5 hours ago by Stuart Mason Dambrot Experimental setup of the plug-and-play Clavis2 system. This type of interferometric setup does not necessitate continuous polarization control and alignment, and therefore guarantees excellent system stability for quantum communications. Courtesy: Anna Pappa, LTCI, CNRSTlcom ParisTech <\/p>\n
(Phys.org) Cryptography the practice and study of techniques for secure communication in the presence of third parties, referred to as adversaries has a long and varied history. In ancient Greece, for example, the Spartan military may have used the so-called scytale transposition cipher to encrypt and decrypt messages. Steganography (hiding the existence of a message) was also first developed at that time as, according to Herodotus, a message tattooed on a slave's shaved head and then hidden under regrown hair and is still in use in the form of invisible ink, microdots, and digital watermarks. That said, applying complexity cryptography to quantum communication is and will continue to be essential and while quantum cryptographic primitives are in principle more secure than classical protocols, demonstrating this in a practical system has proven difficult. <\/p>\n
Recently, however, scientists at Laboratory for Communication and Processing of Information (LTCI), Paris a joint research lab between Centre National de la Recherche Scientifique (CNRS) and Tlcom ParisTech have experimentally implemented a quantum coin flipping protocol that performs better than any classical system over a distance suitable for deployment in metropolitan area optical networks. Based on an enhanced commercial quantum key distribution (QKD) device, the approach is nearly perfectly secure against bounded adversaries a feature the researchers state make it a practical toolbox for designing secure quantum communications systems. <\/p>\n
Researcher Anna Pappa discussed the paper she and her co-authors published in Nature Communications with Phys.org beginning with the challenge of addressing the historical difficulty of demonstrating the known information-theoretic security advantages of quantum versions of coin flipping and other primitives (basic cryptographic algorithms used to construct more complicated cryptographic tools) relative to classical protocols in a practical communication scenario. \"Quantum cryptography is a relatively new field that emerged after Bennett and Brassard's groundbreaking paper in19841, which introduced the idea of using quantum mechanics to enhance classical cryptographic protocols like key distribution and coin flipping,\" Pappa tells Phys.org, adding that the main difference between classical and quantum computing is that in the latter, information is contained in the physical properties of the exchanged messages. <\/p>\n
\"This provides a strong advantage but also hinders straightforward applications of quantum protocols,\" Pappa point out. \"Historically, many protocols that were in theory secure were completely broken in practice, because of limitations in current technology. For example,\" she illustrates, \"previous coin flipping protocols necessitated a single-photon source or an entangled source in order to be secure but the first is not currently available, while the second cannot be easily deployed for long-distance communications since entanglement is very fragile, and cannot be maintained for long periods of time due to quantum memory limitations. In our research, while we are exploiting the effects of superposition in quantum mechanics, we do not use entangled states and this is what makes our implementation easily implementable with standard photonic sources.\" <\/p>\n
Another important factor, she notes, is that coin flipping is a protocol used when participants do not trust each other, which makes correcting transmission errors more difficult. At the same time, trusted setting protocols like quantum key distribution (QKD) have in recent years achieved security for distances of more than 100 kilometers. This is due to the fact that measuring a quantum system disturbs that system, and any third party trying to gain knowledge of the key can therefore be detected by the two communicating users. <\/p>\n
The researchers also faced the challenge of experimentally implementing a quantum coin flipping protocol that performs strictly better than classically possible over a distance suitable for communication over metropolitan area optical networks. \"The Clavis2 platform that we used was developed by IdQuantique, a company based in Geneva, Switzerland that works closely with researchers worldwide in order to test and verify their systems,\" Pappa recalls. \"There were many challenges that we faced during the implementation of our coin flipping protocol using a commercial plug-and-play system originally designed to perform key distribution between two parties (commonly referred to as Alice and Bob) who trust each other and want to establish a common secret key. <\/p>\n
\"In quantum coin flipping,\" Pappa explains, \"the players do not trust each other, since both want to win the coin flip, so they try to cheat by numerous ways for example, by increasing the average photons contained in the pulses, or by declaring that they lost some message when they do not like the result of the protocol. Furthermore, they could try to exploit the physical properties of the system, like an asymmetry in the creation of the different quantum states used, or in the detection of the different states. We therefore needed to account for all imperfections of the system and come up with detailed security proofs in order to show the quantum advantage of our implementation.\" <\/p>\n
Relatedly, the scientists sought to provide combined quantum coin flipping protocols that were almost perfectly secure against bounded adversaries. \"We wanted to find a way to provide security against an adversary of unknown abilities, so we used two schemes that are secure against adversaries of limited power that is, noisy storage and computationally bounded and enhanced them with our protocol.\" To do this, they we analyzed the bounded protocols and found the exact step where an unbounded adversary would be able to perfectly cheat, and then strengthened that step using our protocol.\" <\/p>\n
\"A problem that we faced,\" Pappa tells Phys.org, \"was that, since the players do not trust each other, they cannot perform error-correction and other procedures that necessitate collaboration between the parties, therefore limiting the tolerance to errors. We therefore needed to make some changes to the system in order to observe a quantum advantage for a considerable channel length. For example,\" Pappa explains, \"the detectors on Bob's side had to be replaced because they had low detection efficiencies and high dark counts, and we could not observe any quantum advantage for any channel length. By substituting them with better quality detectors, we managed to experimentally demonstrate a quantum advantage for a channel distance of 15km. In addition,\" she continues, \"photon source attenuation was very high. This meant that essential Clavis2 procedures could not be executed, requiring significant reprogramming.\" Finally, Pappa notes, in order for the players to decide on the protocol parameters, the scientists had to perform detailed careful system analyses to identify system component losses and errors in order to, for example, estimate how many times they need to run the protocol and how much attenuation needs to be applied. <\/p>\n
<\/p>\n