Abstract: We introduce and analyse approximate quantum secret sharing in a formal cryptographic setting, wherein a dealer encodes and distributes a quantum secret to players such that authorized structures (sets of subsets of players) can approximately reconstruct the quantum secret and omnipotent adversarial agents controlling non-authorized subsets of players are approximately denied the quantum secret. In particular, viewing the map encoding the quantum secret to shares for players in an authorized structure as a quantum channel, we show that approximate reconstructability of the quantum secret by these players is possible if and only if the information leakage, given in terms of a certain entanglement-assisted capacity of the complementary quantum channel to the players outside the structure and the environment, is small.
Introduction
Quantum secret sharing (QSS) is a fundamental protocol in quantum cryptography where a dealer distributes a quantum state among several players. The goal is to ensure that only authorized subsets of players can reconstruct the secret, while unauthorized subsets gain no information. Traditional QSS research often focuses on "perfect" scenarios with ideal threshold schemes. However, real-world quantum channels are inherently noisy, and adversaries may be "omnipotent," capable of attacking any share they control.
This paper addresses the gap between idealized models and practical implementations by providing a formal framework for Approximate Quantum Secret Sharing (QSS). The authors move beyond perfect reconstruction to quantify the trade-off between how well authorized players can recover a secret and how much information is leaked to an adversary.
Methodology
The authors model the encoding and distribution of a secret as a quantum channel. To analyze approximate reconstructability, they employ several tools from quantum information theory:
- Adversary Modeling: An adversary is defined as an entity that can apply any physical map (channel) to the shares of non-authorized players.
- Distance Measures: The diamond norm ($D_\diamond$) and Uhlmann fidelity ($F_\diamond$) are used to quantify the closeness of a reconstructed state to the original secret.
- Beny-Oreshkov Duality: This mathematical pillar relates a channel (accessible to authorized players) to its "complementary channel" (accessible to the environment or adversary).
- Modified Entanglement-Assisted Capacity ($\tilde{C}$): A specific capacity metric based on sandwiched Rényi relative entropy of order 1/2 is introduced to represent the "strength" of the adversarial model.
Key Insights: The Secrecy-Reconstructability Duality
A central theme of the work is the "Quantum Mantra": quantum information cannot be learned without disturbing it. If an adversary gains information about the secret, they necessarily leave a "mark" that prevents perfect reconstruction by authorized players.
The paper’s most significant insight is the formalization of this duality. It shows that $\epsilon$-secrecy (the degree to which an adversary is denied information) and $\epsilon$-reconstructability (the fidelity of the recovered state) are mathematically linked. Specifically, the entanglement-assisted capacity of the adversarial channel serves as a universal bound, dictating the fundamental error floor for legitimate players.
Main Results
The authors establish several key theoretical results:
- Theorem 1: For any $(t, n)$ QSS scheme, a specific level of secrecy is mathematically equivalent to a corresponding level of reconstructability in terms of fidelity.
- Quantitative Bounds: A concrete bound for recovery error ($\delta$) in terms of diamond distance is established: $\delta \leq \sqrt{1 - \exp(-C)}$, where $C$ is the capacity of the adversarial model.
- Fidelity-Distance Relations: Refined inequalities are provided to ensure the theory holds across different metrics of state "closeness," such as fidelity and diamond norm.
Limitations and Future Work
While this work provides a robust general theory, several avenues for future research remain. The current focus is on entanglement-assisted capacity; exploring how other capacities (e.g., private or unassisted quantum capacity) influence approximate QSS is an open question. Additionally, applying these theoretical bounds to specific quantum error-correcting codes and modeling hardware-specific noise (e.g., in superconducting qubits) would bridge the gap toward practical engineering benchmarks.
Summary updated