In space, photons can travel for hundreds and hundreds of kilometers with a small probability of getting lost. This is a luxury that doesn’t apply to fiber. Imagine that you have a photon source that releases 10 billion photons every second. With a fiber connection in which a bit less than one out of twenty photons gets lost every kilometer, after 500 kilometers the rate has dropped to roughly one photon arriving every second. After 1000 kilometers of fiber there is such an immense drop in the rate that you might not even be able to measure any photon in your own lifetime anymore: after 1000 kilometers of fiber around one photon will arrive every 300 years. Therefore, the signal needs to be refreshed.
Refreshing the signal with classical nodes
In a pre-quantum network, such as the ones in China and Japan, this is the job of classical nodes. At the node, the photons that arrive are collected and their state is measured. Afterwards, fresh photons in the right state are sent towards the next node in the chain.
The next stage: quantum repeaters
In stages beyond the pre-quantum network, the aim is to refresh the signal with a different kind of node: a quantum repeater. True quantum repeaters have not been realized yet, but in the near future the basis for such repeaters will be laid on the Dutch quantum network. In order to understand how quantum repeaters work, it’s important to understand a different concept first: quantum teleportation.
Sangouard, N. et al. (2011). Quantum repeaters based on atomic ensembles and linear optics. Rev. Mod. Phys. 83, 33-80.