Self-Tuning Transmitter for Quantum Key Distribution Using Machine Intelligence

TL;DR

This paper presents an autonomous quantum key distribution transmitter that uses machine intelligence, specifically a genetic algorithm, to optimize laser parameters for high performance without prior detailed knowledge.

Contribution

It introduces a self-tuning QKD transmitter employing genetic algorithms to optimize laser dynamics automatically, simplifying setup and enhancing performance.

Findings

Achieved autonomous optimization of phase coherence and quantum bit error rate.

Matched state-of-the-art performance levels without prior parameter knowledge.

Demonstrated effective use of genetic algorithms for laser system tuning.

Abstract

The development and performance of quantum technologies heavily relies on the properties of the quantum states, which often require careful optimization of the driving conditions of all underlying components. In quantum key distribution (QKD), optical injection locking (OIL) of pulsed lasers has recently been shown as a promising technique to realize high-speed quantum transmitters with efficient system design. However, due to the complex underlying laser dynamics, tuning such laser system is both a challenging and time-consuming task. Here, we experimentally demonstrate an OIL-based QKD transmitter that can be automatically tuned to its optimum operating state by employing a genetic algorithm. Starting with minimal knowledge of the laser operating parameters, the phase coherence and the quantum bit error rate of the system are optimized autonomously to a level matching the state of the art.