Dept. of  Electrical  and  Computer  Engineering, Texas A&M University

home personnel research news conferences journals funding collaboration


Kirchhoff's-law-Johnson-noise (KLJN) secure key distribution

Information-theoretically (unconditionally) secure key exchange scheme based on: Johnson-noise (Fluctuation-dissipation theorem); Kirchhoff's laws; and the properties of Gaussian stochastic processes.

Brief History of the KLJN system

2005, August-September: Creation of the first unconditionally secure KLJN scheme and submitting it to both PLA and the arXiv preprint database on September 18.

2005, September 30: Science magazine featured the arXiv manuscript. Janos Bergou mentions the possibility of wire resistance attack (which is however not a crack of security only a minor information leak).

2005, December 19: Defense against the man-in-the-middle-attack. The simplest defense against active (invasive) attacks utilizing the robustness of classical physical information, which is essential there.

2006, February: Due the invitation by a Division-Director at National Science Foundation (NSF), an NSF grant application was submitted. Support for 3 years was requested to experimentally demonstrate the KLJN  concept by the end of 2009. The application was returned by the Program Manager without allowing to review it by the board. The argument was: irrelevant topic. This unexpected move by NSF triggered the "Szegedin Whisper project" in Hungary, and resulted in the successful experimental demonstration there 10 months later, which was 2 years earlier than it was planned in the returned NSF grant application, see below.

2006, February: The Bergou-type wire resistance attack submitted by Jacob Scheuer and Amnon Yariv. (The corrected calculations indicating much weaker (1000 times less) signal for Eve was published in PLA by Kish and Scheuer in 2010, see below).

2006, March: Telecloning (teleportation) and networks (with Robert Mingesz's attack against the first version and defense against that).

2006, October: First concept of KLJN key exchange via a ("hot") wire that is already in use for other purposes (power, phone, internet, etc.)

2006, November: Feng Hao (that time, a PhD student in UK) launches the Hao-attack utilizing temperature differences between Alice and Bob. Even though this type of information leak has turned out to be below the measurement limit at practical applications with today's KLJN technology, the attack is important from a fundamental security point of view and it is so far the only passive attack where the author did not make an error in the mathematical evaluation of the problem.

2006, December: The Szegedien Whisper Project (our "Manhattan" project with Robert Mingesz and Zoltan Gingl) at the Noise Research Lab of the Department of Experimental Physics, University of Szeged, Hungary. The US invention for providing information theoretically (unconditionally) secure key exchange via wire connection got realized first in Hungary because NSF returned the US grant application without even reviewing it, see above. All the known attacks of that time were tested and the KLJN system was successfully defended against them.

2007, May: The KLJN scheme was presented as a plenary talk at the Fluctuations and Noise (FaN) symposium (SPIE) in Florence. At the same time, the New Scientist magazine featured the the successful experiments.

2008: Unconditional security of computers and instruments with KLJN (with Olivier Saidi). Derek Abbott (Univ. Adelaide, AU) created the Wikipedia page "Kish cypher".

- As a "fallout" of KLJN research, Noise-based Logic schemes, including their brain logic version, were created to make a classical physical, potential alternatives for quantum computers and artificial intelligence.
- Pao-Lo Liu launched unsuccessful passive attacks against KLJN. Moreover, he made also two very interesting innovations. He brought an active element, the circulator, into the picture, which resulted in several new, circulator-based KLJN schemes. He created a linear-superposition theorem based description of KLJN which allows a computer-based KLJN emulation. The Liu KLJN emulation robust against passive attacks in the steady state; we failed with all breaking attempts, even though Eve can separate the signals propagating in the two directions. However, according to Pao-Lo, his scheme is not secure during the transient regime of the key exchange.

2010: During a visit by Jacob Scheuer, we submitted to PLA the corrected version of their wire resistance attack calculations and found a 1000 times smaller leak signal.

2011: During a visit by Tamas Horvath and Jacob Scheuer, we explored the possibility of privacy amplification. The astronomically small error probability of KLJN makes privacy amplification very fashionable even without error correction protocols before that.

- With Ferdinand Peper, the very first smart grid related KLJN paper was published.
- Intensive fights break out on Wikipedia between enthusiastic (effectively) anonymous KLJN-opposers (quantum-supporters) and KLJN-supporters who rarely understood the actual security and physics aspects of KLJN. Subsequently, several books with "Kish cipher" title by various authors were published as the printouts of the Wikipedia entry. In March 2012, LBK requested the deletion of the Kish cypher entry because of its unprofessional nature and his name attached to it. Then, the request for deletion was refused. Triggered by this situation, LBK signed a contract with World Scientific to write a book with the same title: Kish cypher.

-Inspired by discussions with Horace Yuen's and by his "keyed" quantum communication scheme, new efforts were made and seven more advanced types of KLJN schemes with enhanced security and/or speed were created.
- A year later (2013 April) after the unsuccessful one by LBK, there was a new (this time anonymous) request on the Talk page to delete the Wikipedia entry. Then LBK published a public letter on the Talk page in which he strongly supported the deletion. Finally, the entry was deleted.
- The first package of Physical Unclonable Function (PUF) hardware key applications of KLJN for unconditionally secure locks and keys; with Chiman Kwan.
- In April, Charles Bennett and Jess Riedel launched a general, multi-faced attack against the KLJN scheme in a manuscript on arXiv. The reason why this attack is mentioned in the history (instead of the debate section) is that it triggered a very extensive and detailed study and response in  PLOS ONE  (with Claes Granqvist and Derek Abbott), including information theoretic security proofs of KLJN against various types of attacks. It was concluded that the information theoretic (unconditional) security of the KLJN method has not been successfully challenged. 
-Derek Abbott and Gabor Schmera (US Navy, SPAWAR Center) published a short, peer reviewed, minimalist description of the KLJN scheme in Scholarpedia about its basics to fill the vacuum after the deleted Wikipedia entry.

- Our most general proof of the unconditional security of KLJN (with Claes Granqvist) became invited paper in the journal Quantum Information Processing.
- Yessica Saez completed her works on error removal processes and the best error removal method yielded astronomically low error probabilities (<10^-12) for KLJN, which opened the possibility for intensive privacy applications without the need for error correcting algorithms. Yessica defended her PhD, which was the first PhD at Texas A&M in the topic of KLJN.
- The Gunn-Allison-Abbott "directional coupler attack" published in a Nature-journal triggered the publication of several papers showing that the attack was flawed at all levels (physics, security, circuit theory, and experiments). Yet, the attack was useful academically because its analysis clarified interesting open problems, such as no waves in cables at low frequencies and potential new information leaks due to external disturbance.
- We (with Claes Granqvist) introduced the Second Law Attack, the most efficient cable resistance attack and, in the same paper, a defense against all types of wire resistance attacks by introducing a proper temperature difference between the resistors.
Zoltan Gingl and Robert Mingesz proved that only Gaussian noises can provide perfect security in KLJN. Thus the original KLJN situation is not ony satisfactory but also unique for security.
- Xiaolin Cao's visit triggered two papers about using KLJN to provide unconditional security for Autonomous Vehicle Systems.

- Elias Gonzalez had successfully continued his creative studies (since 2012) how to secure various smart grid topologies with KLJN and how to classify the level of security in mixed networks. In due course, Robert Balog power distribution expert joined the efforts.
- Barry Chen had successfully continued his extensive studies with an industrial cable simulator and tested the current injection and cable capacitance attacks. He also demonstrated the best defense method against active attacks: a computer model of the system that is running in the background, driven by measured data, and its output compared with measurements. 
- Gergely Vadai, Robert Mingesz and Zoltan Gingl discovered a new KLJN operation, where there is a steady power flow between the parties during secure key exchange.
- Inspired by the Vadai-Mingesz-Gingl findings, the Random-Resistance-Random-Temperature KLJN scheme is introduced. So far, there is no known way to extract any information leak from this system. - New method to provide a strongly increased security lifetime of key usage for keys exchanged via KLJN or QKD.
- The first version of the KLJN book for World Scientific was completed.

- The first ever KLJN patent (with Elias Gonzalez and Robert Balog) is granted. It is about distributing unconditionally secure encryption keys over a wired network of hosts.
- Gunn-Allison-Abbott invented a passive attack against KLJN that utilizes transients. Even though the attack offers only a small information leak and the defense against it is very easy, the attacks is important because this is the first transient attack. Unfortunately, the paper has many problematic claims about security and physics. This situation triggered a rebuttal by us showing that these claims are flawed at all levels (security theory, circuit theory, and physics). Yet, the attack idea is valuable.
- Barry and Elias successfully defended their PhDs, increasing the number of KLJN-based PhD theses to three.
- The earlier PUF application got extended to an unconditionally secure credit card chip scheme (with Kamran Entesari, Claes Granqvist, and Chiman Kwan). The critical remarks of Robert Mingesz and Gergey Vadai, as referees, triggered the removal of unnecessary redundance from the original protocol and made it simpler.
- The proofreadings and corrections of the KLJN book (World Scientific) were completed by the end of the year.

- The KLJN book (including paths of developments and personal history) is published by World Scientific in both electronic and hardcover form.

back to the KLJN homepage