No-Switching Quantum Key Distribution Using Broadband Modulated Coherent Light

Abstract

We realize an end-to-end no-switching quantum key distribution protocol using continuous-wave coherent light. We encode weak broadband Gaussian modulations onto the amplitude and phase quadratures of light beams. Our no-switching protocol achieves high secret key rate via a post-selection protocol that utilizes both quadrature information simultaneously. We establish a secret key rate of 25 Mbits/s for a lossless channel and 1 kbit/s for 90% channel loss, per 17 MHz of detected bandwidth, assuming individual Gaussian eavesdropping attacks. Since our scheme is truly broadband, it can potentially deliver orders of magnitude higher key rates by extending the encoding bandwidth with higher-end telecommunication technology.Quantum key distribution (QKD) [1] is a technique for generating a shared cryptographic key between two parties, Alice and Bob, where the security of the shared key is guaranteed by the laws of quantum mechanics. QKD based on continuous variables (CV) [2], in particular coherent state QKD [4,5,6,7,8,9], promises significantly higher secret key rates in comparison to single photon schemes [1,3]. They are relatively simple to implement, in contrast to QKD protocols employing "non-classical" states [10]. Coherent states are readily produced by a stabilized laser and can be detected using high quantum efficiency detectors. Confidence in the practicability of coherent state QKD protocols has increased since it was shown that the security of these protocols can be ensured for channel losses greater than 50% using post-selection [6] or reverse reconciliation [7] procedures. In principle, it is therefore possible to generate a secure key even in the presence of arbitrarily high loss. This development, coupled with potentially high secret key rates, render coherent state QKD protocols viable contenders for real-world cryptographic applications.Our coherent state QKD protocol builds on previous protocols presented in [5,6,7] and is an advance on random switching by simultaneously measuring both measurement bases [8]. The QKD protocol operates as follows. Alice draws two random numbers x A and p A from two Gaussian probability distributions with zero mean and variances of V (x A ) and V (p A ) respectively. Alice prepares a coherent state |x A + ip A and sends it to Bob. As a result of losses in the quantum channel, vacuum noise is coupled into the transmitted state. On receiving the state, Bob simultaneously measures both the amplitude (x B ) and phase (p B ) quadratures of the state via a 50/50 beam splitter. At this stage, Alice and Bob share correlated random data from which they can generate a secret key. They use postselection [6] to reverse any initial "information advantage" a potential eavesdropper (Eve) might have obtained, and perform information reconciliation and privacy amplification to distill a final secret key. Although no-switching coherent state QKD protocols have been demonstrated to be secure against coherent (collective) attacks [11] and progress has been made towards proving the uncondi...

Witnessing effective entanglement in a continuous variable prepare-and-measure setup and application to a quantum key distribution scheme using postselection

“…To determine the quadratures X and Y, Bob has to use a phase reference as local oscillator for the homodyne measurements [6,30]. This phase reference is sent along with the signal state in our experiment, as it is done in most continuous variable quantum cryptography experiments [31,32,33,34,35,36]. The analysis of quantum correlations does not directly apply to this new situation.…”

Experimental demonstration of post-selection-based continuous-variable quantum key distribution in the presence of Gaussian noise

“…Alice and Bob then both announce the absolute values of their encodings |S A | and measurement results |m B | respectively. This is in contrast to previous zero excess noise protocols where only Alice announces her absolute value [3,7]. Alice and Bob then post-select information for which they have a mutual information advantage over Eve and discard information for which they do not.…”

“…These protocols are based on the random Gaussian modulation of coherent states using either homodyne [4] or heterodyne [5] detection and both have been experimentally demonstrated [6,7,8,9,10]. At present PS-based CV-QKD has practical advantages in terms of key distillation and has been demonstrated experimentally for up to 90% channel loss [7].…”

“…For PS CV-QKD, the addition of excess noise into the analysis is quite difficult. The original protocol [3] only considered pure or vacuum states in its scheme and so far all post-selection protocols since have concentrated on the unrealistic case of zero excess noise [7,16,17]. Recently however, excess noise using a hybrid protocol, consisting of both post-selection and either direct or reverse reconciliation, was considered for the case of collective attacks [18].…”

“…These protocols are based on the random Gaussian modulation of coherent states using either homodyne [4] or heterodyne [5] detection and both have been experimentally demonstrated [6,7,8,9,10]. At present PS-based CV-QKD has practical advantages in terms of key distillation and has been demonstrated experimentally for up to 90% channel loss [7].Reverse reconciliation CV-QKD, due to its inherent nature, easily incorporates excess noise into the protocols, and security proof have been demonstrated in the case of individual Gaussian attacks [4,5], non-Gaussian attacks [11], collective attacks [12,13] (with their Gaussian optimality [14]) and coherent states using homodyne detection [15]. For PS CV-QKD, the addition of excess noise into the analysis is quite difficult.…”