Optimization of CV-QKD Under Practical Constraints

2026-05-03

AI Summary

This work addresses the performance degradation of continuous-variable quantum key distribution (CV-QKD) in practical deployments, which arises from hardware limitations including finite-tap FIR filters, constraints on the average photon number, and limited-precision digital-to-analog and analog-to-digital converters (DAC/ADC). To overcome these challenges, the study introduces reinforcement learning into CV-QKD systems for the first time, enabling end-to-end joint optimization under multiple hardware constraints. This approach transcends the conventional modular design paradigm by simultaneously optimizing all relevant components while maintaining system stability, thereby achieving a significant increase in secret key rate. The proposed method establishes a new framework for realizing high-performance, practical CV-QKD systems.

Abstract

Using reinforcement learning, we optimize for practical hardware constraints, including limited FIR filter taps at the transmitter and receiver, mean photon number and finite DAC/ADC resolution. Under these realistic conditions, the proposed approach achieves significant performance improvements.

Problem

• CV-QKD
• practical constraints
• FIR filter taps
• photon number
• DAC/ADC resolution

Innovation

• reinforcement learning
• CV-QKD
• hardware constraints
• FIR filter
• DAC/ADC resolution