Evaluating Relayed and Switched Quantum Key Distribution (QKD) Network Architectures

Basic Information

• Paper ID: 2509.24440
• Title: Evaluating Relayed and Switched Quantum Key Distribution (QKD) Network Architectures
• Authors: Antonis Selentis, Nikolas Makris, Alkinoos Papageorgopoulos, Persefoni Konteli, Konstantinos Christodoulopoulos, George T. Kanellos, Dimitris Syvridis
• Classification: cs.CR (Cryptography and Security)
• Submission Date: Submitted to Journal of Lightwave Technology
• Paper Link: https://arxiv.org/abs/2509.24440

Abstract

This paper evaluates the performance of two network-level quantum key distribution (QKD) architectures: Relayed QKD, which relays keys to non-adjacent nodes through multi-link QKD paths, and Switched QKD, which uses optical switches to dynamically connect arbitrary QKD modules to form direct QKD links. Switched QKD offers the advantage of end-to-end quantum key distribution, while relayed QKD relies on trusted nodes. However, switched QKD depends on arbitrary matching of QKD modules. The study first experimentally evaluates the performance of commercial DV-QKD modules, benchmarking standard/matched module pairs and non-matched module pairs from three vendors. The analysis reveals significant differences in key generation rates between matched and non-matched module pairs. Based on experimental findings, a comprehensive theoretical analysis is conducted to evaluate the network-level performance of both architectures.

Research Background and Motivation

Problem Background

1. Quantum Computing Threat: Quantum computers pose a threat to traditional network security, potentially leading to data breaches. QKD is one of the most mature countermeasures.
2. Network Scalability Requirements: Traditional point-to-point QKD configurations become inefficient as network scale increases. A fully connected network of N nodes requires N(N-1)/2 QKD module pairs, with costs and complexity growing dramatically with network size.
3. Distance Limitation Issues: P2P QKD links have distance limitations, and long-distance key exchange requires intermediate trusted nodes.

Research Motivation

1. Architecture Selection Dilemma: There is a need to choose between relayed and switched QKD architectures, but systematic performance comparisons are lacking.
2. Module Compatibility Issues: Switched QKD requires arbitrary QKD module pairing, but commercial devices are typically provided as matched pairs. The performance impact of non-matched pairings has not been sufficiently studied.
3. Network Optimization Requirements: Guidance on optimal architecture selection for different network scenarios is lacking.

Core Contributions

1. Experimental Evaluation: First systematic assessment of performance differences in commercial QKD modules from three vendors under matched and non-matched pairing configurations
2. Theoretical Analysis Framework: Derivation of optimal key management configurations and analytical formulas for uniform ring networks
3. Performance Comparison Methodology: Establishment of quantitative comparison methods for relayed and switched QKD architectures
4. Application Guidance: Provision of specific architecture selection recommendations for different network scenarios
5. Key Findings: Confirmation of significant impact of non-matched QKD module penalties on switched QKD efficiency

Methodology Details

Task Definition

This research aims to compare the secret key rate (SKR) performance of relayed and switched QKD network architectures under different network parameters. Inputs include network topology parameters (number of nodes N, adjacent node distance Le) and QKD system parameters, with outputs being the maximum minimum SKR consumption rate.

Architecture Design

Relayed QKD Architecture

• Operating Principle: Relays keys to non-adjacent nodes through multi-link QKD paths, relying on trusted intermediate nodes
• Key Management: KMS server maintains quantum key pools (QKPs) and implements relay functions (e.g., XOR operations)
• Topology Requirements: Each node in a ring network establishes QKD links with adjacent nodes

Switched QKD Architecture

• Operating Principle: Uses optical switches to dynamically connect arbitrary QKD modules, forming direct QKD links
• Time Division Multiplexing: Employs TDM to share N QKD module pairs, achieving full connectivity key generation
• Scheduling Strategy: Establishes QKD links in stages by k-hop distance, serving node pairs at the same distance simultaneously

Mathematical Modeling

Relayed QKD Performance Formula

For an N-node ring network, the consumption rate is:

CR = 8·G(Ae)/(N²-1), Ae = ac·Le

where G(Ae) is the adjacent node SKR generation function and ac is the attenuation coefficient.

Switched QKD Performance Formula

The consumption rate is:

CS = 1 bit / [R·(N-1)/2 + ∑(k=1 to (N-1)/2) 1/G(Ak+O)·10^(-P/10)]

where O is the optical switch loss penalty and P is the non-matched module performance penalty.

Technical Innovations

1. Experimental Method Innovation: Design of comparative experiments with bar configuration (matched pairs) and cross configuration (non-matched pairs)
2. Theoretical Modeling: First derivation of analytical performance formulas for both architectures in uniform ring networks
3. Scheduling Optimization: Proposal of optimal periodic scheduling strategy for switched QKD
4. Comprehensive Evaluation: Integration of experimental and theoretical analysis, considering practical factors such as optical switch loss and reconfiguration time

Experimental Setup

Experimental Equipment

Three commercial QKD systems from different vendors were tested:

1. Toshiba QKD 4.2-MU/MB: Phase encoding + decoy state (PDS), link budget 24dB
2. ID Quantique Clavis XG: Time-bin-phase encoding (TBP), link budget 30dB
3. Think Quantum Quky: Polarization encoding (PB), link budget 20dB

Experimental Configuration

• Matched Configuration (Bar): Standard matched pairs A1-B1 and A2-B2
• Non-matched Configuration (Cross): Cross-connected A1-B2 and A2-B1
• Test Duration: Approximately 11 hours of continuous measurement
• Attenuation Settings: Different attenuation values set according to vendor specifications

Evaluation Metrics

• SKR (Secret Key Rate): Key generation rate (bps)
• QBER (Quantum Bit Error Rate): Quantum bit error rate (%)
• SKR Δ: Performance deviation percentage between matched and non-matched configurations

Theoretical Analysis Parameters

• Network Scale: N = 5 to 25 nodes
• Adjacent Distance: Le = 1 to 35 km
• Attenuation Coefficient: ac = 0.24 dB/km
• Optical Switch Loss: O = 2 dB
• Reconfiguration Time: R = 5 minutes

Experimental Results

Main Experimental Findings

Matched vs. Non-matched Performance Comparison

Key Experimental Conclusions

1. PDS Modules: Non-matched pairing results in nearly two orders of magnitude performance degradation, indicating the module is not optimized for arbitrary pairing
2. TBP Modules: Negligible performance deviation (±5%), suitable for switched QKD applications
3. PB Modules: Moderate performance deviation (±23%), still acceptable but requiring consideration of impact

Theoretical Analysis Results

Relative Performance Comparison

Using relative consumption rate gain f = 100·(CS-CR)/max(CS,CR) for evaluation:

1. Switched QKD Advantage Scenarios:
   • Dense networks (short distance, large number of nodes)
   • Le ≤ 5 km for all N values: f > 25%
   • N ≥ 17 and Le < 5 km: f > 50%
2. Relayed QKD Advantage Scenarios:
   • Long-distance networks
   • N = 9 and Le > 20 km: f < -50%
   • N ≥ 21 and Le > 10 km: f < -50%

Parameter Sensitivity Analysis

Non-matched Penalty Impact:

• P = 0 dB (ideal case): Switched QKD advantage significantly expands
• P = 10 dB (high incompatibility): Relayed QKD becomes the only viable option
• P = 2 dB (reference setting): Balanced performance, both architectures have applicable scenarios

Receiver Dynamic Range Impact:

• K = 50 km flat region: Switched QKD advantage extends to greater distances
• K = 0 km (no flat region): Relayed QKD almost universally superior to switched

Related Work

QKD Network Architecture Research

1. Trusted Node Networks: China's 46-node metropolitan area network uses 3 interconnected trusted nodes as the center of a star topology
2. Space-Ground Integrated Networks: 4600 km space-ground quantum communication network uses 32 trusted relays to extend coverage range
3. Software-Defined QKD: Software-defined quantum networks using QKD-secure SDN controllers and encrypted messaging

Scheduling Algorithm Research

1. ILP Optimization: Integer linear programming-based optimal design algorithms for switched and relayed QKD networks
2. Resource Allocation: Resource-efficient QKD network schemes with joint routing, channel, and key rate allocation
3. Time Slot Allocation: Routing, wavelength, and time slot allocation based on auxiliary graphs in metropolitan quantum optical networks

Conclusions and Discussion

Main Conclusions

1. Architecture Applicability:
   • Switched QKD: Suitable for dense networks (short distance, multiple nodes), providing end-to-end quantum key distribution
   • Relayed QKD: Suitable for long-distance and medium-to-large-scale networks, lower cost but dependent on trusted nodes
2. Module Compatibility Critical: Performance penalties from non-matched QKD modules significantly impact switched QKD efficiency
3. Design Guidance: Flat SKR generation regions actually favor the switched QKD architecture

Limitations

1. Network Topology Constraints: Analysis based on uniform ring networks; actual networks are more heterogeneous
2. QKD Technology Assumptions: Primarily based on BB84 protocol and specific vendor parameters
3. Cost Analysis Missing: Insufficient consideration of hardware costs and deployment complexity
4. Security Trade-offs: Insufficient analysis of the trade-off between trusted nodes and end-to-end security

Future Directions

1. Hybrid Architectures: Hybrid schemes combining advantages of both architectures
2. Module Standardization: Development of QKD modules supporting arbitrary connections with minimal performance penalties
3. Heterogeneous Networks: Extension of analysis to non-uniform, complex topology networks
4. Dynamic Optimization: Algorithms for dynamically selecting optimal architectures based on network state

In-Depth Evaluation

Strengths

1. Experimental Comprehensiveness: First systematic evaluation of matched/non-matched performance of commercial QKD modules, filling an important gap
2. Theoretical Rigor: Derivation of precise analytical formulas, providing quantitative comparison basis for both architectures
3. High Practical Value: Provides clear architecture selection guidance with direct value for QKD network deployment
4. Methodological Innovation: The methodology combining experimental and theoretical analysis serves as an excellent model

Weaknesses

1. Network Model Simplification: Uniform ring network assumption limits the generalizability of results
2. Vendor Dependency: Conclusions partially depend on specific vendor equipment characteristics, potentially lacking universality
3. Dynamic Factors Overlooked: Insufficient consideration of network load variations, failure recovery, and other dynamic factors
4. Cost-Benefit Analysis Missing: Lack of detailed economic analysis and deployment cost considerations

Impact

1. Academic Contribution: Provides important theoretical and experimental foundation for QKD network architecture research
2. Industry Guidance: Valuable reference for QKD equipment manufacturers and network operators
3. Standards Promotion: May promote establishment of QKD module compatibility standards
4. Technology Development: Identifies key directions for QKD technology advancement

Applicable Scenarios

1. Metropolitan Networks: Particularly suitable for architecture selection in urban quantum communication networks
2. Data Center Interconnection: Provides guidance for short-distance, high-density connection scenarios
3. Critical Infrastructure: Offers design reference for quantum-secure networks in government, finance, and other critical sectors
4. Research Networks: Provides technical support for construction of quantum communication research networks

References

This paper cites 21 important references covering QKD security theory, commercial system implementation, network architecture design, scheduling algorithm optimization, and other aspects, providing a solid theoretical foundation and technical background for the research.

Overall Assessment: This is a high-quality QKD network architecture comparison study that provides valuable guidance for practical deployment through rigorous experimental and theoretical analysis. Despite some limitations, its innovative methodology and practical conclusions make it an important contribution to the field.