OptiQKD: A Machine Learning-Optimized Framework for Real-Time Parameter Tuning in Quantum Key Distribution

TL;DR

OptiQKD is proposed, a protocol-agnostic machine learning framework specifically engineered to maximize the Secure Key Rate (SKR) and minimize the Quantum Bit Error Rate (QBER) for the BB84, E91, and COW protocols.

Abstract

Despite the robust security guarantees of Quantum Key Distribution (QKD), its practical deployment is significantly challenged by the dynamic nature of quantum channels and the complexity of real-time parameter optimization. In this paper, we propose OptiQKD, a protocol-agnostic machine learning framework specifically engineered to maximize the Secure Key Rate (SKR) and minimize the Quantum Bit Error Rate (QBER) for the BB84, E91, and COW protocols. OptiQKD integrates Temporal Convolutional Networks (TCNs) for high-accuracy and short-horizon forecasting of channel-state fluctuations with a Reinforcement Learning (RL) controller for autonomous and real-time parameter selection. This optimization stack is strictly constrained by standard composable-security assumptions to ensure that performance gains do not compromise the underlying quantum security. We evaluate the framework by simulating critical environmental stressors, including depolarizing and amplitude-damping noise, under realistic device constraints, including channel loss, detector efficiency, and dark counts. Our results demonstrate substantial protocol-agnostic improvements: the median SKR increases by 20--30%, while the median QBER is reduced from 3.0% to 1.5% through predictive state optimization. These findings establish that OptiQKD provides an efficient, security-preserving mechanism for dynamic parameter tuning, paving the way for more resilient and high-throughput practical QKD deployments.

Figures (6)

• Figure 1: The OptiQKD Framework integration loop. The architecture synchronizes the physical QKD layer with a two-stage ML optimization stack. Historical telemetry is processed by the TCN module to forecast channel states, which the RL agent then uses to execute real-time parameter updates, maximizing the secure key rate while suppressing errors.
• Figure 2: Architecture of the TCN module within the OptiQKD framework, illustrating the dilated convolution layers and residual pathways used for state tracking.
• Figure 3: Detailed internal architecture of the Adaptive Control Layer within the OptiQKD framework. The module employs an Actor-Critic PPO reinforcement learning agent to translate predicted states into optimal real-time parameter updates.
• Figure 4: Protocol performance comparison of our ML-optimized OptiQKD framework versus unoptimized benchmarks. Subfigure (a) (top panel) shows Secure Key Rate (bps) improvements, while subfigure (b) (bottom panel) illustrates the corresponding Quantum Bit Error Rate (QBER) reduction across BB84, E91, and COW.
• Figure 5: Protocol-specific QBER progression under linearly scaling noise conditions. The unoptimized protocols degrade rapidly, with E91 breaching the theoretical 11% abort threshold. Conversely, the ML-optimized OptiQKD framework suppresses the peak error rate for all protocols to $\approx 5\%$.