This is the longer (partially unpublished) version of response; the shorter version (http://arxiv.org/abs/physics/0605013) is published in Physics Letters A. We point out that the claims in the comment-paper of Scheuer and Yariv are either irrelevant or incorrect. We first clarify what the security of a physically secure layer means. The idealized Kirchoff-loop-Johnson-like-noise (KLJN) scheme is totally secure therefore it is more secure than idealized quantum communication schemes which can never be totally secure because of the inherent noise processes in those communication schemes and the statistical nature of eavesdropper detection based on error statistics. On the other hand, with sufficient resources, a practical/non-ideal realization of the KLJN cipher can arbitrarily approach the idealized limit and outperform even the idealized quantum communicator schemes because the non-ideality-effects are determined and controlled by the design. The cable resistance issue analyzed by Scheuer and Yariv is a good example for that because the eavesdropper has insufficient time window to build a sufficient statistics and the actual information leak can be designed. We show that Scheuer's and Yariv's numerical result of 1% voltage drop supports higher security than that of quantum communicators. Moreover, choosing thicker or shorter wires can arbitrarily reduce this voltage drop further; the same conclusion holds even according to the equations of Scheuer and Yariv.
This is the longer (partially unpublished) version of response; the shorter version (http://arxiv.org/abs/physics/0605013) is published in Physics Letters A. We point out that the claims in the comment-paper of Scheuer and Yariv are either irrelevant or incorrect. We first clarify what the security of a physically secure layer means. The idealized Kirchoff-loop-Johnson-like-noise (KLJN) scheme is totally secure therefore it is more secure than idealized quantum communication schemes which can never be totally secure because of the inherent noise processes in those communication schemes and the statistical nature of eavesdropper detection based on error statistics. On the other hand, with sufficient resources, a practical/non-ideal realization of the KLJN cipher can arbitrarily approach the idealized limit and outperform even the idealized quantum communicator schemes because the non-ideality-effects are determined and controlled by the design. The cable resistance issue analyzed by Scheuer and Yariv is a good example for that because the eavesdropper has insufficient time window to build a sufficient statistics and the actual information leak can be designed. We show that Scheuer's and Yariv's numerical result of 1% voltage drop supports higher security than that of quantum communicators. Moreover, choosing thicker or shorter wires can arbitrarily reduce this voltage drop further; the same conclusion holds even according to the equations of Scheuer and Yariv.
In a forthcoming paper in IEE Proceedings Information Security, Feng Hao claims that temperature inaccuracies make the key exchange scheme based on Kirchhoff Loop with Johnson-like Noise insecure. First we point out that this claim is irrelevant for the security of the idealized/mathematical communicator because it violates basic assumptions. Furthermore, in general practical applications, as it has been pointed out in the very first paper, the use of thermal noise is disadvantageous therefore the issue of temperature and its accuracy is unimportant even for the security of common practical realizations. It is important to emphasize that any deviation from the idealized scheme can lead to a finite level of security. Thus, if the above-mentioned author had stressed the inaccuracy of the resistor values; his claim would have been practically valid. However the impact of such systematic errors can be kept under control. Here we cite our relevant analysis (Phys. Lett. A359, (2006) 741-744) of 1% voltage inaccuracy along the line due to wire resistance effects, which can be considered as a general result for inaccuracies of this order. The necessity to build a statistics to extract information during the properly chosen clock time leads to extremely poor eavesdropper statistics and this situation provides a good practical raw bit security that can be orders of magnitude beyond the raw bit security of idealized quantum communicators.
Abstract Recently, Gunn, Allison and Abbott (GAA) [http://arxiv.org/pdf/1402.2709v2.pdf] proposed a new scheme to utilize electromagnetic waves for eavesdropping on the Kirchhoff-law-Johnson-noise (KLJN) secure key distribution. We proved in a former paper [Fluct. Noise Lett. 13 (2014) 1450016] that GAA’s mathematical model is unphysical. Here we analyze GAA’s cracking scheme and show that, in the case of a loss-free cable, it provides less eavesdropping information than in the earlier (Bergou)-Scheuer-Yariv mean-square-based attack [Kish LB, Scheuer J, Phys. Lett. A 374:2140-2142 (2010)], while it offers no information in the case of a lossy cable. We also investigate GAA’s claim to be experimentally capable of distinguishing—using statistics over a few correlation times only—the distributions of two Gaussian noises with a relative variance difference of less than 10-8. Normally such distinctions would require hundreds of millions of correlations times to be observable. We identify several potential experimental artifacts as results of poor KLJN design, which can lead to GAA’s assertions: deterministic currents due to spurious harmonic components caused by ground loops, DC offset, aliasing, non-Gaussian features including non-linearities and other non-idealities in generators, and the timederivative nature of GAA’s scheme which tends to enhance all of these artifacts.
It is shown that the original Kirchhoff-loop-Johnson(-like)-noise (KLJN) cipher is naturally protected against the man-in-the-middle (MITM) attack, if the eavesdropper is using resistors and noise voltage generators just like the sender and the receiver. The eavesdropper can extract zero bit of information before she is discovered. However, when the eavesdropper is using noise current generators, though the cipher is protected, the eavesdropper may still be able to extract one bit of information while she is discovered. For enhanced security, we expand the KLJN cipher with the comparison of the instantaneous voltages via the public channel. In this way, the sender and receiver has a full control over the security of measurable physical quantities in the Kirchhoff-loop. We show that when the sender and receiver compare not only their instantaneous current data but also their instantaneous voltage data then the zero-bit security holds even for the noise current generator case. We show that the original KLJN scheme is also zero-bit protected against that type of MITM attack when the eavesdropper uses voltage noise generators, only. In conclusion, within the idealized model scheme, the man-in-the-middle-attack does not provide any advantage compared to the regular attack considered earlier. The remaining possibility is the attack by a short, large current pulse, which described in the original paper as the only efficient type of regular attacks, and that yields the one bit security. In conclusion, the KLJN cipher is superior to known quantum communication schemes in every respect, including speed, robustness, maintenance need, price and its natural immunity against the man-in-the-middle attack.
Abstract There is an ongoing debate about the fundamental security of existing quantum key exchange schemes. This debate indicates not only that there is a problem with security but also that the meanings of perfect, imperfect, conditional and unconditional (information theoretic) security in physically secure key exchange schemes are often misunderstood. It has been shown recently that the use of two pairs of resistors with enhanced Johnsonnoise and a Kirchhoff-loop ‒ i.e., a Kirchhoff-Law-Johnson-Noise (KLJN) protocol ‒ for secure key distribution leads to information theoretic security levels superior to those of today’s quantum key distribution. This issue is becoming particularly timely because of the recent full cracks of practical quantum communicators, as shown in numerous peer-reviewed publications. The KLJN system is briefly surveyed here with discussions about the essential questions such as (i) perfect and imperfect security characteristics of the key distribution, and (ii) how these two types of securities can be unconditional (or information theoretical).
A secure key distribution (exchange) scheme is unconditionally secure if it is unbreakable against arbitrary technological improvements of computing power and/or any development of new algorithms. There are only two families of experimentally realized and tested unconditionally secure key distribution technologies: quantum key distribution (QKD), the base of quantum cryptography, which utilizes quantum physical photonic features, and the Kirchhoff-Law–Johnson-Noise (KLJN) system that is based on classical statistical physics (fluctuation–dissipation theorem). The focus topic of this paper is the thermodynamical situation of the KLJN system. In all the original works, the proposed KLJN schemes required thermal equilibrium between the devices of the communicating parties to achieve perfect security. However, Vadai et al., in (Nature) Sci. Rep. 5, 13653 (2015) show a modified scheme, where there is a non-zero thermal noise energy flow between the parties, yet the system seems to resist all the known attack types. We introduce an attack type against their system. The attack utilizes coincidence events between the line current and voltages. We show that there is a non-zero information leak toward the Eavesdropper, even under idealized conditions. As soon as the thermal equilibrium is restored, the system becomes perfectly secure again. In conclusion, perfect unconditional security requires thermal equilibrium.