[Paper Review] Exploiting the randomness of the measurement basis in quantum cryptography: Secure Quantum Key Growing without Privacy Amplification

Hannes R. Böhm, Paul S. Böhm+3 more|arXiv (Cornell University)|Aug 30, 2004

Quantum Computing Algorithms and Architecture3 citations

TL;DR

This paper proposes a novel quantum key distribution protocol that enhances the BB84 scheme by using the inherent randomness of measurement basis choices to encrypt classical communication during key sifting, eliminating the need for privacy amplification. By preventing an eavesdropper from reconstructing the sifting process, the protocol reduces the eavesdropper's mutual information on the final key, enabling secure key growing even at high quantum bit error rates.

ABSTRACT

We suggest that the randomness of the choices of measurement basis by Alice and Bob provides an additional important resource for quantum cryptography. As a specific application, we present a novel protocol for quantum key distribution (QKD) which enhances the BB84 scheme by encrypting the information sent over the classical channel during key sifting. We show that, in the limit of long keys, this process prevents an eavesdropper from reproducing the sifting process carried out by the legitimate users. The inability of the eavesdropper to sift the information gathered by tapping the quantum channel reduces the amount of information that an eavesdropper can gain on the sifted key. We further show that the protocol proposed is self sustaining, and thus allows the growing of a secret key.

Motivation & Objective

• To exploit the randomness of measurement basis choices in quantum cryptography as a resource beyond basis selection.
• To address the security vulnerability in BB84 where public basis announcements leak information to eavesdroppers.
• To develop a protocol that allows secure quantum key growing without privacy amplification by encrypting basis information during sifting.
• To reduce the mutual information an eavesdropper can gain on the final key by making the sifting process unreplicable without the shared secret.
• To explore the feasibility of operating at higher quantum bit error rates than traditional BB84 by avoiding the key length reduction caused by privacy amplification.

Proposed method

• The protocol extends BB84 by using a shared secret key to encrypt basis information during the sifting phase via XOR operations.
• Alice and Bob each apply an XOR between their local basis choices and a shared secret segment before transmitting basis information over the classical channel.
• The encrypted basis information prevents an eavesdropper from determining which raw key bits contributed to the final sifted key.
• The protocol relies on the fact that without the shared secret, an eavesdropper cannot reproduce the sifting process, even with full knowledge of the raw key.
• The mutual information between the eavesdropper and the sifted key is shown to decrease with increasing raw key length, due to the binomial distribution of contributions from raw key bits to each sifted bit.
• The protocol assumes an authenticated classical channel for security, though it is argued that authentication may not be strictly necessary due to the randomization of basis data.

Experimental results

Research questions

• RQ1 Can the randomness of measurement basis choices in BB84 be exploited as a cryptographic resource beyond basis selection?
• RQ2 Does encrypting basis information during sifting reduce the eavesdropper’s ability to reconstruct the sifted key, even with full access to the raw key?
• RQ3 Can a QKD protocol achieve secure key growing without privacy amplification by leveraging basis randomness?
• RQ4 How does the mutual information between an eavesdropper and the final key scale with the length of the raw key in this protocol?
• RQ5 Is the protocol viable at higher quantum bit error rates compared to standard BB84, particularly when privacy amplification significantly reduces key rates?

Key findings

• The protocol prevents an eavesdropper from reproducing the sifting process because the basis information is encrypted with a shared secret, rendering it unguessable without the key.
• The mutual information between the eavesdropper and the final key can be made arbitrarily small by increasing the length of the raw key, due to the spreading of contributions from raw key bits across multiple sifted key bits.
• The probability that a specific raw key bit contributes to a given sifted key bit follows a binomial distribution, approximated by a Gaussian centered at i = 2l with standard deviation σ = √i / 2.
• For a large number of sifted key bits, the number of raw key bits contributing significantly to a single sifted bit increases, reducing the eavesdropper’s ability to infer the correct bit values.
• The protocol is self-sustaining and enables secure key growing without privacy amplification, suggesting potential advantages in high-error-rate scenarios.
• While a complete security proof is not provided, the protocol shows strong indications of resilience against intercept-and-resend attacks, as the eavesdropper cannot reliably determine which basis choices led to the final key.