Mode Mismatch Mitigation in Gaussian-Modulated CV-QKD
🤖 AI Summary
In Gaussian-modulated continuous-variable quantum key distribution (CV-QKD), pulse shaping limitations at the transmitter induce mode mismatch between the local oscillator and signal pulses, degrading performance. To address this, this work introduces— for the first time—machine learning into transmitter-side pulse waveform design, proposing a data-driven adaptive optimization framework. A neural network jointly optimizes the temporal pulse shape and modulation parameters, integrated with high-fidelity pulse shaping and channel transmission simulations to suppress mode mismatch. Experimental evaluation demonstrates a 47% improvement in asymptotic secure key rate at 50 km standard optical fiber under typical channel conditions, while maintaining strong generalization across diverse channel scenarios. This approach establishes a novel paradigm for overcoming practical bottlenecks in CV-QKD deployment and extends the applicability of intelligent algorithms to physical-layer design in quantum communications.
Application Category
📝 Abstract
Technical limitations in pulse shaping lead to mode mismatch, which significantly reduces the secure key rate in CV-QKD systems. To address this, a machine learning approach is employed to optimize the transmitter pulse-shape, effectively minimizing mode mismatch and yielding substantial performance improvements.
Problem
Research questions and friction points this paper is trying to address.
Mitigate mode mismatch in Gaussian-modulated CV-QKD systems
Optimize transmitter pulse-shape using machine learning
Improve secure key rate by reducing mode mismatch
Innovation
Methods, ideas, or system contributions that make the work stand out.
Machine learning optimizes transmitter pulse-shape
Minimizes mode mismatch in CV-QKD
Improves secure key rate performance
🔎 Similar Papers
No similar papers found.
