Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Optimizing Quantum Key Distribution Network Performance using Graph Neural Networks

Akshit Pramod Anchan, Ameiy Acharya, Leki Chom Thungon

TL;DR

The paper addresses the challenge of optimizing QKD network performance under dynamic conditions by applying Graph Neural Networks to model QKD topologies as dynamic graphs. It develops a Transformer-based GNN with GATv2 attention, edge-feature processing, and a link-prediction decoder to estimate viable quantum links, trained with negative sampling and binary cross-entropy loss. Empirical results show substantial improvements in total key rate (up to ~X-fold) and stable low QBER across network scales, with mid-sized networks offering the best balance between prediction accuracy and performance. The work demonstrates the potential of a global, graph-based optimization paradigm for scalable, adaptive, and secure quantum communication networks, paving the way for more intelligent routing and resource management in quantum internet deployments.

Abstract

This paper proposes an optimization of Quantum Key Distribution (QKD) Networks using Graph Neural Networks (GNN) framework. Today, the development of quantum computers threatens the security systems of classical cryptography. Moreover, as QKD networks are designed for protecting secret communication, they suffer from multiple operational difficulties: adaptive to dynamic conditions, optimization for multiple parameters and effective resource utilization. In order to overcome these obstacles, we propose a GNN-based framework which can model QKD networks as dynamic graphs and extracts exploitable characteristics from these networks' structure. The graph contains not only topological information but also specific characteristics associated with quantum communication (the number of edges between nodes, etc). Experimental results demonstrate that the GNN-optimized QKD network achieves a substantial increase in total key rate (from 27.1 Kbits/s to 470 Kbits/s), a reduced average QBER (from 6.6% to 6.0%), and maintains path integrity with a slight reduction in average transmission distance (from 7.13 km to 6.42 km). Furthermore, we analyze network performance across varying scales (10 to 250 nodes), showing improved link prediction accuracy and enhanced key generation rate in medium-sized networks. This work introduces a novel operation mode for QKD networks, shifting the paradigm of network optimization through adaptive and scalable quantum communication systems that enhance security and performance.

Optimizing Quantum Key Distribution Network Performance using Graph Neural Networks

TL;DR

The paper addresses the challenge of optimizing QKD network performance under dynamic conditions by applying Graph Neural Networks to model QKD topologies as dynamic graphs. It develops a Transformer-based GNN with GATv2 attention, edge-feature processing, and a link-prediction decoder to estimate viable quantum links, trained with negative sampling and binary cross-entropy loss. Empirical results show substantial improvements in total key rate (up to ~X-fold) and stable low QBER across network scales, with mid-sized networks offering the best balance between prediction accuracy and performance. The work demonstrates the potential of a global, graph-based optimization paradigm for scalable, adaptive, and secure quantum communication networks, paving the way for more intelligent routing and resource management in quantum internet deployments.

Abstract

This paper proposes an optimization of Quantum Key Distribution (QKD) Networks using Graph Neural Networks (GNN) framework. Today, the development of quantum computers threatens the security systems of classical cryptography. Moreover, as QKD networks are designed for protecting secret communication, they suffer from multiple operational difficulties: adaptive to dynamic conditions, optimization for multiple parameters and effective resource utilization. In order to overcome these obstacles, we propose a GNN-based framework which can model QKD networks as dynamic graphs and extracts exploitable characteristics from these networks' structure. The graph contains not only topological information but also specific characteristics associated with quantum communication (the number of edges between nodes, etc). Experimental results demonstrate that the GNN-optimized QKD network achieves a substantial increase in total key rate (from 27.1 Kbits/s to 470 Kbits/s), a reduced average QBER (from 6.6% to 6.0%), and maintains path integrity with a slight reduction in average transmission distance (from 7.13 km to 6.42 km). Furthermore, we analyze network performance across varying scales (10 to 250 nodes), showing improved link prediction accuracy and enhanced key generation rate in medium-sized networks. This work introduces a novel operation mode for QKD networks, shifting the paradigm of network optimization through adaptive and scalable quantum communication systems that enhance security and performance.

Paper Structure

This paper contains 17 sections, 9 equations, 4 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Proposed Methodology
  3. Experimental Setup
  4. Transformer Graph Convolution Layer
  5. GATv2 Attention Layer
  6. Edge Feature Processing
  7. Link Prediction Decoder
  8. Negative Sampling Handling
  9. Training Objective
  10. Results
  11. Network Scaling Characteristics
  12. QKD Performance Metrics
  13. Link Prediction Accuracy
  14. Comparison Across Node Counts
  15. Discussion
  16. ...and 2 more sections

Figures (4)

  • Figure 1: Simplified Demonstration of BB84 Communication Protocol
  • Figure 2: Model architecture with (1) Transformer-based graph convolution, (2) GATv2 attention layer, (3) Edge feature processing MLP, and (4) Link prediction decoder.
  • Figure 3: Three-stage quantum key distribution (QKD) network pipeline: (1) Network construction with BB84 protocol simulation, (2) Graph neural network (GNN) training with transformer and graph attention layers, and (3) Performance evaluation with quantum channel metrics.
  • Figure 4: Quantum Network Performance Visualization 1: Training Loss Evolution, 2: Validation AUC Evolution, 3: Key Rate vs Distance, 4: QBER Distribution