arXiv:2609.35339
Zero Knowledge Proofs in Quantum Networks
Tuhin Paul, Srijani Das, Manasi Patra, Ramij Rahaman
Abstract
Zero-knowledge proofs (ZKPs) enable the verification of a statement without revealing any information beyond its validity and constitute a fundamental primitive in cryptography and information theory. However, existing constructions rely on computational assumptions and are predominantly confined to bipartite settings, leaving their information-theoretic realization in bipartite or network scenarios largely unexplored. Here we develop a framework for zero-knowledge verification based on the indistinguishability of quantum states under operational constraints. Exploiting the fundamental limitations imposed by local operations, we show that a verifier is inherently restricted from extracting information about the underlying state while retaining the ability to verify correctness. We construct explicit protocols for multiparty quantum networks that achieve information-theoretic security, ensuring that no subset of collaborating parties can gain knowledge beyond the validity of the statement, independent of their joint computational power.
AI Summary
The paper asks whether a prover can demonstrate knowledge of the identity of shared entanglement without revealing that identity. It studies promised ensembles of Bell and GHZ states, with the prover holding one subsystem and knowing the label of each shared copy. The constructions progress from outcome-prediction tests to device-independent certification. A single-basis Bell test admits separable simulations; adding a complementary basis excludes that example but still permits higher-dimensional simulations with commuting measurements. The device-independent Bell protocol instead uses a secret, label-dependent Pauli correction to map every input to a singlet, followed by CHSH verification at $S\geq2\sqrt2-\epsilon$. The multipartite version maps four GHZ variants to a canonical GHZ state and tests $n+1$ parity constraints for odd $n$. Its Bell operator has local bound $n-1$ and quantum maximum $n+1$. Zero-knowledge arguments use the corrected state's independence from the original label and simulators prepared with a fixed reference state. The paper gives an ideal multipartite soundness bound $[n/(n+1)]^k$ for $k$ independent rounds. It also presents GHZ self-testing arguments, a noise analysis reporting verifier-view trace distance at most $\varepsilon_{\mathrm{ZK}}/n$ for Bell deficiency $\varepsilon_{\mathrm{ZK}}$, and an entropy-accumulation treatment of coherent attacks. The setting assumes specified candidate ensembles and secret state labels; verification against a known state is excluded. The multipartite device-independent construction requires odd $n$. Ideal simulator arguments assume auxiliary registers initially independent of the resource, while coherent-attack entropy rates require independently certified single-round bounds.
Central claim may not hold: The device-independent GHZ protocol reveals the global test index to the prover, enabling constant-output verifier devices and a context-dependent prover response to satisfy every parity test without entanglement or state-identity knowledge. The soundness analysis assumes uniform hidden-state mixtures and a special informed/uninformed model of partial knowledge without establishing security for the broader preparation distributions and quantum side information claimed. The robustness lemmas falsely require approximate states to be incoherent target/orthogonal mixtures and claim a linear trace-distance bound contradicted by explicit mixtures and coherent superpositions of GHZ-basis eigenstates. The device-independent self-testing proof applies a fixed-Pauli uniqueness lemma to arbitrary Jordan blocks without proving the necessary measurement rigidity. The final soundness theorem miscalculates the local parity-game success probability as (n-1)/(n+1), whereas the correct attainable value is n/(n+1).
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- Comments
- 18 pages, 1 figure