Quantum Advanced Quantum Networks Simulation

Entanglement distribution over fibre: loss, repeater chains, swapping and fidelity

Presets

Parameters

Live readouts

Running check...
The science

Why repeaters. A photon survives a fibre of length L with probability eta = 10^(-alpha L/10) (alpha is the attenuation in dB/km, typically about 0.2 at 1550 nm). Without repeaters, no protocol can beat the PLOB bound R <= -log2(1 - eta) bits per channel use (Pirandola et al., 2017), i.e. about 1.44 eta, which falls exponentially with distance. Splitting the path into n links with heralded entanglement and memories turns this exponential into a much gentler polynomial-like cost.

Rate model. Each link succeeds per attempt with p = 10^(-alpha L/(10 n)). With ideal quantum memories, all n links are collected in parallel: the expected number of rounds until all are ready is the expected maximum of n geometric variables, E = sum_k (-1)^(k+1) C(n,k) / (1 - (1-p)^k). Each of the n-1 entanglement swaps (Bell-state measurements) succeeds with probability ps, and if any fails all links are discarded (a pessimistic restart policy), so rounds = E / ps^(n-1). One round lasts 1/R0 + (L/n)/v, the source period plus the one-way herald delay at v = 2x10^5 km/s.

Fidelity. Links are Werner states of fidelity F0 with Werner parameter w = (4F0 - 1)/3. Swapping two links gives F = F1 F2 + (1-F1)(1-F2)/3; after n links, F = (1 + 3 w^n)/4. Entanglement survives only while F > 1/2, which is why many noisy links need purification (not modelled).

Checks in the page. (1) A Monte Carlo of the expected maximum of n geometric waiting times against the inclusion-exclusion formula; (2) the step-by-step swap recursion against the closed form w^n.

Simplifications and honesty. Perfect memories (the readout shows the memory time you would need), no multiplexing, no purification, no dark counts, uniform links, no swap-depth optimisation (nested protocols are faster). Parameter values are typical-order illustrations, not a specific experiment.