Source-linked AI summary

Revisiting the security of quantum dialogue and bidirectional quantum secure direct communication

Fei Gao, Fen-Zhuo Guo, Qiao-Yan Wen, Fu-Chen Zhu

arXiv:0801.2420v2quant-ph

TL;DR

The paper addresses whether public announcements in bidirectional QSDC and quantum dialogue protocols can leak secret information. It analyzes these protocols from information theory and cryptography, concluding that their transmitted information can be partly leaked and that the resulting insecurity is equivalent to reusing a one-time-pad key.

  • Problem

    The paper examines information leakage from public announcements in bidirectional QSDC protocols, a security issue that can prevent the intended unconditional security.

  • Method

    The paper analyzes representative bidirectional QSDC protocols and compares their information leakage with the behavior of one-time-pad encryption using a reused key.

  • Results

    The analyzed protocols partly leak transmitted secret information through public announcements, with the paper relating this insecurity to one-time-pad encryption with a reused key.

  • Takeaways & Limitations

    Public announcements in bidirectional QSDC protocols must be assessed for information leakage because partial secrecy does not establish secure communication.

Abstract

from arXiv · show

From the perspective of information theory and cryptography, we analyze the security of two quantum dialogue protocols and a bidirectional quantum secure direct communication (QSDC) protocol, and point out that the transmitted information would be partly leaked out in them. That is, any eavesdropper can elicit some information about the secrets from the public annunciations of the legal users. This phenomenon should have been strictly forbidden in a quantum secure communication. In fact, this problem exists in quite a few recent proposals and, therefore, it deserves more research attention in the following related study.

2 School of Science, Beijing University of Posts and Telecommunications, Beijing 100876, China

The paper analyzes two quantum dialogue protocols and a bidirectional QSDC protocol, finding that public announcements can leak part of the transmitted secrets to eavesdroppers.

  • The study analyzes the security of two quantum dialogue protocols and one bidirectional QSDC protocol from information-theoretic and cryptographic perspectives.
  • Any eavesdropper can elicit some information about the users’ secrets from their public announcements.
  • Such leakage is forbidden in quantum secure communication because transmitted secret information should remain protected.
  • The paper reports that this problem occurs in quite a few recent proposals and merits further research.

1 Introduction

The introduction motivates quantum cryptography and QSDC as routes toward stronger security, then identifies information leakage through public communication as a distinct insecurity in bidirectional QSDC protocols.

  • QSDC directly transmits a secret message rather than a random key, while bidirectional QSDC allows secret messages to flow in both directions.
  • Quantum cryptography seeks higher, theoretically unconditional security because increasing computational ability challenges many classical cryptosystems.
  • The paper studies information leakage through the public classical communication required by quantum cryptography.
  • The authors report that transmitted secret information is partly leaked in several bidirectional QSDC protocols.
  • Sections 2–4 analyze three typical bidirectional QSDC protocols and demonstrate information leakage in detail.

2 Analysis of NBA and MZL protocols

The NBA and MZL analysis shows that public Bell-state information narrows the possible coding-operation pairs, leaking part of the four transmitted secret bits without active eavesdropping.

  • The leakage occurs without an active attack because Eve can infer information from the public announcement alone.
  • In the NBA protocol, Bob and Alice encode two secret bits each on an EPR-pair carrier using one of four operations, followed by a public Bell measurement result.
  • The public initial state and measurement result restrict Alice’s and Bob’s coding operations to four possible pairs.
  • 2 bits of information about the 4 transmitted secret bits are leaked to Eve.
  • The paper rejects treating NBA as secure merely because Eve cannot determine the exact value of every transmitted bit.

3 Analysis of JZ protocol

The JZ single-photon protocol lets Alice and Bob exchange one secret bit each, but its public initial-state and measurement announcements still reveal part of those two bits.

  • In JZ, Bob and Alice each encode one secret bit on a single photon using the operations I and iσ_y.
  • The protocol uses the photon’s preparation basis for Alice’s measurement, after which she publicly announces the measurement result.
  • With one photon, the users communicate 2 secret bits in total, one for Alice and one for Bob.
  • Given the public initial state and measurement result, Eve narrows the coding operations to two possibilities containing only 1 bit of uncertainty.
  • Only 1 bit of the 2 transmitted secret bits is communicated securely, while the other bit is leaked unknowingly.

4 Analysis of MXN protocol

The MXN protocol uses GHZ triplets and public Bell-measurement announcements to exchange secret bits, but those announcements reveal substantial information about the encoded secrets.

  • Protocol setup: The three-party MXN protocol uses two shared GHZ triplets, with Alice controlling qubits 1 and 4, Bob qubits 2 and 5, and Charlie qubits 3 and 6.The users perform Bell measurements on pairs (1,4), (2,5), and (3,6), then publicly announce the results.
  • Encoding: Alice encodes 2 secret bits, while Bob and Charlie each encode 1 bit, producing eight possible GHZ states after coding.The encoded state depends on the users’ unitary operations and their corresponding bit values.
  • Information leakage: Public measurement results allow any observer to deduce the coded GHZ state and thereby obtain the other users’ secret bits.The protocol’s public announcements expose the state information needed to infer the encoded operations.
  • Information leakage: In the illustrated case, Eve narrows the four transmitted bits to two possible assignments and therefore learns 1 bit of information.The two assignments are [00,0,1]ABC and [11,1,0]ABC under the stated example.
  • Security result: Only 1 of the 4 secret bits is transmitted securely, while 3 bits are leaked, corresponding to 75 percent information leakage.The same analysis is extended to the N-party version, where N bits are leaked from N+1 transmitted bits.

5 Discussion and conclusion

The discussion identifies partial secret leakage as a recurring security problem in bidirectional QSDC and relates it to reuse of one-time-pad keys. It recommends matching transmitted information to channel security capacity and giving the issue further attention.

  • Security interpretation: The leakage effect is equivalent to using a one-time-pad key two or more times, so the analyzed bidirectional QSDC protocols are not fully secure.The paper connects the public-announcement leakage to the standard prohibition against OTP key reuse.
  • JZ protocol: In the simplified JZ analysis, public ciphertexts reveal only 1 bit of information about 2 transmitted secret bits.This follows because the same key bit is used to encrypt both users’ secret bits.
  • Remedy: The paper proposes estimating each quantum channel’s secure capacity before determining how many secret bits to transmit.For JZ, it states that one qubit can securely communicate only 1 bit of secret information.
  • Remedy: Reducing transmission efficiency to a proper value is presented as a possible way to resolve the leakage problem.The conclusion calls for greater attention to information leakage in subsequent research.
  • Conclusion: The paper argues that transmitted secrets are partly leaked in some bidirectional QSDC protocols, challenging their claimed security.The authors describe this as a widespread misunderstanding in recent proposals.

3 Ekert A K. Quantum cryptography based on Bell's theorem. Phys Rev Lett, 1991, 67: 661-663

The supplied passage contains bibliographic entries for work on quantum cryptography, quantum key distribution, and secure direct communication.

  • Bibliographic entry: The supplied references also include quantum key distribution in 50-km optical fibers and experimental free-space quantum cryptography communication.These entries identify later experimental or implementation-oriented quantum-cryptography work.

10 Lucamarini M, Mancini S. Secure deterministic communication without entanglement. Phys Rev Lett, 2005, 94: 140501

The supplied passage lists references concerning secure direct communication, quantum one-time pads, multiparticle GHZ states, and related protocol discussions.

  • Bibliographic entry: The list also includes quantum secure direct communication using high-dimensional superdense coding, multiparticle GHZ states, and EPR-pair networks.These entries cover several protocol constructions and network variants.
  • Bibliographic entry: The references include secure direct communication with a quantum one-time pad and a comment-and-reply exchange concerning that protocol.The listed comment and reply are attributed to Hoffmann, Boström, Felbinger, and Deng and Long.

18 Nguyen B A. Quantum dialogue. Phys Lett A, 2004, 328: 6-10

This section lists a related-work citation for Nguyen B A.'s 2004 paper “Quantum dialogue.”

  • The surrounding reference material includes Zhang Y S, Li C F, and Guo G C's comment on “Quantum key distribution without alternative measurements.”
  • The cited comment appeared in Phys Rev A in 2001 as volume 63, article 036301.

21 Wójcik A. Comment on "Quantum dense key distribution". Phys Rev A, 2005, 71: 016301

This section assembles references on attacks, security analyses, quantum dialogue, quantum secure direct communication, entanglement, and secrecy systems. The citations span foundational proposals, comments, improvements, and information-theoretic discussions.

  • The references include comments and attack analyses concerning quantum secret sharing, quantum exams, participant attacks, and authenticated quantum direct communication.These entries include works by Gao and colleagues, Lo and Ko, and Zhang and colleagues.
  • Several references address security improvements against Trojan-horse attacks, fake-signal attacks, and other attacks on quantum cryptographic protocols.The cited works discuss multiparty quantum secret sharing and related protocol vulnerabilities.
  • The cited literature covers quantum dialogue and bidirectional or controlled quantum secure direct communication using Bell, single-photon, GHZ, and entanglement-swapping approaches.The references include Bennett and Wiesner’s communication via one- and two-particle operators, single-photon quantum dialogue, GHZ-based protocols, and entanglement swapping.
  • The bibliography also includes studies of information splitting, secrecy-system theory, one-time pads, and classical correlation in quantum dialogue.These references connect quantum communication protocols with information-theoretic and classical-correlation analyses.
Loading 0801.2420v2…