On the accuracy of twirled approximations in repeater chains

Abstract

In the performance analysis of quantum networks, it is common to approximate bipartite entangled states as either being Bell-diagonal or Werner states. We refer to these as twirled approximations because it is possible to bring any state to such a form with a twirling map. Although twirled approximations can simplify calculations, they can lead to an inaccuracy in performance estimates. The goal of this work is to quantify this inaccuracy. We consider repeater chains where end-to-end entanglement is achieved by performing an entanglement swap at each repeater in the chain. We consider two scenarios: postselected and non-postselected entanglement swapping, where postselection is performed based on the Bell-state measurement outcomes at the repeaters. We show that, for non-postselected swapping, the Bell-diagonal approximation is exact for the computation of the Bell-diagonal elements of the end-to-end state. We also find that the Werner approximation accurately approximates the end-to-end fidelity when the infidelity of each initial state is small with respect to the number of repeaters in the chain. For postselected swapping, we find bounds on the difference in end-to-end fidelity from what is obtained with the twirled approximation, for initial states with a general noisy form. Finally, for the example of performing quantum key distribution over a repeater chain, we demonstrate how our insights can be used to understand how twirled approximations affect the secret-key rate.