Quantum Repeater

Quantum repeaters solve one of the most fundamental obstacles to building a global quantum internet: photons get lost in fiber, and you cannot simply amplify a quantum signal the way classical networks do.

Why photons disappear in fiber

Optical fiber attenuates light at roughly 0.2 dB per kilometer. Over 100 km that is a 20 dB loss, meaning 99% of photons never arrive. Classical networks handle this trivially: amplifiers read the signal and retransmit it. Quantum mechanics prohibits exactly that strategy. The no-cloning theorem proves that an unknown quantum state cannot be copied. An amplifier that works by copying the signal would have to clone an unknown qubit, which is impossible. So photon loss is not just an engineering inconvenience; it is a fundamental barrier that requires a genuinely different solution.

Entanglement-based repeaters

The solution is to replace signal amplification with entanglement distribution. Instead of sending a qubit directly from Alice to Bob over 1,000 km, the network creates entanglement between neighboring nodes separated by shorter, manageable distances, then connects those short-range links into a long-range link.

The two key operations are:

Entanglement swapping. Suppose node A shares an entangled pair with node B, and node B also shares an entangled pair with node C. Node B performs a Bell state measurement on its two qubits. This measurement destroys node B’s qubits but projects nodes A and C into an entangled state with each other, even though they have never interacted. The entanglement has been “swapped” across the network.

Entanglement purification. Entanglement distributed over real fiber is noisy. Purification (also called distillation) takes multiple low-fidelity entangled pairs and uses local operations and classical communication to produce fewer, higher-fidelity pairs. It is a way of concentrating entanglement quality at the cost of quantity.

Three generations of repeater

Researchers categorize repeater designs into three generations based on how they handle errors and losses.

Generation 1 repeaters use heralding to compensate for loss but rely on post-selection rather than quantum error correction. When an entangled photon arrives at a node, a heralding signal (a classical message) confirms it arrived. Links are built one at a time, verified by heralding, then swapped. These are the most practical near-term designs but are slow because each link must succeed before proceeding.

Generation 2 repeaters add quantum memory. Nodes store one end of an entangled pair while waiting for the other link to succeed. Memory allows links to be established in parallel rather than sequentially, dramatically improving throughput. The bottleneck shifts to memory coherence time: the memory must hold fidelity long enough for neighboring links to complete.

Generation 3 repeaters incorporate full quantum error correction, encoding logical qubits into many physical qubits and correcting errors on the fly. These would achieve the performance of an ideal repeater but require fault-tolerant quantum hardware at each node, a capability that remains far from practical.

Rate-distance tradeoff

Even below the loss limit, repeaters face a rate-distance tradeoff. Each link attempt succeeds with some probability $p$ that falls with distance. Chaining $n$ links together without memory requires all $n$ to succeed simultaneously, which happens with probability $p^n$. This drops exponentially with the number of links.

Memory changes this. A node can hold a successful link while the neighboring link is retried. The probability of connecting two memoried links is proportional to $p^2$ (at worst), not to the probability of simultaneous success. This converts exponential scaling to polynomial scaling, which is the core engineering motivation for quantum memory in repeaters.

The time to establish end-to-end entanglement also determines the useful storage time requirement: the memory must hold fidelity for as long as it takes the slowest neighboring link to succeed. For a 1,000 km network split into 10-km segments, each segment succeeds quickly, but synchronization across segments demands memory coherence times of milliseconds to seconds.

Current demonstrations

Direct fiber-based quantum communication is limited to roughly 100-200 km in practice. Beyond that, losses make the raw rate of arriving photons too low to be useful. Several groups have demonstrated key pieces of repeater technology in isolation: entanglement swapping, purification, and short-lived quantum memories.

Satellite-based links sidestep the fiber loss problem by transmitting photons through vacuum, where loss is far lower over long distances. China’s Micius satellite demonstrated entanglement distribution between ground stations separated by over 1,200 km. This is not a repeater in the strict sense (the satellite acts as a trusted relay), but it demonstrates that intercontinental quantum links are physically achievable.

Hardware platforms for repeater nodes

Different physical systems are being pursued as the hardware basis for quantum repeater nodes. Each involves tradeoffs between storage time, efficiency, and how well the system interfaces with fiber-optic photons.

Atomic ensembles and rare-earth crystals naturally couple to photons and can store entanglement in collective atomic states. The DLCZ protocol, proposed in 2001, laid out a concrete scheme using atomic ensembles that remains influential. Rare-earth crystals such as europium-doped yttrium orthosilicate have achieved storage times of seconds with moderate efficiency.

Nitrogen-vacancy (NV) centers in diamond operate at room temperature and can store a qubit for milliseconds in the electron spin or seconds in nearby nuclear spins. They interface with photons in the visible range, which requires wavelength conversion to telecom wavelengths for fiber transmission.

Trapped ions and neutral atoms in optical tweezers offer some of the longest coherence times among any physical qubit. Both can in principle be integrated with photon interfaces, making them candidates for high-performance repeater nodes, though the engineering complexity is substantial.

Why this matters for the quantum internet

Quantum repeaters are the missing piece between today’s point-to-point quantum key distribution links and a true quantum internet. Without them, QKD networks must rely on trusted nodes: classical computers that decrypt and re-encrypt at each relay, which creates security vulnerabilities. A genuine quantum repeater preserves the quantum state end-to-end, enabling unconditionally secure key distribution and, eventually, distributed quantum computing across continental distances.

See also

• Quantum Internet
• Quantum Key Distribution
• Quantum Teleportation
• Entanglement
• No-Cloning Theorem