Quantifying the effect of polarization drift on continuous-variable QKD security
Almeida, M.
; Guilherme, Í. G.
;
Pinto, A. N.
;
Silva, N. A.
Quantifying the effect of polarization drift on continuous-variable QKD security, Proc SPIE Photonics Europe, Strasbourg, France, Vol. , pp. - , April, 2026.
Digital Object Identifier:
Abstract
Continuous-Variable (CV)-Quantum Key Distribution (QKD) addresses the key distribution problem encoding the information using coherent states, ensuring security by monitoring the noise of the exchanged states. This approach offers high key rates and compatibility with standard telecommunication infrastructure, with fiberdeployed CV-QKD systems extensively demonstrated. However, the long-term stability of these systems remains challenged by the polarization drift. While polarization effects are often assumed negligible or actively compensated, their impact on the secret key rate (SKR) over extended operation has not been fully quantified. In this work, we present a comprehensive analysis of the temporal evolution of the Stokes parameters and SKR under varying polarization drift variances, adopting a complete stochastic approach to emulate the polarization drift. Through extensive simulations, we identify three distinct regimes: an ideal-case approximation with negligible effects; a transition region with progressive signal distortion, biased channel parameter estimation, and SKR degradation; and a worst-case scenario leading to angular displacements near 90◦, collapsed correlation, and null SKR. The polarization coherence time rapidly decreases with increasing polarization drift variance, from nearly one hour at 10−14 to sub-second durations above 10−11. These findings offer critical insights into the interplay between polarization dynamics and system security, informing the design of robust CV-QKD deployments. While active compensation of polarization drift using dynamic controllers require continuous monitoring, passive strategies such as polarization-diversity receivers and digital signal processing-based techniques provide practical alternatives for field-deployed CV-QKD systems.