Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A security framework for quantum key distribution with imperfect sources

Guillermo Currás-Lorenzo, Margarida Pereira, Go Kato, Marcos Curty, Kiyoshi Tamaki

TL;DR

The paper addresses the security gap in QKD when bit-and-basis encoders are imperfect, which introduces side channels and inter-pulse correlations that invalidate standard proofs. It presents a finite-key security proof against coherent attacks that tolerates general encoding imperfections with only a bound on side-channel leakage $\epsilon$, by unifying loss-tolerant and quantum-coin methods and using target/reference states. A key result is a tight bound on the phase-error rate $e_{\rm ph}$ that remains robust under high loss and arbitrary qubit-flaw strength $δ$, provided $\epsilon$ bounds leakage; this enables high-rate BB84, three-state, and MDI-QKD in finite-key regimes. Overall, the framework relaxes the need for full device characterization while maintaining unconditional security, with practical implications for decoy-state and MDI-QKD implementations, and clear directions for tightening $\epsilon$ bounds in real devices.

Abstract

Imperfect bit-and-basis encoders compromise the security of quantum key distribution (QKD) systems via modulation flaws, side channels and inter-pulse correlations, which invalidate standard security proofs. Existing results addressing such imperfections suffer from critical limitations: they either consider only specific flaws, offer an unreasonably poor performance, or require the protocol to be run very slowly. Here, we present a finite-key security proof approach against coherent attacks that incorporates general bit-and-basis encoding imperfections (including modulation flaws, side channels and inter-pulse correlations) while achieving significantly better performances than previous approaches and requiring only partial characterization.

A security framework for quantum key distribution with imperfect sources

TL;DR

The paper addresses the security gap in QKD when bit-and-basis encoders are imperfect, which introduces side channels and inter-pulse correlations that invalidate standard proofs. It presents a finite-key security proof against coherent attacks that tolerates general encoding imperfections with only a bound on side-channel leakage , by unifying loss-tolerant and quantum-coin methods and using target/reference states. A key result is a tight bound on the phase-error rate that remains robust under high loss and arbitrary qubit-flaw strength , provided bounds leakage; this enables high-rate BB84, three-state, and MDI-QKD in finite-key regimes. Overall, the framework relaxes the need for full device characterization while maintaining unconditional security, with practical implications for decoy-state and MDI-QKD implementations, and clear directions for tightening bounds in real devices.

Abstract

Imperfect bit-and-basis encoders compromise the security of quantum key distribution (QKD) systems via modulation flaws, side channels and inter-pulse correlations, which invalidate standard security proofs. Existing results addressing such imperfections suffer from critical limitations: they either consider only specific flaws, offer an unreasonably poor performance, or require the protocol to be run very slowly. Here, we present a finite-key security proof approach against coherent attacks that incorporates general bit-and-basis encoding imperfections (including modulation flaws, side channels and inter-pulse correlations) while achieving significantly better performances than previous approaches and requiring only partial characterization.
Paper Structure (16 sections, 21 equations, 3 figures)

This paper contains 16 sections, 21 equations, 3 figures.

Table of Contents

  1. Introduction
  2. Results
  3. Partial state characterization assumption
  4. Security analysis
  5. Discussion
  6. Conclusion
  7. METHODS
  8. Mixed and non-identically-distributed states
  9. Determining the value of $\epsilon$
  10. Pulse correlations
  11. A more general form of Eq. (1)
  12. Characterizing qubit flaws
  13. Protection against detector imperfections side channels
  14. Weak coherent sources
  15. Choice of target and reference states
  16. ...and 1 more sections

Figures (3)

  • Figure 1: Asymptotic secret-key rate obtainable using our proof as a function of the distance (km) for the BB84 (solid lines) and three-state (dashed-dotted lines) protocols. We assume $\delta=0.063$honjoDifferentialphaseshiftQuantum2004xuExperimentalQuantum2015 and consider several values of $\epsilon$.
  • Figure 2: Asymptotic secret-key rate obtainable using our proof as a function of the distance (km) for the BB84 protocol, compared with that of the original quantum coin (GLLP) gottesmanSecurityQuantum2004loSecurityQuantum2007koashiSimpleSecurity2009 and reference technique (RT) pereiraQuantumKey2020pereiraModifiedBB842023 analyses. We consider the values $\epsilon = 10^{-6}$ and $\delta \in \{0,0.063\}$.
  • Figure 3: Finite-size secret-key rate against general attacks obtainable using our proof as a function of the total number of emitted pulses $N$. We consider the BB84 protocol, $\epsilon = 10^{-6}$, $\delta=0.063$, and we set the correctness and secrecy parameters of the final key to $\epsilon_{\rm corr} = \epsilon_{\rm secr} = 10^{-10}$.