Noisy quantum input loophole in measurement-device-independent entanglement witnesses

TL;DR

This paper investigates how noise in quantum inputs affects measurement-device-independent entanglement witnesses, revealing conditions under which false positives can occur due to noisy quantum inputs.

Contribution

It introduces a general framework for analyzing the impact of noisy quantum inputs on MDI-EWs, relaxing previous ideal assumptions and identifying conditions for reliable entanglement detection.

Findings

Uniform noise can be tolerated under certain conditions

Non-uniform and entangling noise may cause fake entanglement detection

Conditions for detecting entanglement in Werner states under local noise

Abstract

Entanglement witnesses form an effective method to locally detect entanglement in the laboratory without having the prior knowledge of the full density matrix. However, separable states can be erroneously indicated as entangled in such detections in the presence of wrong measurements or loss in detectors. Measurement-device-independent entanglement witnesses (MDI-EWs) never detect fake entanglement even under wrong measurements and for a particular kind of lossy detectors. A crucial assumption in the case of faithful detection of entanglement employing MDI-EWs is that the preparation devices producing "quantum inputs" - which are inputs additional to the quantum state whose entanglement is to be detected - are perfect and there is no noise during their transmission. Here, we relax these assumptions and provide a general framework for studying the effect of noise on the quantum inputs, invoking uniform and non-uniform noise models. We derive sufficient conditions on the uniform noisy map for retaining the characteristic of MDI-EWs. We find that in the context of non-uniform and entangling noise, fake entanglement detection is possible even by MDI-EWs. We also investigate various paradigmatic models of local noise and find conditions of revealing entanglement in the class of Werner states.