On noise in swap ASAP repeater chains: exact analytics, distributions and tight approximations

TL;DR

This paper provides exact analytical formulas and tight approximations for the fidelity distribution in quantum swap ASAP repeater chains, enhancing understanding of entanglement distribution and optimizing fidelity with cut-off policies.

Contribution

It introduces exact formulas for fidelity moments, a generating function approach, and tight approximations for the fidelity distribution in quantum repeater chains, including optimization methods.

Findings

Exact formulas for moments of fidelity up to 25 segments.

Tight exponential approximations for average fidelity.

Analytical calculation of secret-key rate with and without binning.

Abstract

Losses are one of the main bottlenecks for the distribution of entanglement in quantum networks, which can be overcome by the implementation of quantum repeaters. The most basic form of a quantum repeater chain is the swap ASAP repeater chain. In such a repeater chain, elementary links are probabilistically generated and deterministically swapped as soon as two adjacent links have been generated. As each entangled state is waiting to be swapped, decoherence is experienced, turning the fidelity of the entangled state between the end nodes of the chain into a random variable. Fully characterizing the (average) fidelity as the repeater chain grows is still an open problem. Here, we analytically investigate the case of equally-spaced repeaters, where we find exact analytic formulae for all moments of the fidelity up to 25 segments. We obtain these formulae by providing a general solution in terms of a generating function; a function whose n'th term in its Maclaurin series yields the moments of the fidelity for n segments. We generalize this approach as well to a global cut-off policy -- a method for increasing fidelity at the cost of longer entanglement delivery times -- allowing for fast optimization of the cut-off parameter by eliminating the need for Monte Carlo simulation. We furthermore find simple approximations of the average fidelity that are exponentially tight, and, for up to 10 segments, the full distribution of the delivered fidelity. We use this to analytically calculate the secret-key rate, both with and without binning methods.