[Paper Review] Secure Reusable Base-String in Quantum Key Distribution

Kai Wen, Fu‐Guo Deng+1 more|ArXiv.org|Jun 26, 2007

Quantum Information and Cryptography3 citations

TL;DR

This paper proposes a novel quantum key distribution protocol that enables unconditional security and high efficiency by reusing a shared, secret base string through privacy amplification. By leveraging entanglement distillation and error rate estimation, the protocol achieves a reusable base string generation rate up to 25% bit error rate, significantly higher than the 11% threshold for key generation, enabling long-term key expansion from minimal initial resources.

ABSTRACT

Protecting secure random key from eavesdropping in quantum key distribution protocols has been well developed. In this letter, we further study how to detect and eliminate eavesdropping on the random base string in such protocols. The correlation between the base string and the key enables Alice and Bob to use specific privacy amplification to distill and reuse the previously shared base string with unconditional security and high efficiency. The analysis of the unconditional secure reusable base string brings about new concept and protocol design technique.

Motivation & Objective

• To address the security of shared base strings in BB84-type QKD protocols, which are often treated as static keys but vulnerable to eavesdropping.
• To develop a method for reusing a previously shared base string across multiple QKD rounds without compromising unconditional security.
• To extend privacy amplification techniques to the base string itself, enabling distillation of high-fidelity, reusable base pairs from noisy channels.
• To establish a new protocol framework that enhances efficiency by eliminating basis choice overhead and enabling iterative key generation.
• To demonstrate that base string reuse is not only possible but more robust than key generation, with a higher error tolerance threshold.

Proposed method

• The protocol uses 2n EPR pairs for base strings and 2n for communication, grouped into blocks of two entangled pairs per block.
• Alice applies a controlled-Hadamard (CH) gate using her base pair qubit as control and the communication pair qubit as target, entangling the two.
• Bob applies a similar CH gate on his side after receiving the qubits, enabling correlation between base and communication pairs.
• A random permutation of blocks is used to ensure exchangeability, enabling application of the quantum de Finetti representation for security analysis.
• Error rates on communication and base pairs are estimated via check blocks; entanglement purification protocols (EPP) are applied to distill high-fidelity pairs.
• CSS codes are used for error correction and privacy amplification on the key, while classical linear codes are used for base string distillation, simplifying the process.

Experimental results

Research questions

• RQ1 Can a shared, secret base string in QKD be securely reused across multiple rounds without compromising unconditional security?
• RQ2 What is the maximum channel error rate that still allows for secure and efficient reuse of the base string?
• RQ3 How can privacy amplification be adapted to distill reusable base strings from noisy quantum channels?
• RQ4 What is the relationship between the error rate on communication qubits and the phase error rate on base pairs, and how can it be exploited for security?
• RQ5 Can the base string be reused more efficiently than the final key, and if so, by how much?

Key findings

• The maximal error rate allowing non-zero generation rate of reusable base string is 25%, significantly higher than the 11% threshold for key generation.
• The base string distillation rate is given by $ R_b(e) = 1 - H(2e) $, which remains positive for $ e < 0.25 $, enabling long-term reuse.
• The key generation rate is $ R_k(e) = 1 - 2H(e) $, with a maximum tolerance of $ e \< 0.11 $, making base string reuse more robust.
• A single initial $ 2n $-bit base string can generate a total key length of $ L_k = \frac{nR_k(e)}{1 - R_b(e)} $, which grows without bound as long as $ R_b(e) < 1 $.
• The reuse of base strings enables iterative key generation, where the same base string is used repeatedly to generate new keys, drastically improving efficiency.
• Base string distillation uses classical linear codes instead of CSS codes, simplifying the protocol and reducing computational overhead.