Short answer: Sending entangled light through the atmosphere and from orbit loses signal to diffraction, absorption and turbulence. A 2017 Science experiment reported entangled photon pairs distributed to two ground sites 1203 kilometres apart. A 2020 npj Quantum Information paper and a 2024 Communications Physics paper analyse constellation design and turbulence effects in theory.
Why use satellites for entanglement?
The 2020 npj Quantum Information abstract says satellite links are viable in the near term because optical signals attenuate less from satellite to ground and because quantum memories currently have short coherence times. That short coherence prevents ground-based entanglement distribution at high rates over long distances through the atmosphere or fibre. The authors treat this as the reason for studying satellite-based links.
What was demonstrated in 2017?
The Science paper states that owing to channel loss, the distance previously achieved for entanglement distribution was limited to about 100 kilometres. The authors report satellite-based distribution of entangled photon pairs to two locations separated by 1203 kilometres on Earth, through two satellite-to-ground downlinks with summed length varying from 1600 to 2400 kilometres. They observed survival of two-photon entanglement and a violation of the Bell inequality by 2.37 plus or minus 0.09 under strict Einstein locality conditions. They state that the effective link efficiency is orders of magnitude higher than direct bidirectional transmission of the two photons through telecommunication fibres.
What does the theory say about constellations?
The 2020 paper proposes a global-scale quantum internet made of orbiting satellites providing on-demand entanglement distribution to ground stations, which could also act as untrusted nodes for long-distance quantum key distribution. The authors develop a technique for finding satellite configurations with continuous coverage that balance the number of satellites against entanglement-distribution rates. They compute configurations for a polar-orbit constellation and analyse the resulting satellite-to-ground loss and rates for several ground station setups, and compare these rates with ground-based repeater schemes. This is a design study, and the abstract presents it as theoretical tools and guidance.
What about turbulence?
The 2024 Communications Physics paper studies atmospheric turbulence in continuous-variable entanglement distribution and quantum teleportation between a ground station and a satellite. It examines degradation from diffraction, atmospheric attenuation, turbulence and detector inefficiency, for both downlink and uplink. Because teleportation fidelity with the distributed resource is not sufficient on its own, the authors add an intermediate station, either for state generation or for beam refocusing. Their results show feasibility of free-space entanglement distribution and teleportation in downlinks up to the low Earth orbit region, and in uplinks with the help of the intermediate station. They also compare microwave and optical channels in bad weather and study horizontal ground-to-ground and inter-satellite paths.
How to read this
One of these papers is an experiment and two are theoretical studies, so the constellation and turbulence results are modelled and not measured. This post relies on abstracts and visible summary text and has not reviewed full texts. It makes no claim about commercial services or deployed networks.
FAQ
How far did the 2017 experiment distribute entanglement?
Between two ground locations 1203 kilometres apart, according to the Science abstract.
Why is the downlink easier than the uplink?
The 2024 paper finds feasibility in downlinks up to low Earth orbit, but needs an intermediate station for uplinks. The abstract does not give a full physical explanation.
What is a satellite quantum internet proposal?
A constellation providing entanglement on demand to ground stations, as proposed in the 2020 paper, which is a design study.
Does this replace fibre links?
The abstracts do not say so. The 2017 paper compares efficiency with direct fibre transmission of the same photons.
All field notes