictive routing for adapting to fluctuating network states. This comprehensive synthesis of routing algorithms, performance trade-offs, and unresolved issues offers practical guidance for engineering secure, adaptable, and large-scale QKD networks capable of meeting real-world deployment demands.