The Signal
Waterloo Researchers Set a Direct-Link Ceiling for Gaussian Repeaters
A Physical Review Letters paper by University of Waterloo researchers finds that Gaussian repeater chains cannot beat direct transmission over a pure-loss optical channel. The proof covers Gaussian operations, homodyne measurements and clas
Quantum.world Team · Written with AI · fact-checked ·
Gaussian repeater chains gain no quantum or private capacity over direct transmission in a pure-loss optical channel, according to a theorem published October 23 in Physical Review Letters. The result rules out a broad class of intermediate-node schemes as a way to overcome photon loss: adding Gaussian processing stations does not raise the channel’s capacity above the direct-link benchmark.
The paper, “No-Go Theorem for Gaussian Quantum Repeaters from Fractional Extendibility,” is by Rabsan Galib Ahmed and Graeme Smith of the University of Waterloo. It addresses a question about whether repeaters built from experimentally accessible Gaussian tools can improve communication rates without using non-Gaussian operations.
A ceiling for Gaussian repeater chains
The theorem covers repeater chains using Gaussian operations, homodyne measurements and arbitrary classical communication. Those tools cannot increase the quantum or private capacity of a pure-loss attenuation channel beyond what direct transmission allows. The paper also reports that arbitrary Gaussian local operations and classical communication among repeater nodes cannot improve the channel’s two-way capacities.
“In this work, we prove a no-go theorem for Gaussian quantum repeaters in a quantum network,” Ahmed and Smith write. The result concerns capacity—the ultimate rate permitted by the channel model—not whether a particular repeater can improve an experiment’s performance against practical limitations such as imperfect equipment or restricted resources.
Fractional extendibility supplies the proof
The authors’ argument introduces fractional extendibility, a generalization of the extendibility framework used to analyze quantum states. In their treatment, the quantity tracks a constraint on correlations as states pass through attenuation channels and Gaussian operations. The framework’s monotonicity and composition properties let the authors follow that constraint across a repeater chain, then bound the communication rate at the endpoints.
That mathematical approach matters because a repeater is not simply a single amplifier inserted into a line. Nodes can prepare shared states, perform measurements, exchange classical messages and coordinate operations along the route. The proof is designed to cover that chain of permitted Gaussian steps, rather than only one specific relay design.
The result is limited to a defined channel and toolbox
The theorem is not a claim that all quantum repeaters fail, or that Gaussian methods have no role in quantum communications. Its boundary is the pure-loss attenuation channel and the specified Gaussian operations. It does not establish that a real network with noise, hardware imperfections and other channel models has exactly the same capacity bound.
For researchers designing optical repeaters, the implication is direct: a protocol seeking rates beyond direct transmission must leave the Gaussian-only toolbox under these assumptions. The paper identifies non-Gaussian operations as necessary for surpassing the direct-link rate in its model, but does not show that adding them alone guarantees a higher rate. The remaining question is how the fractional-extendibility method—and the no-go result—extends to more complicated Gaussian networks and less idealized channels.