When designing quantum communication protocols, we often assume we know the exact noise profile of our quantum wires. In practice, however, environmental conditions fluctuate unpredictably, and channels can even be subject to malicious manipulation by an adversary.
To build a robust quantum internet, we must design codes that can transmit quantum information reliably without knowing the exact channel ahead of time. A powerful body of work addresses this challenge by analyzing compound channels (where the channel is unknown but fixed) and arbitrarily varying channels (where the channel changes dynamically or adversarially). Together, these papers show how universal coding strategies and pre-shared entanglement can stabilize quantum capacities and guarantee reliable transmission even in the worst-case scenarios.
The Challenge of the Unknown: Compound Channels
Imagine a scenario where a sender (Alice) wants to transmit entanglement to a receiver (Bob), but the exact quantum channel connecting them is drawn from a known set of possibilities. Alice and Bob must choose a single encoding and decoding strategy that works perfectly regardless of which channel from the set is actually deployed. This setup is known as a compound quantum channel.
In Entanglement transmission capacity of compound channels, Bjelaković et al. proved that reliable entanglement transmission is achievable under this type of uncertainty. They derived a precise mathematical formula for the capacity of finite compound channels, establishing that a single "universal" quantum code can safeguard quantum states against a whole family of noise profiles.
An intuitive question arises: if Bob could somehow measure the environment and figure out exactly which channel was used, could he decode the message more effectively? Surprisingly, the answer is no. In a companion paper, Entanglement Transmission and Generation under Channel Uncertainty: Universal Quantum Channel Coding, Bjelaković et al. demonstrated that having an "informed decoder" does not increase the entanglement transmission capacity.
The intuition behind this result lies in the fragile nature of quantum superposition. If a code space is not robust enough to protect the transmitted qubits from being entangled with the unknown environment during transmission, the quantum information is irreversibly damaged. Once that coherence is lost, no amount of clever post-processing or channel knowledge at the receiver's end can reconstruct it.
Defeating the Jammer: Arbitrarily Varying Channels
What if the channel changes with every single qubit sent, and these changes are dictated by an active adversary (a "jammer")? This is modeled as an Arbitrarily Varying Quantum Channel (QAVC).
Ahlswede et al. tackled this problem in Entanglement transmission over arbitrarily varying quantum channels, establishing a quantum analogue of a famous classical result known as the Ahlswede dichotomy. They showed that the capacity of an adversarial channel exhibits a stark threshold effect: it is either completely zero, or it is equal to its "random coding capacity" (the capacity achievable if Alice and Bob share a small amount of randomness to coordinate their codebooks).
The capacity drops to zero when a channel is symmetrizable—meaning the adversary can manipulate the channel states to perfectly mimic legitimate inputs, leaving the receiver with no physical way to distinguish between Alice's actual message and the jammer's interference.
However, the adversarial problem is not as intractable as it seems. In Fully Quantum Arbitrarily Varying Channels: Random Coding Capacity and Capacity Dichotomy, Boche et al. analyzed a fully quantum jammer who can even use entangled states across multiple channel uses to disrupt communication. They proved a remarkable equivalence: the capacity of a fully quantum QAVC is mathematically identical to the capacity of a compound quantum channel, provided Alice and Bob use a tiny, logarithmic amount of shared randomness. This means that complex, dynamic adversarial jamming can be neutralized and analyzed using the steadier, static framework of compound channels.
Entanglement as a Stabilizer
If the capacity of an adversarial channel can suddenly plummet to zero due to the capacity dichotomy, how can we engineer reliable systems? The answer lies in a beautiful irony: we can use pre-shared entanglement to protect the transmission of classical and quantum information.
Janßen et al. demonstrated in Entanglement-assisted classical capacities of compound and arbitrarily varying quantum channels that providing Alice and Bob with a background supply of shared entanglement completely eliminates the capacity dichotomy.
Why does shared entanglement stabilize the channel? In an unassisted QAVC, the adversary exploits symmetry to confuse the receiver. But when Alice and Bob share entangled states, they possess non-local correlations that are completely inaccessible to the jammer. By performing measurements on their respective halves of the entangled pairs, they can establish a secure, coordinate-free baseline that the adversary cannot spoof. This breaks the channel's symmetrizability, transforming a highly volatile, discontinuous capacity into a smooth, predictable function of the channel noise.
This stabilizing effect extends to multi-user networks as well. In The Multiple-Access Channel With Entangled Transmitters, Pereg et al. explored how entanglement shared between two distinct transmitters can enhance communication rates over a multiple-access channel (MAC). They showed that this shared resource allows transmitters to coordinate their signals through quantum measurements, making the achievable transmission rates exceptionally robust and independent of the specific error criteria used to evaluate the network.
Resource Matching and the Reconstruction Floor
The interplay between uncertainty, adversarial capacity, and shared resources points to a deeper geometric principle: quantum communication protocols succeed when their constituent resources are properly matched.
Long before the exact capacity formulas for QAVCs were derived, Li et al. identified this bottlenecking behavior in Probabilistic teleportation and entanglement matching. They analyzed quantum teleportation over partially entangled channels and showed that the maximum success probability is strictly limited by the "weakest link" between the entanglement of the channel itself and the entanglement of the sender's measurement. If a sender performs an overly strong measurement on a narrow, weakly entangled channel, the excess measurement capability is wasted. Optimal throughput requires precisely matching the sender's "sending ability" to the channel's "width."
This concept of resource matching provides a fundamental bound for modern quantum security protocols. In Approximate reconstructability of quantum states and noisy quantum secret sharing schemes, Ouyang et al. established a strict duality between an adversary's capacity and a legitimate recipient's reconstruction error.
They proved that in a noisy quantum secret sharing scheme, the fidelity loss experienced by authorized players is directly dictated by the entanglement-assisted capacity of the adversarial channel. This works because of the monogamy of entanglement: a quantum state can only be highly entangled with one system at a time. The maximum rate at which an adversary can extract information through their channel sets a concrete physical floor on how accurately the legitimate recipients can reconstruct the secret. If the channel allows information to leak to the adversary, it mathematically forces a corresponding reconstruction error for the receivers, cementing the role of entanglement capacity as a universal currency for channel robustness.
Key Foundational Work
Probabilistic teleportation and entanglement matching (2000) — contributed 6 concept elements to this direction.
Entanglement transmission capacity of compound channels (2009) — contributed 4 concept elements to this direction.
Entanglement Transmission and Generation under Channel Uncertainty: Universal Quantum Channel Coding (2009) — contributed 1 concept elements to this direction.