An Unsupervised Machine Learning to Optimize Hybrid Quantum Noise Clusters for Gaussian Quantum Channel

This work focuses on optimizing the hybrid quantum noise model to improve the capacity of Gaussian quantum channels using Machine Learning (ML) generated clusters. The work specifically leverages Gaussian Mixture Model (GMM) and the Expectation-Maximization (EM) algorithm to model the complex noise characteristics of quantum channels. Hybrid quantum noise, which includes both quantum shot noise and classical Additive-White-Gaussian Noise (AWGN), is modeled as an infinite mixture of Gaussian distributions weighted by Poissonian parameters. The study proposes a method to reduce the number of clusters within this noise model, simplifying visualization and improving the accuracy of channel capacity estimations without compromising essential noise characteristics. Key contributions include the reduction of Gaussian clusters while maintaining error tolerances and using the EM algorithm to update quantum channel parameters, leading to more accurate channel capacity. The approach is validated through simulations, demonstrating that ML-enhanced quantum noise clustering significantly improves the channels performance in satellite-based quantum communication systems, specifically for Quantum Key Distribution (QKD). The work demonstrates that GMM and EM algorithms provide a practical solution for modeling quantum noise in real-time applications, advancing the optimization of quantum communication networks.