# Techniques: Establishing shared secret keys on quantum line networks: protocol and security

## What the paper contributes (one paragraph)

The paper studies multi-user secret key establishment on a quantum line network (a "Qline"), where qubit generation and measurement happen at the two ends of a single fibre and intermediate nodes perform only single-qubit unitary transforms. It presents the key establishment protocol, which first runs a quantum-assisted secret sharing phase yielding keys that XOR to zero and then discloses one party's key to pair the others, scheduled round-robin. Its main result is a complete, formal security proof of this protocol in the abstract cryptography framework, leveraging existing prepare-and-measure QKD security to show the Qline protocol is 3-epsilon-secure. Because the proof is in abstract cryptography, the resulting security is composable, so the keys can be combined with authentication, one-time-pad encryption, or other key encapsulation mechanisms. The architecture removes the trusted-node and key-routing assumptions of QKD networks, lowering trust requirements for last-mile quantum networking. The arxiv id is 2304.01881.

## Techniques offered

- **Composable security proof in the abstract cryptography framework** - A formal proof that the Qline key establishment protocol securely realises its ideal key functionality, giving composable security usable with downstream cryptographic tasks.
  - guarantee: interactive composable security bound; the protocol is 3-epsilon-secure under the stated assumptions, where epsilon is the security parameter of a standard QKD protocol.
  - quote: "Under Assumptions~\ref{assump:general},~\ref{assump:Alice},~\ref{assump:Bob}~and~\ref{assump:qline}, the key establishment protocol on Qline is $3\epsilon$-secure in the abstract cryptography framework, where $\epsilon$ is the security parameter of a standard QKD protocol under the same"

- **Quantum key establishment on quantum line networks (Qline)** - A networking protocol that lets any pair of nodes on a single quantum communication channel establish a shared secret key without trusted nodes or key routing through intermediate parties.
  - guarantee: unconditional key secrecy against eavesdropping, inherited from prepare-and-measure QKD; intermediate parties never learn other parties' keys.
  - quote: "The most important feature is that  Qline can enable key exchange among participants but unlike QKD networks, there is no need for a trusted node set-up assumption."

- **Quantum-assisted secret sharing on a line** - A secret sharing primitive in which single-qubit states are passed along the line so the parties obtain keys whose XOR is zero, forming the building block from which pairwise keys are derived.
  - guarantee: correctness of the shared correlation, the three keys satisfy K_A XOR K_B XOR K_C = 0; secrecy follows from the security proof above.
  - quote: "The first step leads the parties to output three keys $K_A$, $K_B$ and $K_C$ such that $K_A\oplus K_B \oplus K_C = 0$. This is the secret-sharing phase."

## Where it could apply

- Target primitive(s): none (this is a quantum communication and cryptography protocol, not a gate-model algorithm primitive such as VQE, QAOA, QPE, sampling, or Trotterised simulation).
- Target application group(s): platforms_tooling, networking and quantum cryptography (key establishment, secret sharing, secure multi-party computing on line networks).
- Code family / hardware assumptions: no quantum error correction; prepare-and-measure photonic hardware with BB84 states, single-photon source at Alice, single-photon detector at Bob, and intermediate nodes restricted to single-qubit unitaries (for example polarisation modulators).

## Caveats

The techniques are quantum-communication and cryptography results rather than capabilities suppliable to a fault-tolerant or gate-model computational result. The security proof reuses existing QKD security and depends on the prepare-and-measure QKD assumptions plus a Qline-specific assumption; the analysis covers only the quantum communication and sifting phase, deferring error correction and privacy amplification to standard QKD post-processing.