Effectiveness of the DEJMPS purification protocol in noisy entangled photon systems, a Monte Carlo simulation

TL;DR

This study uses Monte Carlo simulations to evaluate the DEJMPS entanglement purification protocol's effectiveness on noisy photon pairs, providing detailed fidelity and yield maps across various noise levels.

Contribution

It offers the first comprehensive numerical analysis of both fidelity and yield improvements for DEJMPS under combined amplitude-damping and dephasing noise.

Findings

Single purification round can increase fidelity by up to 0.07 in high-noise regimes.

Yield can decrease by up to 55% after purification.

Fidelity gains increase with noise levels, yields decline more sharply under combined noise.

Abstract

Entanglement purification is a critical enabling technology for quantum communication, allowing high-fidelity entangled pairs to be distilled from noisy resources. We present a comprehensive Monte Carlo study of the DEJMPS purification protocol applied to polarization-entangled photon pairs subject to both amplitude-damping noise (gamma) and dephasing noise (p). By sweeping (gamma, p) over a two-dimensional grid and performing repeated stochastic trials, we map out the average fidelity and average yield surfaces of the purified output, as well as the net gains (DF) and losses (DY) relative to the unpurified baseline. Our results show that a single round of DEJMPS purification can boost entanglement fidelity by up to 0.07 in high-noise regimes, while incurring a yield penalty of up to 0.55. Fidelity gains grow monotonically with both gamma and p, whereas yields decline more sharply under combined noise. Contour and 3D surface plots of DF(gamma, p) and (DY, gamma, p) vividly illustrate the trade-off between quality and quantity of distilled pairs. This two-parameter Monte Carlo characterization provides practical guidance for optimizing purification depth and operating points in real-world photonic networks, and represents, to our knowledge, the first detailed numerical charting of both fidelity and yield improvements across a continuous noise landscape for DEJMPS.