Cryptographic key distribution system, method for generating cryptographic keys in the cryptographic key distribution system, and communication system

By using chaotic wave correlation mixing of irreversible time-varying silicon chips in optical encryption terminals, the problem of traditional encryption technology generating and allocating cryptographic keys in secure communication is solved, and the secure generation and allocation of keys is realized, and the keys are generated and allocated in a safe manner is achieved. Provides unconditional security when facing quantum computers.

CN113169863BActive Publication Date: 2024-11-19CUP SCI INC +2
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN201980060319.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Priority Date
2018-09-14
Filing Date
2019-09-16
Publication Date
2024-11-19
Estimated Expiration
2039-09-16

AI Technical Summary

Technical Problem

The prior art is difficult to effectively generate and allocate cryptographic keys that cannot be eavesdropped in secure communications, and traditional encryption technologies have security challenges when facing quantum computers.

Method used

The security of the key is ensured by using the relevant mixing of chaotic waves in an irreversible time-varying silicon chip in an optical encryption terminal, and the optical encryption pattern and electromagnetic radiation source are used to generate and detect radiation amount and photometrics to ensure the security of the key. .

Benefits of technology

It realizes the secure generation and allocation of cryptographic keys between optical encryption terminals, avoids interception by eavesdroppers, and provides unconditional security when facing quantum computers.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN113169863B_ABST
    Figure CN113169863B_ABST
Patent Text Reader

Abstract

In one aspect, the present invention provides an optical encryption terminal for generating and distributing cryptographic key signals in a cryptographic key distribution system having at least two optical encryption terminals. The optical encryption terminal includes an electronic processing unit, and the optical encryption terminal is configured to selectively receive an optical input signal generated by an electromagnetic radiation source, and an optical input signal generated by another optical encryption terminal, and selectively output a first optical output signal to a detection element and a second optical output signal to another optical encryption terminal, wherein the first optical output signal is based on the optical input signal generated by another optical encryption terminal and is transformed according to an optical encryption pattern set at the optical encryption terminal. In addition, the optical encryption terminal is configured to determine the cryptographic key signal based on at least one radiation amount and / or light amount associated with the optical output signal detected by the detection element using the electronic processing unit.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates generally to optical encryption and, more particularly, to the generation and distribution of cryptographic key signals in a cryptographic key distribution system.In some aspects, the present invention relates to perfect secrecy cryptography by correlated mixing of chaotic waves in an irreversibly time-varying silicon chip. Background Art

[0002] Cryptography is the study of techniques to develop secure communications in the presence of a third party called an adversary. Typically, a cryptographic scheme shows two main components: the message to be sent and a cipher representing the algorithm that is applied and reversed to encode and decode the message, respectively.

[0003] Modern cryptography techniques are based on ciphers that are increasingly difficult to reverse without a key, which represents additional information in the form of a string of characters and / or numbers that must be possessed by both parties sharing the encrypted information.

[0004] With the digitization of many areas of everyday life, the adoption of digital data (e.g., the Internet and the Internet of Things IoT) as an increasingly dominant element of communication and commerce between different parties, devices and objects, there is a constant need for reliable tools that allow different parties to exchange sensitive information without being intercepted by third parties. Obviously, communication between two parties using a key is considered secure as long as the password is uniquely and exclusively known only to the two parties in communication. In addition, since the key must be known in advance to the parties in communication, the key must be distributed among the parties in communication in a secure manner without giving a third party (i.e., at least one eavesdropper) the opportunity to intercept the distributed key. Summary of the invention

[0005] The aim is therefore to provide various parties (individual users, computers, personal electronic devices, network objects and devices) with a device for generating and distributing cryptographic keys that cannot be cloned or intercepted by any type of eavesdropper.

[0006] The above-mentioned problems and objects are solved in a first aspect of the present invention by an optical encryption terminal for generating and distributing cryptographic key signals in a cryptographic key distribution system, wherein the cryptographic key distribution system has at least two optical encryption terminals. According to some illustrative embodiments of the present invention, the optical encryption terminal includes an electronic processing unit, wherein the optical encryption terminal is configured to selectively receive a first optical input signal generated by an electromagnetic radiation source and a second optical input signal generated by another optical encryption terminal, and selectively output a first optical output signal to a detection element and a second optical output signal to another optical encryption terminal, the first optical output signal being based on the second optical input signal transformed according to an optical encryption pattern set at the optical encryption terminal. In addition, the optical encryption terminal is configured to determine the cryptographic key signal based on at least one radiation amount and / or light amount associated with the first optical output signal detected by the detection element using the electronic processing unit.

[0007] In a second aspect of the present invention, a cryptographic key distribution system is provided, comprising: a first optical encryption terminal, which is configured to transform an optical input signal into an optical output signal according to a first optical encryption pattern set at the first optical encryption terminal; a second optical encryption terminal, which is configured to transform an optical input signal into an optical output signal according to a second optical encryption pattern set at the second optical encryption terminal, the first and second optical encryption terminals being optically coupled so as to exchange optical output signals with each other; a first electromagnetic radiation source, which is configured to provide a first input signal of electromagnetic radiation to the first optical encryption terminal; a second electromagnetic radiation source, which is configured to provide a second input signal of electromagnetic radiation to the second optical encryption terminal; a first detection element, which is configured to detect at least one first radiation amount and / or associated with the first optical output signal of the first optical encryption terminal a light quantity, wherein the first optical output signal is based on a second input signal continuously transformed by the second and first optical encryption terminals; a second detection element, which is configured to detect at least one second radiation quantity and / or light quantity associated with the second optical output signal of the second optical encryption terminal, wherein the second optical output signal is based on the first input signal continuously transformed by the first and second optical encryption terminals; a first electronic processing unit, which is coupled to the first detection element, and the first electronic processing unit is configured to determine a first cryptographic key signal based on at least one first radiation quantity and / or light quantity detected by the first detection element; and a second electronic processing unit, which is coupled to the second detection element, and the second electronic processing unit is configured to determine a second cryptographic key signal based on at least one second radiation quantity and / or light quantity detected by the second detection element.

[0008] In a third aspect of the present invention, a method for generating cryptographic keys in a cryptographic key distribution system having a first optical encryption terminal and a second optical encryption terminal is provided, the method comprising: providing electromagnetic radiation from a first electromagnetic radiation source to the first optical encryption terminal as a first optical input signal; transforming the first optical input signal into a first optical output signal according to a first optical encryption pattern formed at the first optical encryption terminal; providing the first optical output signal to the second optical encryption terminal as a second optical input signal; transforming the second optical input signal into a second optical output signal according to a second optical encryption pattern formed at the second optical encryption terminal; detecting at least one first radiation amount and / or light amount associated with the second optical output signal; and determining a first cryptographic key based on the detected at least one first radiation amount and / or light amount.

[0009] With reference to the above-mentioned first to third aspects of the present disclosure, some illustrative embodiments of the present disclosure will be described in more detail.

[0010] After a complete explanation of the present disclosure, those skilled in the art will understand that the optical encryption terminal of the first aspect can allow the generation and distribution of cryptographic key signals in a cryptographic key distribution system between two optical encryption terminals, wherein cryptographic keys are generated and shared between two parties in communication through the optical encryption terminal based on optical signals exchanged between the two optical encryption terminals in the cryptographic key distribution system. Here, the generated and shared cryptographic keys can be completely generated only after the signal is subjected to the action of the optical encryption terminal in communication. Therefore, an eavesdropper will not intercept a cryptographic key that is not completely generated in the communication path between the two communicating parties. It can be determined that the complete cryptographic key signal is based on only the combined action of the two optical encryption terminals in the optical communication.

[0011] According to some demonstrative embodiments of the first aspect, the optical encryption pattern may include a plurality of regions of a first material randomly arranged in a second material, wherein the first and second materials differ from each other in at least one optical property.

[0012] Therefore, optical encryption of exchanged optical signals can be provided, increasing the security level of cryptographic key signals to be generated and distributed in a cryptographic key distribution system employing at least two optical encryption terminals of the first aspect.

[0013] According to some further advantageous examples herein, the first material may have greater absorption and / or transmission and / or diffraction properties with respect to electromagnetic radiation than the second material.

[0014] Hence, a simple and cheap way of implementing an optical encryption pattern in an optical encryption terminal may be provided.

[0015] According to other advantageous embodiments of the present invention, the plurality of areas of the first material may be arranged according to a predefined pattern, such as a user individual pattern, for example a fingerprint of a user of the optical encryption terminal. Thus, when using, for example, a user's fingerprint, a unique optical encryption pattern may be achieved.

[0016] According to some other exemplary embodiments of the first aspect, the optical encryption pattern may be formed in the photonic nanostructure.

[0017] By using photonic nanostructures with random and / or chaotic scattering components, optical encryption patterns for multiple purposes can be realized in a simple and inexpensive manner. Alternatively, optical encryption patterns for single purposes can be realized, for example, by temporary scattering components, i.e., scattering components that maintain a certain scattering behavior only during a certain time interval, which is long enough that a single cryptographic key can be generated at each of the first and second encryption terminals, i.e., the period of time during which electromagnetic radiation travels back and forth between the first and second encryption terminals.

[0018] According to some advantageous embodiments of the present invention, the photonic nanostructure can be implemented, for example, in silicon-on-insulator technology, where light is confined to the plane of the chip by total internal reflection. According to some specific illustrative examples, the scattering components can be obtained by drilling holes in the top layer of silicon, with a diameter between 0.1*λ and λ, where λ is the wavelength of light in the medium (e.g., in the C-band and L-band, i.e., the wavelength range is 1530nm to 1625nm). The top surface and holes of the chip can be uncoated, or coated with other materials, such as polymers (e.g., PMMA, SU8) or dielectric layers (e.g., SiO2), to achieve robustness and packaging requirements. Other embodiments can be implemented in random scattering materials, including high refractive contrast semiconductors (e.g., also GaAs, GaP, Si3N4, INP, etc.), treated to randomly scatter light in or out of the plane of the chip, or colloidal solutions of strongly scattering particles in a polymer matrix (e.g., PMMA, polyimide, SU8, PDMS, etc.), with the strongly scattering particles having a size comprised between 0.1*λ and λ (e.g., particles made of metal, alumina, TiO2, silicon, etc.). Additionally, or alternatively, optical encryption patterns can be implemented in biological materials or by directly using human interfaces, such as, for example, a human fingerprint or retina.

[0019] Regarding the second aspect, the cryptographic key distribution system thus provided can allow for the generation and distribution of cryptographic key signals between first and second optical encryption terminals, wherein the first and second cryptographic key signals are generated and shared between two parties in communication via the first and second optical encryption terminals based on optical signals exchanged between the first and second optical encryption terminals of the cryptographic key distribution system. Here, the generated and shared first and second cryptographic key signals can be fully generated only after the signals are subjected to the action of both the first and second optical encryption terminals in the communication. Therefore, an eavesdropper will not intercept the first and second cryptographic key signals that can only be determined after the combined action of the first and second optical encryption terminals in a known optical communication.

[0020] According to some demonstrative embodiments of the second aspect, each of the first and second optical encryption patterns may include a plurality of regions of a first material arranged in a second material according to a predetermined pattern, wherein the first and second materials differ from each other in at least one optical property.

[0021] Thus, optical encryption of exchanged optical signals can be provided, increasing the security level of cryptographic key signals to be generated and distributed in a cryptographic key distribution system.

[0022] According to some further advantageous examples herein, the first material may have greater absorption and / or transmission and / or diffraction properties with respect to electromagnetic radiation than the second material.

[0023] Hence, a simple and cheap way of realizing the first and second optical encryption patterns may be provided.

[0024] According to other advantageous embodiments herein, the plurality of areas of the first material may be arranged according to a fingerprint of a user of the first or second optical encryption terminal.

[0025] Therefore, after using the user's fingerprint, a unique optical encryption pattern can be achieved.

[0026] According to some other exemplary embodiments of the second aspect, each of the first and second optical encryption patterns may be formed in a photonic nanostructure.

[0027] By means of the photonic nanostructure, each of the first and second optical encryption patterns can be realized in a simple and inexpensive manner.

[0028] According to some advantageous embodiments of the present invention, the photonic nanostructure can be implemented, for example, by silicon-on-insulator technology, in which light is confined to the plane of the chip by total internal reflection. The scattering component is obtained by drilling holes in the top layer of silicon, with a diameter between 0.1λ and λ, where λ is the wavelength of light in the medium. The top surface and holes of the chip can be uncoated or coated with other materials, such as polymers (e.g., PMMA, SU8) or dielectric layers (e.g., SiO2) to achieve robustness and packaging requirements. Other embodiments can be implemented in random scattering materials, including high refractive contrast semiconductors (e.g., also including GaAs, Gap, Si3N4, INP, etc.), treated to randomly scatter light in or out of the plane of the chip, or colloidal solutions of strongly scattering particles in polymer matrices (e.g., PMMA, polyimide, SU8, PDMS, etc.), with the strongly scattering particles having a size contained between 0.1*λ and λ (e.g., particles made of metal, alumina, TiO2, silicon, etc.). Optical encryption patterns can also be implemented in biomaterials or by directly using human interfaces, such as human fingerprints or retinas.

[0029] Regarding the third aspect, the provided method can allow for the generation of cryptographic keys in a cryptographic key generation and distribution system, wherein a first cryptographic key is generated at a first optical encryption terminal based on an optical signal exchanged between the first and second optical encryption terminals of the cryptographic key distribution system between two parties in communication through the first and second optical encryption terminals. Here, the generated first cryptographic key can be completely generated only after the signal has passed through the actions of both the first and second optical in communication. Therefore, an eavesdropper will not intercept the first cryptographic key determined only at the first optical encryption terminal.

[0030] According to some illustrative embodiments of the present invention, the method may also include: providing electromagnetic radiation from a second electromagnetic radiation source as a third optical input signal to the second optical encryption terminal; transforming the third optical input signal into a third optical output signal according to the second optical encryption pattern; providing the third optical output signal as a fourth optical input signal to the first optical encryption terminal; transforming the fourth optical input signal into a fourth optical output signal according to the first optical encryption pattern; detecting at least one second radiation amount and / or light amount associated with the fourth optical output signal; and determining a second cryptographic key based on the detected at least one second radiation amount and / or light amount.

[0031] According to some advantageous embodiments herein, the first and second cryptographic keys may be identical when the electromagnetic radiation of the first and second electromagnetic radiation sources provide the same optical input signal.

[0032] The present disclosure will now be described with reference to the accompanying drawings. Various structures, systems and devices are schematically depicted in the accompanying drawings so as to illustrate only and not to obscure the present disclosure due to details known to those skilled in the art. However, the drawings are included to describe and explain illustrative examples of the present disclosure. The words and phrases used herein should be understood and interpreted as having meanings consistent with the understanding of those words and phrases by those skilled in the relevant art. No special definition of a term or phrase, i.e., a definition different from the ordinary or customary meaning understood by those skilled in the art, is intended to be implied by the consistent use of the term or phrase herein. To the extent that a term or phrase is intended to have a special meaning, i.e., not a meaning understood by a technician, such a special definition should be clearly stated in the specification in a definition manner that directly and clearly provides the special definition of the term or phrase. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Additional effects and advantages will be apparent from the following detailed description taken in conjunction with the accompanying drawings, in which:

[0034] Figure 1 schematically illustrates a cryptographic key distribution system according to some demonstrative embodiments of the invention;

[0035] Figure 2a and 2b A process flow is shown for a method of generating and distributing cryptographic keys in a cryptographic key distribution system according to some demonstrative embodiments of the invention;

[0036] Figure 3 schematically illustrates an optical encryption terminal according to some illustrative embodiments of the present disclosure;

[0037] Figure 4 schematically illustrates an optical encryption terminal according to some further illustrative embodiments of the present disclosure;

[0038] Figure 5 schematically illustrates a cryptographic key distribution system according to some demonstrative embodiments of the present disclosure;

[0039] Figures 6a to 6e schematically illustrates an optical encryption terminal, and measurements performed at the optical encryption terminal, according to some particular illustrative embodiments of the present disclosure;

[0040] Figure 7 schematically illustrates an optical encryption terminal according to other illustrative embodiments of the present disclosure;

[0041] Figures 8a to 8d An illustrative process for encrypting and decrypting data in a cryptographic key generation and distribution system according to some illustrative embodiments of the present disclosure is shown;

[0042] Fig. 9 shows a block representation of a communication system according to the present invention; and

[0043] Fig.10a shows experimental data illustrating the power density spectra of the sources used by Alice and Bob,

[0044] Fig.10b Experimental data illustrating the power density spectrum measured by Alice are shown;

[0045] Fig.10c shows a graph illustrating the relative differences between the spectra measured by Alice and Bob; and

[0046] Fig.10d Shown by Fig.10b The spectrum of the generated key is represented as a binary square point matrix;

[0047] Fig.11 for Fig.11 (a) to (c) schematically illustrate a protocol scheme for perfect confidentiality key generation over a classical channel;

[0048] Fig.12 for Fig.12 The diagrams in (a) to (f) schematically illustrate the security of the protocol against time domain and spectral attacks;

[0049] Fig.13 Schematically illustrates integrated fingerprint silicon chip design and chaos analysis according to some particular illustrative embodiments of the present disclosure;

[0050] Fig.14 Schematically shows the wave analysis of a fingerprint chip from a FDTD simulation performed by the inventors;

[0051] Fig.15 An experiment on a single fingerprint chip is shown; and

[0052] Fig.16 Experiments on key distribution and attacks in the spectral domain are shown. DETAILED DESCRIPTION

[0053] In the following detailed description, at least some illustrative embodiments of the above-mentioned first to third aspects of the present disclosure will be described in more detail with reference to the accompanying drawings.

[0054] Figure 1A cryptographic key distribution system 100 is schematically shown, in accordance with some demonstrative embodiments of the present disclosure. Cryptographic key distribution system 100 includes a first optical encryption terminal 110A and a second optical encryption terminal 110B, each configured to transform an optical input signal into an optical output signal.

[0055] like Figure 1 As shown, the first optical encryption terminal 110A may include a first optical pad 112A, and the first optical pad 112A has a first optical encryption pattern 114A, which may be disposed at the first optical encryption terminal 110A. According to some illustrative examples herein, the first optical encryption pattern 114A may be provided by a first optical scattering and / or diffraction pattern disposed at the first optical pad 112A. According to some specific illustrative examples herein, the first optical encryption pattern 114A may be permanently formed in the optical pad 112A, for example, by etching, cutting, scratching, etc. a target or desired pattern into the first optical pad 112A, the first optical scattering and / or diffraction pattern may be formed in the first optical pad 112A. The first optical scattering and / or diffraction pattern may be formed as a photonic nanostructure, an exemplary manufacture of which is as described above.

[0056] Alternatively, the first optical encryption pattern 114A may be temporarily formed during the time when the cryptographic key distribution system 100 is used to generate and distribute cryptographic keys. For example, a user of the first optical encryption terminal 110A may temporarily couple the first optical encryption pattern 114A to the first optical encryption terminal 110A at least during the time when at least one cryptographic key is to be generated and / or distributed in the cryptographic key distribution system 100. Figure 7 Some illustrative and non-limiting example implementations of the first optical encryption pattern are described.

[0057] According to some illustrative embodiments of the present disclosure, the second optical encryption terminal 110B may include a second optical pad 112B having a second optical encryption pattern 114B, which may be disposed at the second optical encryption terminal 110B. According to some illustrative examples herein, similar to the first optical encryption pattern 114A at the first optical pad 112A, the second optical encryption pattern 114B may be provided by a second optical scattering and / or diffraction pattern, which may be disposed at the second optical pad 112B. According to some particular illustrative examples herein, the second optical encryption pattern 114B may be permanently formed in the second optical pad 112B, for example, by etching, cutting, scratching, etc. a target or desired pattern into the second optical pad 112B, the second optical scattering and / or diffraction pattern may be formed in the second optical pad 112B. Alternatively, the second optical encryption pattern 114B may be temporarily formed at least during the time when the cryptographic key distribution system 100 is used to generate and distribute cryptographic keys. For example, a user of the second optical encryption pattern 114B may temporarily couple the second optical encryption pattern 114B to the second optical encryption terminal 110BA at least during a period of time when at least one cryptographic key is to be generated and / or distributed in the cryptographic key distribution system 100. Figure 7 Some illustrative and non-limiting example implementations of the second optical encryption pattern 114B will be described.

[0058] According to some demonstrative embodiments of the present disclosure, a first electromagnetic radiation source 120A may be provided, the first electromagnetic radiation source 120A being configured to provide a first input signal OIS-A of electromagnetic radiation to the first optical pad 112A. According to some demonstrative embodiments of the present disclosure, the first electromagnetic radiation source 120A may be an integral element of the first optical encryption terminal 110A. This does not impose any limitation on the present disclosure, and alternatively, the first electromagnetic radiation source 120A may be provided as an external element relative to the first optical encryption terminal 110A. For example, the first source 120A may be temporarily coupled to the first optical encryption terminal 110A during the time when at least one cryptographic key is generated in the cryptographic key distribution system 100.

[0059] According to some demonstrative embodiments of the present disclosure, the first input signal OIS-A of electromagnetic radiation output by the first electromagnetic radiation source 120A may be optically coupled to the first optical pad 112A so as to provide the first input signal OIS-A as an optical input signal to the first optical pad 112A. Therefore, it will be understood by those skilled in the art that the first optical encryption terminal 110A may have at least one input channel (not shown) through which the electromagnetic radiation emitted by the first electromagnetic radiation source 120A may be optically coupled to the first optical encryption terminal 110A when the first source 120A is not an integral part of the first optical encryption terminal 110A. For example, the first optical pad 112A may have at least one input channel (not shown) through which the electromagnetic radiation emitted by the first electromagnetic radiation source 120A may be optically input to the first optical pad 112A.

[0060] According to some demonstrative embodiments of the invention, a first detection element 130A may be provided, which is configured to detect at least one first radiation amount and / or photometric amount of a first optical output signal OOS-AB output by the first optical pad 112A toward the first detection element 130A.

[0061] Those skilled in the art will appreciate that the first radiation quantity may represent the sum of radiation quantities of various wavelengths (λ), and the first photometric quantity may be represented by a weighted sum of radiation quantities of various wavelengths, with the weighting factor defining the weight of the weighted sum according to a photopic or scotopic spectral luminous efficiency function. Typically, the radiation quantity and / or photometric quantity may be based on at least one of the luminous flux, luminous intensity, brightness, and illuminance of the first optical output signal OOS-AB. The radiation quantity and photometric quantity are related by the following equation:

[0062] X v =X e *V(λ)*K m ,

[0063] Among them, x v represents the photometric quantity, X e represents the radiation, V(λ) represents the spectral luminous efficiency, and the factor K m Indicates the efficiency factor, such as K for red light m It can be about 683lm / W, and the K of green light m can be about 683.1lm / W, and the K of blue light m It can be about 68.3lm / W.

[0064] According to some demonstrative embodiments of the invention, first detection element 130A may be provided as an integral element of first optical encryption terminal 110A. According to alternative embodiments of the invention, first detection element 130A may be an external element of first optical encryption terminal 110A, first detection element 130A being optically coupled to first optical encryption terminal 110A so as to receive first optical output signal OOS-AB of first optical pad 112A and detect at least one first radiation quantity and / or photometric quantity associated with first optical output signal OOS-AB of first optical pad 112A, the first optical output signal OOS-AB being based on an optical input signal received by first optical encryption terminal 110A as an optical signal output by second optical encryption terminal 110B to first optical encryption terminal 110A.

[0065] According to some demonstrative embodiments of the invention, the detection result of at least one first radiation quantity and / or light quantity obtained by the first detection element 130A may be output as a first measurement signal SA to the first electronic processing unit 116A. According to some illustrative examples herein, the first electronic processing unit 116A may be an integral component of the first optical encryption terminal 110A, and the first electronic processing unit 116A is configured to determine the first cryptographic key signal keyA based on the at least one first radiation quantity and / or light quantity provided to the first electronic processing unit 116A via the first measurement signal SA.

[0066] Reference is now made to the second optical encryption terminal 110B. According to some demonstrative embodiments of the present disclosure, a second electromagnetic radiation source 120B may be provided, and the second electromagnetic radiation source 120B may be provided, the second electromagnetic radiation source 120B being configured to provide a second input signal OIS-B of electromagnetic radiation to the second optical pad 112B. According to some demonstrative embodiments of the present disclosure, the second electromagnetic radiation source 120B may be an integral element of the second optical encryption terminal 110B. This does not impose any limitation on the present disclosure, and alternatively, the second electromagnetic radiation source 120B may be provided as an external element to the second optical encryption terminal 110B. For example, during the time when at least one cryptographic key is generated and / or distributed in the cryptographic key distribution system 100, the second source 120B may be temporarily coupled to the second optical encryption terminal 110B.

[0067] According to some demonstrative embodiments of the present disclosure, the second input signal OIS-B of electromagnetic radiation output by the second electromagnetic radiation source 120B may be optically coupled to the second optical pad 112B to provide the second input signal OIS-B as an optical input signal to the second optical pad 112B. Therefore, it will be understood by those skilled in the art that the second optical encryption terminal 110B may have at least one input channel (not shown) through which the electromagnetic radiation emitted by the second electromagnetic radiation source 120B may be optically coupled to the second optical encryption terminal 110B when the second source 120B is not an integral part of the second optical encryption terminal 110B. For example, the second optical pad 112B may have at least one input channel (not shown) through which the electromagnetic radiation emitted by the second electromagnetic radiation source 120B may be optically input to the second optical pad 112B.

[0068] According to some demonstrative embodiments of the invention, a second detection element 130B may be provided, which is configured to detect at least one second radiation amount and / or photometric amount of a second optical output signal OOS-BA output by second optical pad 112B toward second detection element 130B.

[0069] According to some demonstrative embodiments of the present disclosure, the first and second detection elements 130A and 130B are of the same type. For example, the first and second detection elements 130A and 130B may have equal resolutions, relative amplitudes, and the like.

[0070] Those skilled in the art will appreciate that the second radiation quantity may represent the sum of radiation quantities of various wavelengths (λ), and the second photometric quantity may be represented by a weighted sum of radiation quantities of various wavelengths, the weighting factor defining the weight of the weighted sum according to the photopic or scotopic spectral luminous efficiency function. Typically, the radiation quantity and / or photometric quantity may be based on at least one of the luminous flux, luminous intensity, brightness, and illuminance of the second optical output signal OOS-BA.

[0071] According to some demonstrative embodiments of the invention, second detection element 130B may be provided as an integral element of second optical encryption terminal 110B. According to alternative embodiments of the invention, second detection element 130B may be an external element of second optical encryption terminal 110B, second detection element 130B being optically coupled to second optical encryption terminal 110B so as to receive second optical output signal OOS-BA of second optical pad 112B, and detecting at least one second radiation quantity and / or light quantity associated with second optical output signal OOS-BA of second optical pad 112b, the second optical output signal OOS-BA being based on an optical input signal received by second optical encryption terminal 110B as an optical signal output by first optical encryption terminal 110A toward second optical encryption terminal 110B.

[0072] According to some demonstrative embodiments of the invention, the detection result of at least a second radiation amount and / or light amount obtained by the second detection element 130B can be output to the second electronic processing unit 116B as a second measurement signal SB. According to some illustrative examples herein, the second electronic processing unit 116B can be an integral component of the second optical encryption terminal 110B, and the second electronic processing unit 116B is configured to determine the second cryptographic key signal keyB based on at least a second radiation amount and / or light amount provided to the second electronic processing unit 116B via the second measurement signal SB.

[0073] After reading this disclosure in its entirety, those skilled in the art will appreciate that Figure 1 As shown, the cryptographic key distribution system 100 may include first and second optical encryption terminals 110A, 110B optically coupled to exchange optical output signals with each other, such as via signals OOS-A and OOS-B. Figure 1 . In particular, the electromagnetic radiation provided by the first electromagnetic radiation source 120A can be transformed by the first optical encryption terminal 110A into an optical output signal OOS-A of the first encryption terminal 110A. The optical output signal OOS-A of the first optical encryption terminal 110A can be provided to the second optical encryption terminal 110B as an optical input signal, and the optical input signal is transformed into an optical output signal toward the second detection element 130B through the second encryption pattern 114B of the second optical encryption terminal 110B. In particular, the electromagnetic radiation generated by the first electromagnetic radiation source 120A can be continuously transformed through the first optical encryption pattern 114A of the first optical encryption terminal 110A and the second optical encryption pattern 114B of the second optical encryption terminal 110B before being detected by the second detection element 130B.

[0074] In addition, the electromagnetic radiation of the second electromagnetic radiation source 120B can be detected by the first detection element 130A after being continuously transformed by the second optical encryption pattern 114B of the second optical encryption terminal 110B and the first optical encryption pattern 114A of the first optical encryption terminal 110A. Therefore, the optical key signal can be exchanged between the first and second optical encryption terminals 110A, 110B, thereby exchanging cryptographic key signals keyA and keyB at the first and second optical encryption terminals 110A, 110B.

[0075] According to some illustrative and non-limiting examples of the present invention, the first and / or second detection elements may be implemented by charge coupled devices (CCDs), photoconductive detectors (eg, photodiodes and phototransistors), and the like.

[0076] According to some illustrative and non-limiting examples of the present invention, the first and / or second electronic processing units may be implemented by a computing device (eg, a central processing unit (CPU), etc.).

[0077] According to some illustrative and non-limiting examples of the present disclosure, the first and second optical encryption terminals 110A, 110B may be optically coupled via at least one optical fiber, such as a single-mode optical fiber, a multi-mode optical fiber, etc. For example, the optical fiber may have a core width of 10 microns or less, for example, in the range of about 1 micron to about 10 microns, such as about 7 microns, and a cladding thickness of 125-150 microns.

[0078] According to some illustrative and non-limiting examples of the present disclosure, the source may be a broadband signal generated by a diode amplifying in the C+L band by amplified spontaneous emission, or a broadband signal generated by a tunable signal frequency laser in the standard telecommunication C+L band.

[0079] According to some illustrative and non-limiting examples of the present disclosure, the optical fiber can be directly coupled to at least one of the optical encryption terminals 110A, 110B, that is, collimation optics can be avoided when coupling the optical fiber to at least one of the optical encryption terminals 110A, 110B. In some specific illustrative examples herein, at least one of the optical detection elements 130A, 130B can include a fiber optic spectrum analyzer.

[0080] According to some illustrative embodiments of the present disclosure, Figure 2a The process flow shown is for Figure 1 The cryptographic key distribution system 100 is used to describe methods of generating cryptographic keys according to some demonstrative embodiments of the present disclosure.

[0081] According to some demonstrative embodiments herein, the first optical encryption terminal 110A may be provided with electromagnetic radiation from a first electromagnetic radiation source 120A as a first optical input signal OIS-A (refer to Figure 2a , or step S210 in the above procedure.

[0082] The first optical input signal OIS-A may be transformed into a first optical input signal OOS-A (reference Figure 2a (S215 in).

[0083] The second optical encryption terminal 110B may be provided with the first optical input signal OOS-A as the second optical input signal (refer to Figure 2a The second optical input signal may be transformed into a second optical output signal OOS-BA (refer to S220 in FIG. 1 ) according to the second optical encryption pattern 114B formed at the second optical encryption terminal 110B. Figure 2a (S225 in).

[0084] At least one first radiation quantity and / or light quantity associated with the second optical output signal OOS-BA may be detected by the second detection element 130B (see Figure 2a S230 in ).

[0085] like Figure 1 As schematically shown in FIG. 1 , the second detection element 130B may provide one or more detected radiation quantities and / or light quantities as an electronic measurement signal SB to the second electronic processing unit 116B.

[0086] The first cryptographic key signal keyB (reference 10) may be determined based on at least one first radiation quantity and / or light quantity provided to the second processing unit 116B via the electronic measurement signal SB. Figure 2a in S235).

[0087] about Figure 2b , will combine Figure 2a The processing flow shown in Figure 1 The cryptographic key distribution system 100 is used to describe a method of generating another cryptographic key according to some demonstrative embodiments of the present disclosure.

[0088] According to some demonstrative embodiments of the present disclosure, the second optical pad 112B of the second optical encryption terminal 110B may be provided with electromagnetic radiation from the second electromagnetic radiation source 120B as the third optical input signal OIS-B (refer to Figure 2b S240 in ).

[0089] According to the second optical encryption pattern 114B of the second optical encryption terminal 110B, the third optical input signal can be transformed into a third optical output signal OOS-B (reference signal) from the second optical encryption terminal 110B to the first optical encryption terminal 110A. Figure 2b S245 in ).

[0090] The first optical encryption terminal 110A may be provided with the third optical output signal OOS-B of the second optical encryption terminal 110B as the fourth input signal of the first optical encryption terminal 110A (refer to Figure 2b S250 in the ).

[0091] The fourth optical input signal can be transformed into a fourth optical output signal OOS-AB in the first optical encryption terminal 110A according to the first optical encryption pattern 114A (refer to Figure 2b in S255).

[0092] The fourth optical output signal OOS-AB of the first optical pad 112A of the first optical encryption terminal 110A can be received by the first detection element 130A and at least one second radiation quantity and / or light quantity associated with the fourth optical output signal OOS-AB can be detected (refer to Figure 2b Based on at least one second radiation quantity and / or light quantity, the electronic measurement signal SA can be output by the first detection element 130A to the first electronic processing unit 116A, and a second cryptographic key signal can be determined based on at least one second radiation quantity and / or light quantity (refer to Figure 2b in S265).

[0093] Thus, in the cryptographic key distribution system 100 cryptographic key signals keyA and keyB may be generated at the outputs of the first and second optical encryption terminals 110A and 110B.

[0094] Those skilled in the art will appreciate that Figure 2a and Figure 2b The processing flow shown in , makes it possible to determine key signals keyA and keyB simultaneously or continuously. For example, the key signal keyA can be determined before the key signal keyB is determined, and vice versa.

[0095] According to some illustrative embodiments of the present disclosure, the first and third optical input signals OIS-A and OIS-B may be similar, e.g., identical. For example, the first and second sources 120A and 120B may emit substantially equal electromagnetic radiation, e.g., substantially equal emission power density spectra of the electromagnetic radiation. Here, the second and fourth optical output signals OOS-AB and OOS-BA may be identical and the determined cryptographic key signals keyA and keyB may be identical. Thus, cryptographic keys may be exchanged between users of the optical encryption terminals 110A and 110B.

[0096] about Figure 3 , the optical encryption terminal 312 according to some illustrative embodiments of the present disclosure will be described in more detail below.

[0097] Figure 3 Schematically shown, in a schematic block diagram, is a method for distributing a cryptographic key in a cryptographic key distribution system (e.g., as described above with respect to Figure 1 The optical encryption terminal 312 for generating and distributing cryptographic key signals in the cryptographic key distribution system 100) is provided. The cryptographic key distribution system has the optical encryption terminal 312 and at least one other optical encryption terminal (not shown), wherein the optical encryption terminal 312 includes an electronic processing unit 324.

[0098] According to some illustrative examples, optical encryption terminal 312 may be configured to selectively receive a first optical input signal (given by Figure 3 The arrows 303 and 308 in FIG. 3 represent an optical input signal generated by the electromagnetic radiation source 348 and a second optical input signal (generated by Figure 3 Indicated by arrow 304 in FIG. 3 , it represents an optical input signal generated by another optical encryption terminal (not shown) which is optically input to the optical encryption terminal 312 via the first optical input / output port 351).

[0099] According to some demonstrative embodiments of the present disclosure, the optical encryption terminal 312 may be configured to selectively output an optical output signal (by a second optical input / output port 350) to the detection element 363. Figure 3 301 in the figure), and outputting an optical output signal to another optical encryption terminal (not shown) via a first optical input / output port 351, as indicated by Figure 3 The optical output signals 301 and 305 are based on the corresponding input signals 304 and 302, which are similar to the above-mentioned Figure 1 The optical encryption patterns 114A and 114B are transformed by the optical encryption pattern 316 provided at the optical encryption terminal 312. The optical output signal 301 output to the detection element 363 is based on the optical input signal 304 generated by another optical encryption terminal (not shown).

[0100] Using the electronic processing unit 324, a cryptographic key signal can be determined based on at least one radiation quantity and / or light quantity (similar to the first and second radiation quantities and / or light quantities described above) detected by the detection element 363, coupled to the optical encryption terminal 312 via the first optical input / output port 351 and transformed by the optical encryption pattern 316.

[0101] According to some illustrative embodiments of the present disclosure, optical input signals 303 and 308 may be provided to a second optical input / output port 350 and may be directed to an optical encryption pattern 316 for transforming the optical input signals 303, 308 into an optical output signal 305, which is output from the first optical input / output port 351 to another optical encryption terminal (not shown).

[0102] According to some demonstrative embodiments of the present disclosure, a beam splitter 360 may be provided between the detection element 363 and the second optical input / output port 350 of the optical encryption terminal 312. Alternatively, the beam splitter 360 may be provided between the second optical input / output port 350 and the optical encryption pattern 316 as an integral part of the optical encryption terminal.

[0103] According to some demonstrative embodiments of the present disclosure, electromagnetic radiation generated by electromagnetic radiation source 348 may be collimated into a collimated beam by collimating optics 361 and directed to beam splitter 360 by reflective surface 365. This is not intended to limit the present disclosure in any way, and those skilled in the art will appreciate that at least one of collimating optics 361 and reflective element 365 may be optional.

[0104] According to some illustrative embodiments of the present disclosure, beam splitter 360 may be a polarization beam splitter. According to some particular illustrative examples herein, polarization element 364 may be disposed between source 348 and beam splitter 360. When the orientation of polarization element 364 is selected to be perpendicular to the orientation of polarization beam splitter 360, that is, electromagnetic radiation passing through polarization element 364 may be polarized such that the polarized electromagnetic radiation may not pass through polarization beam splitter toward detection element 363. Thus, interference between the optical output signal 301 transformed by optical encryption pattern 316 at detection optical element 363 and the optical input signal generated by source 348 is avoided.

[0105] Those skilled in the art will understand that Figure 3 The encryption terminal scheme depicted in can be implemented with fiber optic technology, where a single mode or multimode optical fiber is placed after a collimating lens to guide the optical energy along paths 361, 301, and the beam splitter 360 is a fiber circulator. The polarization element 364 can be implemented by a fiber-based polarization controller.

[0106] about Figure 4 , schematically illustrates a cryptographic key distribution system (e.g., as described above with respect to Figure 1 The optical encryption terminal 418 of the cryptographic key distribution system 100 described above (which can be used with the above Figures 1 to 3 Generation and distribution of cryptographic key signals in at least one optical encryption terminal similar to that described in .

[0107] like Figure 4 Schematically shown in FIG. 4 , the optical encryption terminal 418 may include an optical encryption pattern 416, a source 420, a detection element 422, and an electronic processing unit 424. According to some illustrative examples herein, the source 420, the detection element 422, the optical encryption pattern 416, and the electronic processing unit 424 may be integrated into the optical encryption terminal 418. Alternatively, at least one of the source 420 and the detection element 422 may be optional, and the optical encryption terminal 418 may be configured to be coupled to the optional and omitted elements.

[0108] According to some illustrative embodiments of the present disclosure, the source 420 of the optical encryption terminal 418 can generate an electromagnetic radiation signal 426, including and not limited to, for example, a Gaussian wave packet and / or a wave packet whose spectrum has a profile of a Lorentz or Cauchy distribution and / or a Voigt profile, with a characteristic width around a central wavelength. The emitted electromagnetic radiation 426 can be emitted along an optical path 428 toward the optical encryption pattern 416, and then leave the optical encryption terminal 418 toward another optical encryption terminal (not shown).

[0109] According to some illustrative embodiments of the present disclosure, the optical encryption terminal 418 can receive an optical input signal along the optical path 432 and directed toward the detection element 422 from other optical encryption terminals (not shown) exposed to the optical encryption pattern 416, as an optical signal that is the combined effect of the optical encryption pattern 416 passing through the optical encryption terminal 418 and another optical encryption pattern (not shown) of the other optical encryption terminals (not shown).

[0110] According to some demonstrative embodiments of the present disclosure, the detection element 422 of the optical encryption terminal 418 may be configured to determine a power distribution spectrum (PDS) of the received optical signal 430, such as Figure 4 Here, the power density spectrum (pds or PDS) or spectral density S of the optical signal x(t) is shown schematically in FIG. xx (f) Describe the power distribution as the frequency components that make up the signal. pds is represented as a plot with frequency on the horizontal axis and optical power distribution on the vertical axis.

[0111] According to some illustrative but non-limiting embodiments of the present disclosure, the detection element 422 may include a spectrum analyzer or an optical spectrometer.

[0112] According to some demonstrative embodiments of the present disclosure, the electronic processing unit 424 may be configured to determine a cryptographic key signal 436 based on the output spectrum 434 detected by the detection element 422. Thus, the cryptographic key signal may be generated only after the optical signal 430 is detected which has been subjected to the combined effect of the optical encryption pattern 416 and other optical encryption patterns (not shown) of an additional optical encryption terminal (not shown).

[0113] about Figure 5 , a cryptographic key distribution system 510 according to some demonstrative embodiments of the present disclosure will be described.

[0114] According to some illustrative embodiments of this document, Figure 5The cryptographic key distribution system 510 shown may represent a system for generating and distributing cryptographic keys between two users 512 and 514, and in particular a system for making two identical copies available at two optical encryption terminals of users 512 and 514, as described above with respect to Figure 1 The cryptographic key distribution system 100 described above is similar. The two users 512 and 514 can be separated from each other by an arbitrary distance L, which can be as large as 100 kilometers for intra-city networks and larger for inter-city networks. Long distances can be achieved using standard telecommunication optical amplifiers. According to some illustrative embodiments of the present disclosure, electromagnetic radiation signals 527a (electromagnetic radiation signal from user 512 to user 514) and 527b (from user 514 to user 512) can be exchanged between the two users 512 and 514, the two signals 527a, 527b being subjected to transformations associated with optical encryption patterns as schematically shown by area 516 representing the combined effect of the two optical encryption patterns. Each of users 512 and 514 may have an optical encryption terminal for deriving cryptographic keys from the optical spectrum (or power distribution spectrum) 518a, 518b measured at each end, i.e., user 512 measures PDS from electromagnetic radiation 527b, while user 514 measures PDS of electromagnetic radiation 527a. Each of users 512, 514 may employ a combination of the above Figures 1 to 4 The optical encryption terminal.

[0115] According to some illustrative embodiments of the present disclosure, a cryptographic key may be determined based on the measured PDS 518a and 518b, respectively. According to an illustrative but non-limiting example, each value of the spectrum 518a and 518b may be presented for extracting the magnitude (i.e., the mantissa) between zero and one and the absolute value of the exponent of the value. For example, when the mantissa is rounded to the nearest integer and the absolute value of the determined exponent is converted to a binary value, five binary digits may be obtained. For example, a value in the spectrum is measured as the number 2.2*10 13 (in arbitrary units), the mantissa is 0.22, while the absolute value of the exponent is 12. When the mantissa is rounded to the nearest integer, "0" is obtained, and when the absolute value of the exponent is converted to four binary digits, the sequence binary 1100 is obtained. Therefore, the five-bit binary sequence 01100 represents the measurement number 2.2*10 in arbitrary units. 13 , where the last four digits represent the binary representation of the integer 12, and the first digit is the integer rounded to the nearest integer from 0.22. Therefore, when assuming, for example, a measurement resolution of 1000 points per 100 nm optical bandwidth (i.e., a resolution of 0.1 nm), a key of 5 kB length per spectrum can be obtained.

[0116] Alternatively, the measured spectrum can be divided into regions indicated by the dotted lines in the spectra 518a and 518b, and based on the average value within each region, the binary numbers "1" and "0" can be determined by, for example, comparing the average value with a specific threshold or rounding the fraction of the average value relative to the total average value of all regions to the nearest integer and converting it into a binary number. Therefore, each user 512, 514 can obtain the following information: Figure 5 A sequence of binary digits as shown in , for example "1011".

[0117] about Figures 6a to 6e , an optical encryption terminal according to some illustrative embodiments of the present disclosure will be described, and measurements performed by the inventors for the described optical encryption terminal will be introduced and discussed.

[0118] Figure 6a An optical encryption terminal 600 according to some illustrative embodiments of the present disclosure is schematically shown. The optical encryption terminal 600 may include an optical pad 610 having an optical encryption pattern 620 formed therein, a plurality of input ports 630 (e.g., four input ports 632, 634, 636, and 638), and an output port 640 (the number of input ports and output ports is not limited, and those skilled in the art will appreciate that at least one input port and at least one output port may be provided).

[0119] According to some illustrative embodiments of the present disclosure, the optical pad 610 may be implemented as a so-called "fingerprint chip," in which the optical encryption pattern 620 may be implemented as an image of a user's fingerprint encoded into a crystal structure. According to some illustrative examples herein, the optical pad 610 may be a photonic nanostructure, in which the optical encryption pattern 620 is implemented as, for example, a series of circular holes patterned in the optical pad 610, having a shape that conforms to the user's fingerprint. For example, the digital image of the user's fingerprint may be transferred into the photonic structure by techniques known in the art of photonic nanostructure fabrication, including but not limited to electron beam lithography, nanoimprinting, and optical lithography. Figure 6a The fingerprint chip of the optical encryption terminal 600 shown can allow the user to be provided with an individually customized encryption terminal that is impossible to clone to a sufficient degree of accuracy. However, the implementation of the optical encryption pattern 620 as an image of a fingerprint does not impose any limitations on the present disclosure, and those skilled in the art will understand that any other pattern can be used to implement the optical encryption pattern, such as any pattern uniquely associated with a user, or any arbitrary pattern that is not uniquely linked to a user, such as a random pattern. The user's individual pattern can be generated from the user's physical features, such as a footprint of the user's foot, a print of the user's toes, a print of the user's retina, or any other image that the user wishes the optical encryption pattern 620 to be based on.

[0120] In the case of a cryptographic key distribution system (e.g., as described above with respect to Figures 1 to 5 When the optical encryption terminal 600 in any one of the cryptographic key distribution systems described above is used, the optical encryption terminal 600 can be coupled to an electromagnetic radiation source ( Figure 6a Not shown) and the detection element ( Figure 6a On the output side, the optical encryption terminal 600 can be connected to the optical encryption terminal 600 by optically coupling another optical encryption terminal (not shown) to the Figure 6a The output port 640 in is optically coupled to another optical encryption terminal (not shown).

[0121] about Figure 6b , a test setup for measuring the optical encryption terminal 600 as used by the inventors in an actual test setup will be described.

[0122] In tests performed by the inventors, the optical encryption terminal 600 is coupled to an electromagnetic radiation source 656a and a detection element 652a at an input side 650a, wherein the detection element 652a is coupled to an electronic processing unit 654a for outputting a power distribution spectrum. The source 656a and the detection element 652a are selectively coupled to an input port of the plurality of input ports 630. On the output side 650b, the source 656b, the detection element 652b, and the electronic processing unit 654b are optically and electronically coupled to an output port 640 of the optical encryption terminal 600.

[0123] about Figure 6c , measured at the output side 650b according to Figure 6c The power distribution spectra obtained by the four modes indicated by a, b, c and d, in particular as obtained by the electronic processing unit 654b.

[0124] about Figure 6d , according to Figure 6d 650a and 650b. In particular, test configurations a and e represent mutual exchange between input side 650a and output side 650b using the same input / output channel. Similarly, configurations b and f, c and g, and d and h are test configurations in which optical signals are exchanged via corresponding input / output channels. Figure 6c and 6d Comparison between reveals that the power distribution spectra corresponding to test configurations a and e match, while the spectra of configurations b and f match, while the spectra of configurations c and g match, and in the case of d and h the spectra also match.

[0125] about Figure 6e, a 2×2 signal exchange arrangement between configurations a, b, c, and d relative to configurations e, f, g, and h is plotted, showing the signal strength from one configuration relative to the others. In particular, the configurations along the diagonal of the two-dimensional arrangement are nearly identical within a few percent of signal strength. In particular, Figure 6b When the input and output sides of the test arrangement in FIG. 6 adopt matched input / output configurations, namely a and e, b and f, c and g, and d and h, almost the same power distribution spectrum can be measured on each side of the optical encryption terminal 600.

[0126] about Figure 7 , another illustrative embodiment of the present disclosure will be described with respect to implementing an optical encryption pattern as a non-permanent optical encryption pattern.

[0127] refer to Figure 7 , schematically illustrating an optical encryption terminal 700 during operation, the optical encryption terminal 700 having an optical encryption pattern 701 formed by resting a user's finger 703 on an optically transparent medium, such as a prism 710. An electromagnetic radiation source 720, such as a light emitting diode, a laser diode, etc., can emit electromagnetic radiation to the medium 710, wherein in particular, as long as the finger 703 rests on the medium 710, the electromagnetic radiation is reflected and scattered according to the optical encryption pattern 701 temporarily formed on the medium 710. The reflected light encoding the scattered and diffraction pattern of the optical encryption pattern 701 is guided to an output optical device 730, which is used to generate an optical encryption pattern 701 as described above with respect to a cryptographic key distribution system (e.g., as described with respect to Figure 1 The optical encryption terminal 700 is optically coupled to another optical encryption terminal (not shown) in the manner described in the cryptographic key distribution system 100 described. In this way, each time the user places his finger on the medium 710, a unique optical encryption pattern 701 can be formed, in particular in a non-repeating and non-clonable manner. In this way, secure generation and distribution of cryptographic keys can be provided.

[0128] According to some illustrative embodiments of the present disclosure, the optical encryption terminal can be implemented as an integrated fingerprint chip. Those skilled in the art will understand that the integrated fingerprint chip represents an illustrative implementation of a physical one-time pad, which allows the creation and distribution of cryptographic keys in a cryptographic key distribution system using two integrated fingerprint chips of two communicating users. Those skilled in the art will understand that the integrated fingerprint chip utilizes complex random scattering of light and creates a unique physical key that is impossible to copy or clone. The security of exchanging keys in a cryptographic key distribution system using such an integrated fingerprint chip is not based on statistical parameters, but relies on the fact that it is impossible for a third person to access the cryptographic key distribution system created by two different fingerprint chips of two communicating users. In the experiment, the inventor measured that even if two fingerprint chips manufactured under the same experimental conditions were used, two fingerprint chips with slightly different characteristics were produced, which resulted in the generation of completely different and unrelated cryptographic keys. Despite the use of two fingerprint chips with the same experimental conditions, defects were still introduced into the fingerprint chip, resulting in unique characteristics of each fingerprint chip, which ensured the uniqueness of the key generated by the system of two specific fingerprint chips.

[0129] With respect to FIG8 , a situation is schematically shown: the inventors use two fingerprint chips communicating through channels A2 and B2 to generate and distribute the key used by user “Alice” to encrypt picture “A”. Only when the second user “Bob” uses the key measured by matching channels A with B1, the key for successfully decrypting the ciphertext is obtained (refer to Figure 8c ).about Figure 8d , shows an unsuccessful attempt by an eavesdropper "Eve" to intercept the communication between "Alice" and "Bob". Thus, secure communication between "Alice" and "Bob" is achieved.

[0130] In the following, reference will be made to Fig. 9 and FIG. 10 depicts a man-in-the-middle attack performed by an eavesdropper (eg, “Eve” in FIG. 8 ).

[0131] Fig. 9A block representation of a communication system 900 between "Alice" and "Bob" (e.g., "Alice" and "Bob" in FIG. 8 ) is shown in the illustrative case where Alice's optical chip 910A and Bob's optical chip 910B are connected to a multi-channel communication network 920, such as a standard multi-channel communication network. According to some illustrative and non-limiting examples herein, each channel (e.g., a mode of optical fiber) of the multi-channel communication network 920 can be capable of carrying information provided to the communication network 920 by at least one independent source (e.g., at least Alice and Bob). For the present discussion of man-in-the-middle attacks, each block of the communication system 900 is represented by a transfer function, which can be understood as indicating a response to an input signal. In particular, each of the optical chips 910A and 910B is represented by a corresponding transfer function H. a (ω) and H b (ω) represents, and each channel of the multi-channel communication network 920 is represented by a transfer function C na (ω) and C nb (ω) represents, which can also be understood to represent the frequency-dependent input-output relationship for signals entering and exiting the corresponding optical chips 910A, 910B. Fig. 9 In the source is represented by the power density spectrum |S(ω)|^2, and in Fig. 9 Medium alice , P bob , P a and P b represent the power density spectra measured by Alice, Bob and Eve respectively.

[0132] During each communication between Alice and Bob, the source signal |S(ω)|^2 is first modified by Alice and Bob's optical chips. The emerging signal is then coupled to different channels and transmitted to the other side, where it is retrieved. The power density spectrum measured by Alice and Bob is:

[0133]

[0134] where |C(ω)| 2 is the contribution of the entire transmission system, which can be expressed as follows:

[0135]

[0136] Among them C na (ω) and C nb (ω) is the transfer function of each channel connecting Eve to Alice and Bob. In contrast, the power density spectrum measured on the Eve side is as follows:

[0137]

[0138] where x = a, b represents the signal measured by Eve when the communication is initiated by (a) Alice or (b) Bob, respectively. By comparing equation (3) with equations (1)-(2), we immediately observe that due to the incomplete spectral information on the transmission channel (Eve does not have the possibility to determine the entire transmission line C(ω) when measuring), so EVE cannot reproduce the power density spectrum measured by Alice / Bob. In fact, when Eve does the product of the acquired spectrum P e (ω)=P a (ω)·P b (ω), the information about the transmission system retrieved by Eve is:

[0139]

[0140] It does not represent the transmission line contribution |C(ω)| measured by Alice and Bob. 2 To reconstruct the function C(ω), Eve needs to independently measure and store the signal traveling on all possible channels and extract the transfer function C of each channel. nx (ω) and indirectly reconstruct the channel transfer function C(ω) from (2). However, even with recent advances in communication data analysis, this type of measure is impossible: in a standard optical communication network, there are millions of single-mode fibers, and the user does not know a priori in which channel the traffic is routed. Near the user, there is usually a single line, and the information is usually multiplexed in a multimode fiber with thousands of modes; and it is impossible to store the content of the information on each channel in real time. In a wireless network or a free-space link, there are an infinite number of channels and paths that electromagnetic waves can travel; monitoring all of them is impossible.

[0141] It should be noted that even under the assumption that this type of measurement is possible, reconstruction of the information exchanged between Alice and Bob is still impossible. In the case where there is a single optical fiber connecting Alice and Bob and the fiber is single-mode, the product P of the power density spectrum measured by Eve is e (w) = P a (w)·P b (w) is as follows:

[0142]

[0143] By comparing equation (5) with equations (1)-(2), we immediately observe that in the case of n = 1 channel, Eve's measurement does not reproduce the signal exchanged between Alice and Bob, due to the power density spectrum of the source (where Ps =|S(ω)| 2 The power density spectrum of Alice’s source is defined, which appears twice in all Eve’s measurements due to the bidirectional nature of our communication scheme. The source belongs only to Alice and Bob and Eve cannot access it. However, even if Eve has physical access to the source, she still cannot measure P s .Replenish Fig. 9 The quantity P in the block diagram s is in fact an ideal quantity that represents the isolated density spectrum of the source when not connected to any instrument. It is well known that measuring this quantity is impossible: as soon as Eve connects the source to an instrument, in fact, she breaks the isolated nature of the source and introduces into the measurement a spectrally varying coupling coefficient that is not known a priori and affects all results. This problem is the same as measuring the ideal temperature of a body. As soon as the thermometer comes into contact with the body, the temperature of this system changes and the thermometer reads only the equilibrium temperature between the body and the thermometer, not the true temperature inside the body. This type of uncertainty is embedded in the intrinsic nature of our measurements and cannot be avoided.

[0144] To quantitatively illustrate this for the communication scheme described above with respect to the illustrative embodiment of the present disclosure, the inventors conducted an experiment, the results of which are shown in FIG. 10 .

[0145] Fig.10a A graph illustrating the power density spectra of the sources used by Alice and Bob, measured using the same physical device used by Alice and Bob in the key distribution scheme of the present disclosure is shown. The sources employed by Alice and Bob may have substantially the same power density spectra. The inventors repeated the measurements at different times and observed that the measurements were stable.

[0146] Fig.10b A representative power density spectrum measured on the Alice side is shown in logarithmic scale and corresponds to input condition A2B1. In contrast to the source, the power density spectrum varies in different decades, showing a rather complex oscillation pattern.

[0147] Fig.10c The logarithmic scale is reported, and Eve attempts to normalize P eva =P e / P s The relative difference between the spectra measured by Alice and Bob in order to reconstruct Alice's spectrum (in Fig.10c Indicated as Δ alice-bob ) and the relative difference between Alice and Eve (in Fig.10c Indicated as Δ alice-eve1). The spectra are completely different by more than an order of magnitude. This stems from the fact that what Eve measures is only a representation of the spectrum, and the coupling coefficient between the source and the instrument is unknown.

[0148] However, even in the case where the spectra of Alice and Eve are normalized at their maximum values ​​(in Fig.10c Indicated as Δ alice-eve2 ), their shapes remain different in most of them. This stems from the fact that: s The uncertainty, even if small, in the spectrum is 1 / P s It also amplifies exponentially when normalized, especially at points with small amplitudes. It is well known that these types of normalization cannot be used in practice for large bandwidth signals, since the resulting reconstruction is very noisy, especially for strongly oscillating spectra such as Fig. 9 The spectra of the optical chips 910A and 910B in are exchanged between Alice and Bob.

[0149] Replenish Fig.10d The influence of the spectrum differences between Alice, Bob and Eve on the generated key is analyzed. Fig.10b The spectral generated key is displayed as Fig.10d The square matrix of zeros (black dots) and ones (white dots) in the upper panel of the graph. As can be visually observed, the generated bits are uncorrelated. The inventors calculated the statistics using 250 generated keys and obtained that the Hamming distance (i.e., the percentage of different bits) between the keys generated by Alice and Bob is only 5%, while in the case of Eve (also in the case of rescaling the spectrum to the same value), it becomes about 50%, which means the maximum entropy condition, where half of the bits are randomly inverted.

[0150] After the teaching of the present disclosure, it will be understood by those skilled in the art that any cryptographic key distribution system of the present disclosure may be implemented in a communication system for providing communication between a third-party device and a controller employing the cryptographic key distribution system of the present disclosure. According to some illustrative and non-limiting examples, at least one of the first and second electromagnetic radiation sources may be provided as an optoelectronic converter or modulator integrated in the third-party device. In some particular illustrative examples, the communication system may be embedded or implemented in a network of physical devices, vehicles, household appliances, and other items embedded with electronic devices, software, sensors, actuators, and connectivity, which enables these objects to connect and exchange data (the so-called "Internet of Things" or IoT), where each thing is uniquely identifiable through its embedded computing system, but is able to interoperate within the existing Internet infrastructure.

[0151] According to some illustrative embodiments of the present disclosure, the cryptographic key distribution (CKD) system may be applied to a communication system, such as an optical communication system. In an illustrative but non-limiting example herein, the cryptographic key distribution system may be integrated into an optical fiber communication system.

[0152] According to some illustrative embodiments herein, a keying-based digital modulator, such as frequency shift keying (FSK), amplitude shift keying (ASK), quadrature amplitude modulation (QAM), and / or any type of phase shift keying (PSK), can be integrated at the input side and / or output side of at least one encryption terminal of the described cryptographic distribution system. For example, a digital-to-analog converter (DAC) device can be arranged to transform a transmission sequence sent to one of the encryption terminals into an analog waveform, thereby generating an electrical signal, which can then be optically modulated using a modulator and transmitted through an analog optical fiber of an internal CKD. The CKD can transform the signal into an indestructible optical image, which can then be optically converted and further digitally converted using a standard analog-to-digital converter (ADC) device or a differential pulse code modulator system, which can then be reintegrated into the communication system.

[0153] Hereinafter, as an appendix to the foregoing disclosure, the present invention will be described in alternative ways and / or for alternative embodiments. After reading this disclosure in its entirety, those skilled in the art will understand how the following relates to the above-described embodiments, and how the following alternative embodiments are considered equivalent to the above-described embodiments. In order to better understand the following discussion, some background of the present invention is presented in a different manner than the above description, so as to allow a thorough understanding of the present invention and a thorough understanding of the manner in which a skilled person can implement the present invention.

[0154] Now, it will be explained how perfect confidentiality cryptography can be achieved through the correlated mixing of chaotic waves in an irreversible, time-varying silicon chip.

[0155] Preface

[0156] Protecting confidential data is a major challenge of the Internet age. Standard encryption techniques are fast and scalable, but they are broken by quantum algorithms. Quantum cryptography is unclonable and more powerful, but requires quantum equipment that is more expensive, slower, and less scalable than classical optical networks.

[0157] A system is provided that provides a physical implementation of an alternative embodiment described in more detail below to the Vernam cipher by utilizing correlated chaotic wave packets that are mixed in inexpensive and CMOS-compatible silicon chips that are irreversibly modified before and after each communication. For example, each chip may contain a biometric fingerprint of a user of the system and may therefore be different for each user. Those skilled in the art will appreciate that these chips may have the capacity to generate 0.1 Tbit of different keys for each communication and for each mm of length of the input channel. The keys generated according to the protocol of the alternative embodiment described below may require the transmission of an amount of data that may be approximately 1000 times smaller than the length of the message and may be generated at the remote end of each user without being visible in the communication line.

[0158] As explained in more detail below, it will be demonstrated theoretically and experimentally by work performed by the inventors that when the chip is changed, the key cannot be recreated again, not even by the user.

[0159] The discussion of the security of the present protocol, as shown below, follows Kerckhoffs' principle in the case where an ideal attacker with unlimited technical power, control of the communication channels and access to the system before or after the communication copies any part of it, including the chip. After reading this disclosure in its entirety, a person skilled in the art will understand that the second law of thermodynamics and the exponential sensitivity of chaos unconditionally protect the scheme from any attack. Theory and measurements performed by the inventors using classical telecommunication fibers, as presented in this context, show that an attacker of this system is always in a maximum entropy scenario (uncertainty higher than 0.99 bit per bit): the adversary can only guess the messages exchanged by the users by best guesses without intercepting their communications.

[0160] background

[0161] As the information society transmits more and more personal data through public channels, information security is an emerging global challenge. Traditional cryptographic schemes based on the Data Encryption Standard (DES), the Advanced Encryption Standard (AES), and Rivest, Shamir, and Adleman (RSA) utilize short-length public and private keys to encode messages. The main advantage of these algorithms is their speed, and the main disadvantage is their security, which relies on computational and provable security parameters rather than unconditional proofs. The main threat lies in the development of quantum computers, which are expected to crack any of these ciphers in a short period of time.

[0162] A perfect secrecy cryptography called the One-Time Pad (OTP) was invented during the telegraph era and then patented by Vernam. The Vernam cipher encodes a message through a bitwise XOR operation using a random key that is as long as the text to be transmitted, can never be reused in whole or in part, and remains secret. Shannon proved that the scheme, when properly implemented, is unbreakable, providing no information to the attacker except the maximum length of the message. Almost a century later, despite its proven absolute security, OTP has not been adopted in the digital realm due to the lack of a practical and secure way for users to exchange keys. Since the 1980s, research efforts have been devoted to solving this problem with point-to-point quantum key distribution (QKD) algorithms that exploit the non-cloning property of individual photons. While the progress of QKD over the past few decades has been tremendous, there are still critical challenges posed by the limitations of quantum communications. Quantum networks are currently unable to scale globally due to the impossibility of amplifying individual photons; their data transmission is much slower than classical optical communications, which already utilize hundreds of high-bandwidth intercontinental lines with communication speeds close to the optical limit and have undergone massive investments in the coming years.

[0163] Brief Overview

[0164] After a complete reading of the present disclosure, those skilled in the art will appreciate that the inventors have developed a physical implementation of OTP that is compatible with existing optical communication infrastructure and provides unconditional security in key distribution.

[0165] Detailed description

[0166] For a better understanding, the Vernam code on a classical channel will be explained. It is well known that chaos produces time-varying signals that are mathematically unpredictable for a long time. This stems from the sensitivity to different input conditions: even in the limit of ε→O, two nearby input states x(t=0) and x′(0)=x(0)+ε always generate a signal with an exponential divergence distance Δ(t)=|x(t)-x in time. l (t)|~e μt of the trajectories, μ is the maximum Lyapunov coefficient of the flow.

[0167] By exploiting this property, the inventors show that it is possible to create a bidirectional communication channel in which users mix independent time-varying chaotic processes to securely exchange random keys of arbitrary length.

[0168] about Fig.11 , shows a protocol scheme for perfect confidentiality key generation over classical channels. Fig.11(a) shows a communication setup with users Alice and Bob and an attacker Eve, where Alice and Bob participate in a cryptographic key distribution system, such as described above with respect to Figure 1 The cryptographic key distribution system described herein is incorporated by reference. Fig.11 (b) shows the communication and key generation steps, where Eve does not have an additional state E x When actively interfering with the channel, Alice and Bob send different chaotic states A n , B n′ , and always measure the associated mixed chaotic state. Fig.11 In (c), the encryption and decryption scheme is schematically shown.

[0169] exist Fig.11 In the system shown in (a), each of the two users Alice and Bob has an optical encryption terminal for generating and distributing cryptographic key signals in a cryptographic key distribution system having at least two optical encryption terminals, the optical encryption terminal includes an electronic processing unit, wherein the optical encryption terminal is configured to selectively receive a first optical input signal generated by an electromagnetic radiation source and a second optical input signal generated by another optical encryption terminal, selectively output a first optical output signal to a detection element and output a second optical output signal to another optical encryption terminal, the first optical output signal is based on the second optical input signal transformed according to an optical encryption pattern set at the optical encryption terminal; and use the electronic processing unit to determine the cryptographic key signal based on at least one radiation amount and / or light amount associated with the first optical output signal detected by the detection element. According to some specific illustrative examples herein, the optical encryption pattern may include a plurality of regions of a first material randomly arranged in a second material, wherein the first material and the second material differ from each other in at least one optical property. For example, the first material may have greater absorption and / or transmission and / or diffraction properties in terms of electromagnetic radiation than the second material. Additionally or alternatively, the plurality of regions of the first material may be arranged according to a random and / or chaotic and / or predefined scattering pattern.

[0170] According to the specific illustrative and non-limiting examples herein, the optical encryption pattern can be implemented by a fingerprint chip. For example, Alice and Bob can have two chips that generate chaotic light states transmitted on a common classical optical channel. Each light state, indicated as A for Alice n And for Bob, the instruction is B n′ , is a random superposition of light waves at different frequencies:

[0171]

[0172] The above equation includes the uncorrelated random amplitude a nm , b n′m and Phase ψ n′m The state represented by equation (6) is composed of the emission of different frequencies ω1, ω...ω inside each chip. m The chaotic scattering of identical broadband pulses with different input conditions n and n' (position, angle, polarization, time modulation, ...) arbitrarily chosen by Alice and Bob is generated. The pulses do not need to be identical, their differences constitute the main source of uncertainty and set the desired communication bit error rate (BER) in the communication. The scattering system consists of a large number of point scatterers.

[0173] For the purpose of this discussion, assume that the chip meets the following conditions:

[0174] 1. The scattering process inside each chip is completely chaotic, and any emission conditions of light into the scatterer follow chaotic dynamics.

[0175] 2. Alice and Bob's chips are in thermodynamic equilibrium with the environment, and there is no obvious structural change for light propagation during communication in each state.

[0176] 3. Any modification to the distribution of scatterers within the chip results in a new chaotic system that has an exponentially divergent trajectory relative to the previous one.

[0177] 4. The chip is structurally modified in time before and after each communication through two physically irreversible processes (e.g., deformation, adding scatterers, ...) independently chosen and applied by Alice and Bob, creating a new chaotic scattering system in which all trajectories are exponentially different from the previous one.

[0178] The use of static optical scatterers in information security provides computational and provable security in both authentication problems and cryptographic key generation, offering advantages over electronic schemes in terms of bulk physical data storage relative to standard electronic databases.

[0179] However, the security of schemes based on complex scattering structures relies on i) the condition that the physical scattering objects are kept secret from the adversary, and ii) the assumption that these structures are infeasible to cloning.

[0180] However, these arguments do not provide unconditional security. While it has been demonstrated in recent work that it is indeed possible to experimentally clone physically unclonable functions, perfect secrecy requires proving the security of the system in the limit that an adversary has access to the system before or after the communication and has copied parts of the system (although such an operation may be very difficult). In such a case, all existing schemes based on classical scatterers suffer from the same security problems as their electronic counterparts and are caused by the lack of perfect secrecy.

[0181] Those skilled in the art will appreciate that the inventors considered the limitation of perfect secrecy without assuming conditions i) and ii). However, the inventors believe that an adversary has the technology to clone any type of scattering structure.

[0182] Another element of novelty with respect to classical schemes based on physically unclonable functions is the lack of a hard requirement to perform a first encounter (or equivalently secure communication) between users, in addition to the authentication protocol. This situation makes current classical schemes inferior to QKD, which does not require any initial setup beyond user identity authentication. The system presented here, by using a properly defined time-varying complex system, does not require any initial secure communication (besides identity authentication), thus providing a classical alternative to QKD. At each step of the communication, Alice and Bob randomly select input conditions n and n′, respectively, emit optical signals in the system, and transmit them as Fig.11 (b) shows how Alice and Bob measure the output at their end. After the measurement, Alice and Bob choose (with random probability) to keep the emission condition or to change it. The process is then repeated. Due to the reciprocity of the communication network connecting Alice and Bob, Alice and Bob measure the same optical observables (intensity, power density spectrum, ...) if Eve does not perform active manipulation of the signal. For example, when Alice sends a chaotic wave packet An to Bob, he measures an optical observable, such as the combined light state The strength of the association ( is the operator that combines the states after propagation over the channel.) In this configuration, when Alice measures the output at her end, she measures the mutual state have different phases, but the same amplitude and the same optical observable value

[0183] At the end of the chosen sequence, Alice and Bob record all instances of the acquired data being unchanged and communicate this information over a common line, which allows the OTP key to be extracted through the sequence in an overlapping, repeating sequence. The key is generated by converting the intensity or power density spectrum of the exchange into a binary sequence using the Adaptive High Boost (AHB) technique.

[0184] Using this protocol, for length L m The amount of data exchanged to generate an OTP key is N b is the combined state from each transfer The exchange key is then used at each user end to encode and decode the data using the Vernam cipher by bit-by-bit XOR ( Figure 1 c).

[0185] Fig.11 The communication protocol described in can be viewed as a classical version of the original BB84 QKD (Quantum Key Distribution) scheme developed by Bennett and Brassard, where the scatterers act as generators of random states and the intercommunication lines provide the correlated measurement states.

[0186] In a variation of classical schemes based on complex scatterers, the protocol presented here does not require a first encounter or initial secure communication between users (except for authentication), thus providing a classical alternative to QKD.

[0187] In the quantum limit, when a user (let's say Bob) emits a single photon in the chip, the receiver (Alice) measures the photon that emerges from the chip at a random location. If Alice injects the photon back into the same scattering channel, the reciprocity theorem of quantum mechanics guarantees that Bob measures the newly emerged photon in the same input channel he originally used. The process has some similarities to the quantum BB84 scheme, with the scattering channels acting as random polarization states. However, there are differences. In order for users to exchange the same sequence of bits, they can initially agree to use a common dictionary that associates the same bit string with relevant input-output locations in Alice's and Bob's chips. This operation is not required in the classical limit (because the users measure the result of a large set of photons on all channels) and in the BB84 scheme.

[0188] Regarding the perfect secrecy of the system, the inventors provided the following explanation: Shannon proved that the Vernam cipher has perfect secrecy properties if i) the keys exchanged are as long as the messages, ii) each key is used only once and is independent of new keys, and iii) the keys are known only to the users. Fig.11 The scheme of (a) exchanges keys of arbitrary length over classical optical channels at full speed. Therefore, it provides a feasible implementation of the first requirement.

[0189] The covariance matrix Knn′=(AnAn′) of the correlation between Alice chaotic wave packets is a delta function, where (...) represents the average in amplitude and phase:

[0190]

[0191] where σ 2 n =K nn Equation (7) yields the fact that it is not related to amplitude, for n≠n′ and m≠m′, (a nm a′ m′ )=0. Keeping the same conditions at Bob’s end, B nn ′=(B n B n ′)=X 2 δ nn ′.

[0192] This means that the states of Alice and Bob are independent of each other and from the combined state The generated keys are also not relevant. Fig.11 The protocol designed in meets the second requirement.

[0193] Regarding the perfect confidentiality of the key distribution scheme, the inventors explain as follows, considering the ideal case (Kerckhoffs' principle) where the system falls into the hands of an adversary who knows all the details of the encryption / decryption process and has access to the ciphertext. In this case, the only unknowns are the user's key and the input conditions (including any chosen transformations).

[0194] Since the system is classical, Eve can store all the signals emitted by Alice and Bob during the communication, and then she can try to search for the input conditions on each user's chip that produced the state she measured. Once Eve knows the input conditions, she can emit the same state and try to recreate the key.

[0195] However, the second law of thermodynamics prevents this attack. Every time Alice and Bob change the chip using irreversible physical processes, they increase the total entropy of the system and the environment, creating new chaotic structures that are exponentially different from those used in the communication (see points 3-4 above). If Eve accesses the system, she cannot recreate the initial chip and perform any search, because this would require reversing Alice and Bob's transformations in a way that reduces entropy, thus violating the second law.

[0196] Another possibility for Eve is to make an identical copy of the system in all parts of the system and try to search for it the next time they communicate. This task requires Eve to generate chaotic scatterers identical to those of Alice and Bob so that their transformations can be cloned by Eve before communication. Since the chip is not an isolated system and is in equilibrium with the environment (see point 2 above), this task requires replicating the surroundings of Alice and Bob's chips. This condition is essential to enable Eve's Eve copy chip to reach the same equilibrium state as the original chip of Alice and Bob's chips. The second principle of thermodynamics makes this condition impossible. In fact, Eve does not know and cannot replicate the exact time when Alice and Bob performed their transformations. If Eve performs the transformation after Alice or Bob, the environment will be different because it has at least one transformation that is irreversible in time (one of Alice or Bob), which increases its entropy, and vice versa if Eve performs the transformation before the user. Therefore, it is impossible for Eve to clone the same transformation of Alice and Bob. Because of points 3-4 above, Eve will generate new chaotic scatterers that are exponentially different from the ones Alice or Bob are using, and are therefore useless.

[0197] The only possible search is a brute force attack on the key. However, due to the perfect secrecy of the Vemam cipher, this attack does not generate any information. This only gives the attacker one possibility: to extract the key from the information available on the system.

[0198] In any possible attack, the data available to the attacker is the chaotic wave packet A flowing on the communication line. n and B n’ related observables. Eve can measure these states, and from the results of the measurements, can attempt to reconstruct Alice and Bob's secret key. In the following analysis, we show that the result is always a key where each bit has a 50% chance of being correct and a 50% chance of being wrong, regardless of the type of attack employed. This implies perfect confidentiality: the posterior probability that Eve's key represents Alice and Bob's secret key is exactly the same as the prior probability of guessing Alice and Bob's secret key before interception.

[0199] exist Fig.11 (a) In the optical communication scheme, the chaotic state A nThe only experimentally observable quantities related to Bn' are non-instantaneous quantities of frequency or time, such as the power density spectrum or intensity, while instantaneous values ​​of amplitude and phase are not observable. This limitation also applies to interferometric detection and time gating, which require either a periodic signal in time or precise knowledge of the pulse properties in advance. In order to measure a randomly generated chaotic wave packet A that is never replicated and whose shape is unknown, n and B n’ The only possibility is to accelerate the electron at relativistic speeds to follow the dynamical properties of the photon, but this is impossible because it requires an infinite amount of energy.

[0200] refer to Fig.11 A protocol scheme, a person skilled in the art will understand that, generally, a method for generating a cryptographic key in a cryptographic key distribution system having a first optical encryption terminal and a second optical encryption terminal is implemented, the method comprising: providing electromagnetic radiation of a first electromagnetic radiation source as a first optical input signal to the first optical encryption terminal, transforming the first optical input signal into a first optical output signal according to a first optical encryption pattern formed at the first optical encryption terminal, providing the first optical output signal as a second optical input signal to the second optical encryption terminal, transforming the second optical input signal into a second optical output signal according to a second optical encryption pattern formed at the second optical encryption terminal, detecting at least one first radiation amount and / or light amount associated with the second optical output signal, and determining the first cryptographic key based on the detected at least one first radiation amount and / or light amount. According to some other illustrative embodiments of the present invention, the first and second optical encryption patterns can be formed with photonic nanostructures, biomaterials, or by directly using a human interface (e.g., a human fingerprint or retina).

[0201] In some embodiments, the method may further include: providing electromagnetic radiation of a second electromagnetic radiation source as a third optical input signal to the second optical encryption terminal, transforming the third optical input signal into a third optical output signal according to the second optical encryption pattern, providing the third optical output signal as a fourth optical input signal to the first optical encryption terminal, transforming the fourth optical input signal into a fourth optical output signal according to the first optical encryption pattern, detecting at least one second radiation amount and / or light amount associated with the fourth optical output signal, and determining a second cryptographic key based on the detected at least one second light amount. For example, the first and second cryptographic keys are the same. According to some demonstrative embodiments, the first and / or second optical encryption patterns may be modified before and after each communication, wherein the first and / or second optical encryption patterns may be modified by a reversible or irreversible mechanical deformation applied to the first and / or second optical encryption patterns, or a drop of a liquid (e.g., water or any other liquid containing a colloidal deposit as an additional scatterer) may be dropped onto at least one of the first and second optical encryption patterns for temporarily or permanently changing the optical properties of at least one of the first and second optical encryption patterns. In this way, in Fig.11 In the scheme of (a), at least one of Alice and Bob can temporarily or permanently generate a new chaotic scattering system.

[0202] Regarding security against time domain attacks, the inventors analyzed the limitations of Eve to develop a technique to access the instantaneous value of the intensity. In this context, the inventors considered a scenario independent of the source and channel used, which is set to a mathematical Dirac delta δ(t) with a constant spectrum.

[0203] In this case, the strength of the combined state com state is in, is the convolution operator. At each step of the communication, Eve can measure and store the strength of the transmitted state |A n (t)| 2 and |B n ′(t)| 2 , then try to pass through Freely combine states to reconstruct mixed states in is the selected operator. We consider all possible (sum, product, convolution) case. The result of the attack is quantified by the average Shannon information entropy contained in each bit measured by Eve, and is calculated according to the average Shannon information entropy per bit H = -dlog2d-(1-d)log2(1-d), where d is the difference in bits between the key of a user (Alice) and the key reconstructed by Eve. The Shannon entropy H quantifies Eve's uncertainty about each bit measured. When d = 0 (Eve measures the same as Alice) or d = 1 (Eve measures the opposite of Alice), Eve's information entropy is zero because Eve predicts the key exchanged by Alice and Bob without a doubt. In other cases, H is a positive function with a maximum value of H = 1. Under this condition, Eve has 1 bit of uncertainty for each bit measured and zero information about the key.

[0204] refer to Fig.11 , will be used in cryptographic key distribution systems (e.g., the above Figure 1 Here, the key may be generated before, after, or during each step of the communication between Alice and Bob (see Fig.11 (a)). For example, at least one of Alice and Bob may randomly select input conditions n and n!, respectively.

[0205] According to some illustrative embodiments of the present invention, at least one of Alice's and Bob's optical encryption terminals may have at least two input terminals configured to receive optical input signals generated by an electromagnetic radiation source, and one of the at least two input terminals may be selected, and the optical input signal may be provided to the selected optical input terminal.

[0206] After the measurement, Alice and / or Bob can choose (with random probability) to maintain the transmission condition or change it. The process can then be repeated.

[0207] According to some illustrative embodiments of the present invention, in response to measuring an optical signal according to an optical input signal input via a selected optical input terminal, the selection of the input terminal and the measurement of the optical output signal can be repeated until all input terminals are selected. All instances of the measured optical signal can be recorded, and the recorded instances can be provided at each of the first and second optical encryption terminals. Therefore, on both sides of the cryptographic key distribution system, all recorded instances are available, and each of Alice and Bob can be able to extract at least one optical signal of the measured optical signal that is the same at the first and second optical encryption terminals. That is, when each of Alice and BoB measures an optical observable value (e.g., intensity, power density spectrum, etc.) as an optical signal at their respective ends in response to an input signal, a mutual state with different phases but the same amplitude and the same optical observable value is measured. Therefore, Alice and BoB record all instances of acquisition data corresponding to the measured optical signal, and data indicating that the measured optical signals match each other can be identified in the recorded instances, and can be extracted as the extracted optical signal of Alice and Bob.

[0208] Based on the selected input condition, by applying the input signal at the optical encryption terminal, the optical signal is sent into the system via the optical encryption terminals of Alice and Bob, and in response to the input signal through the two optical encryption terminals, the output signal is measured as the output at the optical encryption terminal as the output at the optical encryption terminal (see Fig.11 (b)). Based on the extracted optical signal, a key can be generated at Alice and Bob’s optical encryption terminals by converting the intensity of the exchanged power density spectrum into a binary sequence, for example, by using the Adaptive High Boost (AHB) technique.

[0209] Those skilled in the art will appreciate that, due to the reciprocity of the communication network connecting Alice and Bob, if Eve does not perform active manipulation of the signal, Alice and Bob measure the same optical observables (intensity, power density spectrum, ...). For example, when Alice sends a chaotic wave packet to Bob, he measures an optical observable, such as the intensity associated with the combined light state. In this configuration, when Alice measures the output at her end, she measures a reciprocal state with a different phase but the same amplitude and the same optical observable.

[0210] At the end of the chosen sequence, Alice and Bob record all instances of the acquired data being unchanged and communicate that information over a common line, which allows the OTP key to be extracted via the sequence in the overlapping, repeating sequence.

[0211] By using the Adaptive High Boost (AHB) technique, the key is generated by converting the intensity or power density spectrum of the exchange into a binary sequence.

[0212] Using this protocol, the amount of exchanged data required to generate the OTP key for a message of length Lm is Lm , Nb is the number of bits generated from each transmitted combined state.

[0213] The data is then encoded and decoded bit by bit XOR at each user end using the exchanged key according to the Vernam cipher (see Fig.11 (c)).

[0214] Fig.12 It is shown that when Alice and Bob mix as described by equation (6) and contain an increasing number of different frequencies ω1, ..., ω M The random wave packet of is the average uncertainty of Eve. Fig.12 In , the security of the protocol against time domain and spectral attacks is shown. Fig.12 (a) shows the uncertainty per bit of time-domain key reconstruction for all possible types of attempts measured by an ideal attacker as the user mixes an increasing number M of random wave packets of different frequencies: (red line), + (green line), · (blue line). Fig.12 (b)-(c) show the non-periodic case (see Fig.12 (b)) and the chaotic case (see Fig.12 (c) Statistics of the intensity of a single wave packet. Fig.12 (d)-(f) show the results of the spectral attack, where the uncertainty per bit of Eve is shown as a solid green line and the BER between Alice and Bob’s keys is shown as a solid red line for different standard deviations σAB between the power density spectra of the combined states measured by Alice and Bob, as a function of the number of bits N extracted for each spectral point measured. Fig.12 Graphs report mean values ​​(solid lines) and standard deviations (error bars).

[0215] The chaotic waves caused by chaotic scattering obey the general Gaussian statistics for intensity P(I) = αe -αl (For Alice I = A 2 n , for Bob I = B 2 n′ ), and it is in state A n and B n′ Complete randomization is achieved at sufficiently large M. At lower M, the wave packet A n and B n′It is a non-periodic superposition of waves with no general behavior.

[0216] In the range of small M, the statistics of the wave packets vary from case to case, and the uncertainty of Eve oscillates in fairly large intervals, indicating that on average an attacker can The results of their measurements are combined to infer the key. Using different operators, as intuitively expected, does not generate any information. As the number of frequencies M increases and the state A n and B n′ Each implementation shows the same general characteristics and variance of uncertainty collapse when it becomes a chaotic wave. In this range, Eve's uncertainty becomes unity (H ≥ 0.998 ± 0.01): the information accessible from the transmitted data is not enough to reconstruct the complex chaotic state formed at Alice and Bob's ends, and the system is unconditionally secure. Active attacks introduce deterministic errors in the communication sequence between Alice and Bob, without Eve's information. These errors are small and in the range of 1 / N, where N"1 is the number of chaotic states available in Alice and Bob's chips.

[0217] These errors can also be eliminated by using information reconciliation and privacy amplification, both of which are performed over a public authentication channel. With information reconciliation, Alice and Bob refine the same key on each user side by exchanging a minimal amount of information (e.g., bit-only parity of the block key sequence). The second phase, privacy amplification, is then applied to effectively eliminate the information acquired by Eve during the reconciliation step. Privacy amplification is typically performed by using a universal hash function that generates a new, shorter key about which Eve has zero information.

[0218] Regarding security against spectral attacks, the transfer function H(ω) of the system connecting Alice and Bob is expressed as follows:

[0219]

[0220] Here, and denote the transfer functions of Alice's and Bob's chips, respectively, C(ω) is the contribution of the transmission system, S(ω) is the spectrum of the input source, and α(ω) is the coupling coefficient of the spectrum analyzer, |α(ω)| ≤ 1 for energy conservation. For reciprocity, the coefficient α(ω) is the same for both users. Power density spectrum and is a random spectrum that changes at each transmission step n due to different input conditions chosen by each user. In a spectrum attack, Eve uses identical copies of Alice and Bob’s spectrum analyzers to measure the spectrum P transmitted on the communication line. Alice and PBob , and then try to pass P Alice P Bob / P S The measured combined state is reconstructed at the remote end, where P S is the spectrum of the source. As shown below, the wave packet A n The chaotic nature of and Bn' makes it theoretically impossible to reach a correlation value measured by Alice and Bob in their combined spectra that provides zero information about the secret key shared by Alice and Bob. The uncertainty contained in the mutual power density spectra of Alice and Bob is generated by the following uncorrelated components:

[0221] Source P S =|S(ω)| 2 Statistical fluctuations ΔS of the power density spectrum.

[0222] Noise thermal fluctuations ΔC in the communication channel.

[0223] Fluctuations Δ in the input conditions chosen by Alice and Bob respectively A , Δ B .

[0224] The statistical uncertainty Δ provided by the measurement device used M .

[0225] The coupling coefficient α(ω) of the spectrum analyzer, which is a systematic source of uncertainty that is not known a priori (type B according to NIST guidelines). To evaluate α(ω), the user must measure the absolute value of the spectrum and normalize it. To do this, the user must use another measuring instrument that introduces another coupling coefficient that the user does not know. This parameter can be iterated at infinity without the user knowing the first coupling coefficient. It is the same problem as measuring the exact temperature of a body: the temperature of the thermometer needs to be known in advance, because it reads the equilibrium temperature between the thermometer and the coupled environment of the body.

[0226] In the experimental implementation, the main statistical component of the uncertainty is the fluctuation of the source: Δ A,B,C,M <<Δ s , (as discussed quantitatively in the experimental implementation section). This is also the general case where the user can always choose, since the source is physically in hand and controlled by the user during communication. Within the acceptable bit error rate (BER) range, if necessary, Δ s It can always be increased by adding artificial time fluctuations in the communication.

[0227] Then the total standard uncertainty Δ when measuring the user spectrum PBob is expressed by the following formula:

[0228]

[0229] If a main middle attacker Eve wants to try to reconstruct the spectrum PBob, she needs to make at least three measurements:

[0230] 1. Spectrum P sent by Alice EA yes

[0231]

[0232] C A (ω) is the transfer function of the communication line connecting Alice to Eve. The ideal scenario for Eve is to use exact copies of Alice and Bob's analyzers so that they measure exactly the same α(ω).

[0233] 2. Spectrum P sent by Bob EB yes

[0234] P EB (w) = IS (w) H (n) B(w)·C B (w)·α(w)| 2 , (11)

[0235] Among them C B (ω) is the transfer function of the communication channel between Bob and Eve.

[0236] 3. The source spectrum is P I s =IS(w)α(w)I 2 , which is necessary because of the use of chaotic uncorrelated spectra, which does not allow the estimation of P from measurements 1-2 above S In this case, the attacker's best option is to use an identical copy of the user's spectrum analyzer. This will minimize any other statistical uncertainties introduced at this stage due to, for example, different sensors, different resolutions, and any different components used for the measurements.

[0237] Once measurements 1-3 above are in place, Eve can EA P EB / P EB / P I s Try P Bob reconstruction.

[0238] This operation returns an estimate expressed as:

[0239]

[0240] Among them C A C B= C, and the total uncertainty Δ estimated from each independent measurement performed by Eve l :

[0241]

[0242] Equations (12)-(13) show that in an ideal scenario, Eve's measurement contains: i) at least one systematic multiplicative uncertainty that Alice and Bob have not experienced; and ii) a statistical uncertainty that is at least √3 higher than the statistical uncertainty of either Bob or Alice. These results are theoretical limits for an ideal adversary with infinite technical capabilities and are within the Kerckhoffs limit. Although neither the user nor the attacker knows c(ω), the statistical uncertainty Δ S is the amount that cannot be removed by an adversary because the source is in the physical hands of the user at the time of communication.

[0243] The inventors analyzed the system safety in terms of theoretical limits in the worst case for the user, where for each considered frequency |α(ω)| 2 = 1. This limit is not actually achievable, since it implies a technique that could measure the state without disturbing it, and this would violate at least the projection assumption of quantum mechanics regarding the possible outcomes in the measurement of physical observables. However, the inventors will explain below that even in this case, the present system is unconditionally safe.

[0244] The inventors considered different communication scenarios where Alice and Bob measure related spectra with different statistical fluctuations and developed a multi-bit AHB transform inspired by the concept of random bit generation that optimizes the information extraction from the spectra acquired at each user end. Commercially available line detectors provide at least 16 bits for each spectral point: if we maximize the use of this information in the generated key, we can significantly reduce the communication workload between Alice and Bob for key generation.

[0245] In their safety analysis, the inventors considered a population with a standard deviation σ AB ≤ 8% statistical fluctuation between Alice and Bob’s power density spectra, which is usually satisfied in classical communication networks. The maximum tolerable BER is set equal to the spectrum σ AB This choice means that in all cases considered, the BER is below the tolerable limit of 11% set for QKD schemes.

[0246] Fig.12The diagrams in (d)-(f) show the results in terms of BER and adversary uncertainty per bit, which are given by the theoretical limit with |C(ω)|=1 and for different σ AB These results are based on the P measured by Alice and Bob. Alice , P Bob =P Alice +Δ and P reconstructed by Eve in the theoretical limit E =P Alice +Δ 1 of 10 6 A statistical collection of different chaotic power density spectra is calculated. The spectra have the following statistics:

[0247] like Fig.12 The results shown in the schematics in (d)-(f) show that, independent of the considered communication scenario, when the bit error rate (BER) between Alice and Bob reaches σ AB When the user can extract each measured spectral point N b >10-bit OTP key while maintaining single uncertainty against an attacker (uncertainty higher than 0.99 bit per bit).

[0248] refer to Fig.12 (a), the variance of Eve's uncertainty is negligible. This is generated by using a chaotic spectrum with universal statistics, since the result is independent of the specific sequence considered and applies identically to all sequences with the same universality. b The limit is set to the point where the BER is equal to the standard deviation σAB in the spectrum, then the number of possible bits N extracted for each measured spectral point is (d)N b =11 increases to (e)N b = 12, and (f) N b =14 and can be controlled by reducing the uncertainty Δ between Alice and Bob’s spectra.

[0249] These results show that in the theoretical limit and in the worst case for the user, the attacker has zero mathematical information about the user's secret key and the system is unconditionally secure.

[0250] Key generation throughput

[0251] The number of pulses N to be transmitted for generating the OTP key in the absence of active eavesdropping p yes:

[0252]

[0253] and equal to the message length Lm times the probability P of observing a repeated state at the same time position in Alice and Bob's sequences ov From each combined state The ratio between the number of bits extracted Nb. (ov) At maximum, the maximum key generation throughput is observed.

[0254] The inventors have proved that for user probability P (Alice) and P (Bob) The probability P of different values ​​of (ov) After each step, the state is changed. When P (Alice) =P (Bob) =0.25, the maximum value P can be obtained (ov) = 10%. In this case, by using a 1024-bit standard detector and extracting 10 bits per spectral sample, the protocol needs to transmit N p = L m pulses, corresponding to ≈1 / 1000 of the length of the message.

[0255] In the presence of active eavesdropping, the key generation throughput is reduced since the eavesdropper is introducing additional error states. The inventors demonstrate that the throughput changes introduced by these attacks are in the range of a few percent.

[0256] Regarding the physical implementation, the design and engineering of the chip will be discussed by demonstrating the above points 1-4 required for the communication protocol.

[0257] To provide an integrated biometric silicon chip, in some illustrative embodiments herein, a chaos chip is developed from a human fingerprint, which is highly complex and has been implemented in electronic files almost worldwide.

[0258] refer to Fig.13 , Fig.13 An integrated fingerprint silicon chip design and chaos analysis according to some specific illustrative embodiments of the present disclosure are shown. Fig.13 (a)-(b) show the acquisition and conversion of biometric fingerprint into a chaotic resonator. Fig.13 (c) shows the dynamics of the two trajectories with the smallest digital E displacement representable on a computer. Fig.13 (d) shows the average Lyapunov exponent (μ) in the phase space of possible input conditions and the percentage of the phase space filled with chaotic dynamics for different radii r of the scatterer. Fig.13 (e) shows the dynamics for three identical input conditions in three different infinitesimally small resonators.

[0259] For example, in Fig.13After the biometric scan shown in (a)-(b), the digital image of the user's fingerprint is transformed into a chaotic fingerprint microresonator consisting of a series of point scatterers made of reflective nanodisks of constant radius. Fig.13 (c) shows the dynamics of two example trajectories in a chaotic fingerprint microresonator. Here, the microresonator acts as a chaotic optical ball for the dynamics of an example light ray.

[0260] Those skilled in the art will appreciate that a microresonator may generally represent an optical microcavity, which may be understood as a structure formed by a reflective surface on both sides of a spacer layer or an optical medium or by wrapping a waveguide in a circular manner to form a ring. The former type forms a standing wave cavity, and the latter forms a traveling wave cavity. According to some illustrative examples, the microcavity may be only a few microns thick, with the spacer layer sometimes even in the nanometer range. Those skilled in the art will appreciate that, with a common laser, this forms an optical cavity or an optical resonator, allowing a standing wave to form within the spacer layer, or a traveling wave to travel around in the ring.

[0261] refer to Fig.13 (c)-(e), the inventors optimized the above requirements 1 and 3 of the cryptographic communication protocol. Fig.13 (a) Chaotic properties of the resonator fabricated in the process shown. The obtained fingerprint resonator is characterized by a large number of non-intersecting convex bodies that are randomly displaced in space, representing a finite form of Lorentz air billiards. This family of billiards has strongly chaotic behavior arising from the hyperbolic motion of the exponentially dispersive wave front. Fig.13 (c) shows the minimum floating point number ε=2.2 10 with the same initial orientation n and represented on a computer. -16 Propagation of a ray for the two input conditions (x0,n) (red solid line) and (x0+ε,n) (green solid line) that undergo a spatial shift. After a few collisions and despite the infinitesimal displacement, the dynamics diverge exponentially into two completely different trajectories.

[0262] refer to Fig.13 (d) provides a quantitative analysis of the chaos in the microresonator by calculating the average Lyapunov exponent (μ), which quantifies the average exponential growth of different input conditions and the volume of the phase space of input conditions filled by the chaotic dynamics as a function of the scatterer radius. For example, assuming that there is n s = 322 scatterers, which have a phase space that is completely chaotic (as required in point 1) for scatterers with radius r satisfying r ≥ 0.008L, where L denotes the radius along Fig.13 (c) The resonator width in the y direction. In this system, the fingerprint structure is strongly chaotic, and the average Lyapunov index is in the range of about 0.21 to about 0.35, as shown in Fig.13 (d) as shown.

[0263] This means that after n scattering collisions, the two different input conditions undergo exponential separation with an average rate of ∼e (μ)n , thus providing very quickly uncorrelated trajectories, such as Fig.13 (c) as shown.

[0264] refer to Fig.13 (e) Analyze the behavior of the fingerprint structure with respect to the above requirements of the cryptographic algorithm. This figure shows that by i Dynamics of three identical input conditions (pure blue, red, and green lines) emitted in three different fingerprint resonators realized by randomly shifting the position of i A vector of , where / ∈ i / =∈ min The analysis shows that infinitesimal transformations in the scatterer of the fingerprint induce a chaotic structure with trajectories that diverge exponentially with respect to the old trajectories. This is a general result as well as other possible transformations that can satisfy requirements 3 to 4.

[0265] Now, refer to Fig.14 , a wave analysis of a fingerprint chip from a FDTD simulation performed by the inventors is presented. Here, Fig.14 (a) shows the electromagnetic energy distribution in a resonator made of an air column on a silicon substrate for λ = 1550 nm and an input pulse of 150 fs length. Fig.14 (b) shows the probability distribution of electromagnetic energy inside the resonator. Fig.14 (c) and (d) show the correlation entropy between the binary sequences generated by the transmitted electromagnetic spectra obtained by shifting the emission position of the input light beam within 1 μm from x=y=0 under a displacement of 20 nm for TE polarized excitation and TM polarized excitation, where the average correlation entropy for the displacement is within ±150 nm.

[0266] refer to Fig.14 (a) shows the typical dynamics of light in a fully chaotic fingerprint resonator with L = 7 μm and r = 0.012L, calculated from finite-difference time-domain (FDTD) simulations for an air hole structure on a silicon substrate excited by a TE-polarized 150 fs long pulse centered at λ = 1550 nm. The chaotic scattering of light from this structure completely randomizes the wavefront, generating electromagnetic chaotic wave packets using general Gaussian energy statistics, as shown in Figure 2. Fig.14 As shown in the figure shown in (b).

[0267] refer to Fig.14(c)-(d), a detailed analysis of the correlation between different chaotic states that can be generated by the emission power density spectrum (PDS) from the resonator. It is worth noting that the electromagnetic spectrum is advantageous over the intensity signal because the former tends to be simpler and cheaper to implement experimentally. Here, the inventors calculated the emission electromagnetic spectra for both TE and TM polarized point sources with a duration of 150fs, centered at a wavelength λ=1550nm, and emitted at x=y=0, with displacements y1, y2,... along y within a range of 1μm, with a resolution of 20nm. For each input position, the inventors calculated the transmission energy spectrum generated by the TE and TM input light from the chip and converted it into a binary sequence by the AHB technique. Then, the inventors calculated the entropy correlation matrix H, where the elements H ij =-d ij log2 d ij -(1-d ij )log2(1-d ij ) is the Shannon information entropy of the Hamming distance dij between bit sequences i and j caused by the PDS obtained by inputting offsets yi and yj.

[0268] Those skilled in the art will appreciate that the entropy correlation matrix for a fully chaotic fingerprint resonator is strongly diagonal (see Fig.14 (c)), showing that the bit sequences generated from light emitted at each spatial shift with different input polarizations are completely uncorrelated. Fig.14 (d) provides a quantitative analysis of the correlation length based on the calculation of the average correlation entropy for a displacement of 300 nm. It shows that displacements exceeding 200 nm provide uncorrelated sequences.

[0269] After a complete explanation of the present disclosure, those skilled in the art will understand that Fig.11 In the communication system, by using simple displacement and two polarizations (excluding combination), the inventors can generate (1000 / 0.2··2) for each mm of chip length L. 2 =10 8 If we measure each spectrum using a commercially available line detector with 1024 pixels and convert each pixel into N bits, we can theoretically generate 0.1·NTb different keys per mm of chip unit length L and for each communication.

[0270] According to some specific illustrative examples of this article and as presented in the context of the experimental demonstration below, the fingerprint chip can be manufactured by electron beam lithography, patterning the fingerprint resonator structure onto a silicon substrate developed using a silicon-on-insulator (SOI) platform that is CMOS compatible, industrially available and operates in the standard telecommunication C+L bands.

[0271] refer to Fig.15 , an experiment with a single fingerprint chip is shown and will be discussed below. Here, Fig.15 (a) shows a scanning electron microscope (SEM) image of the fabricated sample. Fig.15 (b) shows the transmission spectra and the generated bit sequences corresponding to four different input positions, each spaced 10 μm apart. Fig.15 (c), (d), (g) and (h) show the results of the experiments from the same chip (see Fig.15 (c)), the same chip in direct and reciprocal configurations (see Fig.15 (d)), original chip (see Fig.15 (g)) and Fig.15 (e) The modified chip, Fig.15 (e) The modified chip and Fig.15 (f) The modified chip and the original structure (see Fig.15 (h)) Correlation entropy table between bit sequences generated from the same input position. Fig.15 (e)-(f) show examples of irreversible transformation sequences characterized by the deposition of water on the chip surface without purification using a colloidal suspension, Fig.15 (f) shows the same chip after drying.

[0272] refer to Fig.15 (a), shows a scanning electron microscope (SEM) image of an illustrative fingerprint chip. The entire structure may be L = 40 μm wide and 100 μm long.

[0273] refer to Fig.15 (b)-(c) show an example of a simple irreversible process that can be applied to a chip. The process involves first depositing a non-purified water droplet on top of the sample, which naturally contains colloids that act as additional scatterers (see Fig.15 (b)). The inventors noticed that the distribution of scatterers in these fluid structures was not reproducible. When the chip was dried naturally, another impurity distribution was obtained on the surface of the chip (see Fig.15 (c)).

[0274] After a complete explanation of the present disclosure, those skilled in the art will understand that according to some illustrative embodiments of the present disclosure, a drop of liquid (e.g., water or any other liquid containing a colloidal reservoir that acts as an additional scatterer), such as a doped hydrogel that can continuously deform with temperature, pressure, light or electrical signals, can be dropped onto the optical encryption pattern before applying an input signal to the optical encryption pattern. Those skilled in the art will also understand the concept of utilizing an elastic-photometamaterial that combines optics and solid mechanics; in this sense, other examples of additional scatterers include optical devices based on metamaterials.

[0275] Metamaterials are compounds beyond those found in nature that involve and assemble multiple elements such as metals and plastic materials. Their precise shape, geometry, size, orientation, and arrangement allow them to manipulate electromagnetic waves beyond what is feasible by using traditional materials. Thus, EM waves can be manipulated by blocking, absorbing, enhancing, or bending the waves.

[0276] As an elasto-optic effect, the skilled person understands a change in the refractive index of an optical fiber caused, for example, by a change in the length of the optical fiber core due to mechanical stress.

[0277] As a photoelastic material, the skilled person understands that metamaterials are able to process / control / modify light by changing its refractive index and the volume of its unit (elasticity) by means of light transformation according to atmospheric conditions such as temperature, pressure, light, electrical signals, etc. Several methods can be used to dope metamaterials to make them more sensitive to atmospheric conditions in order to manipulate EM waves. One example is mesoscopically uniform nanoporous microstructures as an example of combining optics with solid mechanics.

[0278] Furthermore, microgels (Mgs) can be applied directly to non-metamaterials, directly to any structure with functionality that allows such a structure to be molecularly altered by environmental conditions (light, temperature, pressure, etc.). Exploring the freedom of microgels will improve the sensitivity range (wavelength shift) and response time affected by these environmental conditions.

[0279] The inventors characterized the Fig.15 Chip in (a)-(c): The transmitted optical signal with a bandwidth of 100nm centered at the communication wavelength of 1550nm is end-fire coupled with a 60x aspheric lens mounted on an XYZ translation stage, and the repeatable spatial offset is 0.5μm.

[0280] refer to Fig.15 (d) shows the transmission spectra recorded at different input positions indicated on the right, and the corresponding generated bit sequences for each 1024 bits of the original chip (see Fig.15 (a)).

[0281] refer to Fig.15 (e) shows the entropy correlation between 40 different input positions, each offset by 1 μm.

[0282] The inventor pointed out that, in contrast to the above Fig.14 The theoretical predictions shown in (c)-(d) are consistent, and the generated bit sequences are completely uncorrelated with each other.

[0283] refer to Fig.15 (f) Verifies that the chip is Fig.11 The cryptographic scheme requires 2 optical interactions. Fig.15 (f) shows the correlation between the bit sequence created by emitting a signal from the input and collecting the transmission spectrum at the output, and the bit sequence measured by emitting a signal from the output and collecting the spectrum at the input. The chip is stable, with perfectly correlated bit sequences generated under direct and mutual emission conditions (according to Fig.15 (f) diagonal). These sequences are completely uncorrelated with each other (see Fig.15 (f): yellow area).

[0284] refer to Fig.15 (g), respectively, before and after the transformation Fig.15 The entropy correlation between the bit sequences created at the same position in the chip (a) and (b) is analyzed; Fig.15 (h) shows the Fig.15 Correlation between the sequences generated by the chips of (a), (b) and (c). These sequences are completely uncorrelated with each other, and the average correlation entropy per bit is (H) = 0.998 ± 0.001 bit. The experiment shows that the irreversible transformation (a, b, c) generates completely different chip responses.

[0285] After a complete explanation of the present disclosure, those skilled in the art will understand that when using Fig.15 (b) When the chip structure communicates, once the chip is in Fig.15 (c) is irreversibly modified, and an attacker will never be able to recreate any of the generated states. This process can be iterated arbitrarily and can also be integrated with other possible irreversible transformations (including thermal and sample deformation). Fig.15 (g)-(h) experimentally show that these chips meet the requirements of points 3-4 of the communication protocol.

[0286] Now, the on-chip implementation and integration of ultrafast light modulation will be discussed. Fig.14 As shown in (c), the fingerprint chip generates uncorrelated data sequences at each spatial point from TE and TM polarized light. Fig.15 (a) This is demonstrated experimentally by reporting the experimental results for the correlations between 40 different sequences generated by launching sample light of different linear polarizations TE and TM at different spatial locations. Using different input positions and light polarizations opens up the possibility of Fig.15 Block-level diagram of integrated ultrafast optical control shown in (b).

[0287] The initial stage consists of a modified Mach-Zehnder interferometer, excited by an unpolarized source or equivalently a combination of TE and TM polarized light sources. The Mach-Zehnder interferometer consists of a y-junction fiber polarization beam splitter, followed by two ultrafast electro-optic (EO) modulators and a final y-junction for recombination of the two polarization arms. The output fiber is then connected to a 1×N optical switch matrix, which will provide ultrafast selection of the input channels on the fingerprint chip. With this integrated structure, 4·N optical signals can be realized in each communication channel. 2 Both the EO modulator and the switch matrix can provide modulation up to hundreds of GHz and hundreds of channels using currently available technology, thus enabling on-chip integrated applications for ultra-fast cryptographic communications using the protocol developed in this work.

[0288] Now, about Fig.16 Discuss the security of cryptographic transmission and systems against spectral attacks, Fig.16 Experiments on key distribution and attacks in the spectral domain are shown. Fig.16 (a) shows the experimental setup. Fig.16 (b) shows the communication sequence between Alice and Bob. Fig.16 (c)-(e) show examples of data sent during one communication step, including individual random spectra of users as measured from the communication line (see Fig.16 (c)), the combined spectrum measured by Alice and Bob at the remote end (see Fig.16 (d)), and Fig.16 Absolute spectral difference between the combined state measured by the user (solid blue line) and the state reconstructed by an ideal attacker (solid orange line) in (e). Fig.16 (f)-(h) show encryption and decryption experiments, where the panels show the results obtained by Eve's best attempt to recreate the user's key.

[0289] In this regard, the inventors used Fig.16 (a) setup, assembled with cheap and readily available classical optical components. The inventors used a standard single-mode telecommunication fiber SMF-28, which connects two fully chaotic chips with L=40μm and different user fingerprints. The inventors selected different input coupling conditions by mounting the chips on an XYZ translation stage with a 1μm spatial offset. As in previous experiments, the inventors used light pulses with 100nm bandwidth and measured the power density spectrum with 0.1nm resolution. The inventors then selected random input conditions, collected the spectra and generated keys at each user end. The inventors assumed an ideal attacker who controls the communication channel and has unlimited technical capabilities.

[0290] refer to Fig.16(b) , shows a typical communication with a list of input locations for each user selection.

[0291] refer to Fig.16 (c)-(d) analyze the transmission for a set of input conditions and show: the power density spectra transmitted by Alice and BOB (see Fig.16 (c)) and the spectrum measured by the user at the remote end (see Fig.16 (d)). The combined state PDS measured by Alice B22A30 The power density spectrum of the PDS is the same as the spectrum measured by Bob. A30822 Totally relevant. Fig.16 The solid blue line in (e) quantifies the exact difference between the data measured by Alice and Bob. This is calculated in absolute terms by directly extracting the PDS amplitude read by the spectrum analyzer. The spectral difference is within 10 -4 and 10 -1 between.

[0292] refer to Fig.16 (e), the orange solid line shows the difference between the spectrum measured at Bob’s end and the spectrum reconstructed by Eve through the spectral attack. In this case, Eve measures the spectrum P transmitted in the communication line. B22 and P A30 , are identical copies of Alice and Bob’s spectrum analyzers (implemented here by using the same instrument), and then try to B22 P A30 / P S Reconstruct the combined state measured at the remote end, P S A person skilled in the art will understand that the wave packet A n and B n’ The chaotic nature of makes it theoretically impossible to reach the correlation value measured by Alice and Bob in their combined spectrum. Alice / Bob The closest representation of the spectrum of the measured combined state P Eve The theoretical limit is:

[0293]

[0294] Among them, |α(ω)| 2 ≤1 indicates unknown coupling coefficient (system uncertainty), Δ indicates source P S Statistical fluctuations of the spectrum. Fig.16 (e) By measuring P using Alice and Bob's optical spectrum analyzers S The P is collected by connecting the SMF-28 optical fiber directly to the user spectrometer of each user. B22 and P A30to quantify Eq. (13) without using any components that are not present in the communication setting and that might introduce additional uncertainties. Even in this measurement scenario, there are significant differences between Bob and Eve’s spectra.

[0295] refer to Fig.16 (f)-(h), These results are visualized using encryption and decryption experiments. Fig.16 (f), Alice uses AB =7% (this is Fig.16 (a) The average standard deviation generated in the system) optimizes the AHB transform to encode the data being sent using a key generated from a repeating spectral sequence. The encoding process follows the Vemam cipher, with a bit-by-bit XOR between the image and the key. The resulting ciphertext is then sent to Bob, who decodes it using his own generated key (see Fig.16 (g)). To decrypt the image, Bob performs a bit-by-bit XOR between the ciphertext and his own key generated from the combined spectrum measured at his end. The image decoded by Bob is correctly retrieved from the original. Conversely, Eve exploits the Fig.16 The spectral attack described in (e) reconstructs the key used to decode the image, which is just white noise and has no information about the message.

[0296] Those skilled in the art will appreciate that, using this basic configuration and design of the chip (L = 40 μm), approximately 40 μm can be extracted per transfer. 2 ·N b 1024 = 2.N b Mbit different keys, which need to be exchanged The spectrum is assigned to the length L m The OTP of the message. b ≥ 10, which is at least 10240 less than the length of the message.

[0297] According to the above content, the inventor introduces the application of Fig.16 The inventors consider a statistical population of 1000 random bit sequences of 1000 bits each, and they show that keys generated using the nonlinear AHB transform pass all tests, thus validating the scheme against the NIST standards for real-world cryptographic applications.

[0298] In summary, the inventors have designed, analyzed, and experimentally demonstrated a protocol for perfect secret cryptography using a CMOS compatible fingerprint silicon chip. The key properties of such a cryptographic system are:

[0299] The system provides a physical implementation of a one-time pad by using a public classical optical communication network.

[0300] The security of the system is evaluated following the Kerckhoff principle, where the attacker controls the communication channel, has access to the system before or after the communication, and has the technology to accurately copy any part of it. The second law of thermodynamics and the exponential sensitivity of chaos prevent the attacker from obtaining any information about the keys exchanged by the users, regardless of the attacker's technical capabilities and knowledge of the system. The theory confirmed by experiments shows that the result of any attack is a binary sequence of the keys exchanged by the users under maximum entropy.

[0301] The protocol is fully compatible with many techniques for privacy amplification and information reconciliation that have been developed for quantum cryptography schemes.

[0302] The system requires no electronic databases, private keys, or confidential communications beyond the initial communication required to authenticate the user.

[0303] Combined with the technological maturity, speed and scalability of classical optical communications, the results shown by the inventors open a new path to achieve perfect confidentiality cryptography on a global scale at limited cost.

[0304] The particular embodiments disclosed above are illustrative only, as the invention may be modified and implemented in different but equivalent manners apparent to those skilled in the art having the benefit of the teachings herein. For example, the above-described processing steps may be performed in a different order. Furthermore, there are no limitations to the details of the structure or design shown herein except as described in the claims below. Therefore, it is apparent that the particular embodiments disclosed above may be changed or modified, and all such variations are considered within the scope and spirit of the invention. Note that the use of terms describing various processes or structures in this specification and the appended claims, such as "first", "second", "third" or "fourth", is used only as a shorthand reference to these steps / structures and does not necessarily mean that such steps / structures are performed / formed in this ordered sequence. Of course, an ordered sequence of these processes may or may not be required, depending on the exact claim language. Therefore, the protection sought herein is listed in the claims below.

[0305] In addition to the general and detailed descriptions above, reference is made to the accompanying publication "Perfect secrecy cryptography via correlated mixing of chaotic waves in irreversible time-varying silicon chips" by A. Di Falco, V. Mazzone, A. Cruz & A. Fratalocchi, which is also incorporated herein by reference in its entirety.

[0306] appendix

[0307] Perfect-Secret Cryptography via Correlated Mixing of Chaotic Waves in Irreversible Time-Varying Silicon Chips Supplementary Material

[0308] A.di Falco, V.Mazzone, A.Cruz&A.Fratalocchi

[0309] I. Time Domain Attack: Computation

[0310] detail

[0311] We generate chaotic states An(t) and Bn0(t) with m=1,...,M different frequencies, where and λm=m, obtain a time series α with random amplitude nm , b nm ∈[0, 1], phase φ nm , ψ n′m ∈[0, 2π]. Then, each generated state A is converted into n (t) and B n’ (t) is converted to a bit sequence of length L = 10000 bits. The AHB transform initially applies a transform Δ (such as, for example, a convolution kernel) to the data sequence to be converted, and then uses the Gabor transform G to generate a binary sequence. For a given digital sequence Y I (I=1, ..., L), AHB is compactly represented as follows:

[0312]

[0313] In our analysis, we used a differential convolution filter in is a discrete convolution, and a Gabor transform with a threshold of th = 0. The average uncertainty of each bit is evaluated as follows. After the chaotic state is converted into a binary sequence, we calculate the normalized Hamming distance ab between two binary sequences a and b as follows:

[0314]

[0315] This value provides the probability that the two sequences differ in one bit on average. We then convert the generated value into the Shannon information entropy H = -d ab log2 d ab -(1-d ab )log2(1-d ab ). We evaluated the results on the statistics of 10,000 samples.

[0316] II. Active Attack

[0317] In this scenario, Eve measures the data and decides to replace a state with another state from a different chip, where E1, ..., E p The state meets the probability P eve These attacks are particularly dangerous if they allow an adversary to gain information about the key without the user being aware of it.

[0318] Main text Figure 1 The security of the protocol arises from the fact that when active eavesdropping occurs, the users measure unrelated states. If Alice and

[0319]

[0320] Replenish Figure 1 . Error rate introduced into the communication by actively intercepting the retransmission attack as a function of (a) Eve’s state change probability and (b) the number of chaotic states N in each chip. 7 We simulate the analysis of two sequences. Each element of Alice’s and Bob’s sequences is an integer random number n∈[0,N] and n′∈[0,N], representing a chaotic wave packet generated by the input conditions n and n′ in Alice’s and Bob’s chips, respectively. We assume that each chip contains N independent and uncorrelated chaotic states. The generation of a new state in the sequence occurs with probability P and is the same for Alice and Bob. Communication errors are introduced between Alice and Bob’s repeated sequences by replacing the state with a new state of the other chip, simulating Eve’s actions at each step. Eve’s decision to change state occurs with probability P. eve occur.

[0321] Bob issues A i and Bob B j , Eve uses E x Replacing these states, Alice measures and Bob These states are not related: in the best-case scenario, the pseudo-state E x From a chip with a chaotic structure that is exponentially different from the one used by Alice and Bob (point 4 of the communication scheme). If Eve replaces a state, then a user measures the correct state While another user measures an unrelated state, or Main text Figure 1 Two-way communication can identify these error conditions and minimize their presence in the final key.

[0322] Replenish Figure 1 Shows the Peve The results of active attacks with different probabilities, relative to the probability P of the user choosing to change the state after each communication step Alice , P Bob Normalization. We consider the best scenario for Eve, where P Alice =P Bob =P. Figure 1 a shows that even in the best scenario P eve / P = 1, when Eve uses the same probability as Alice and Bob, the error rate introduced in the sequence measured by the user is very small, less than 4%, with only N = 25 different chaotic waves. Figure 1 b shows that the error rate introduced by Eve is inversely proportional to the number of chaotic states N, and for chips with a large number of states N>>1, the error rate introduced by this attack tends to zero.

[0323] III. Sources of uncertainty in communication schemes

[0324] The uncertainty contained in the mutual power density spectrum of Alice and Bob is generated by the following uncorrelated parts:

[0325] Statistical fluctuation of source spectrum Δ s .

[0326] Noise thermal fluctuation Δc of the communication channel.

[0327] Fluctuation Δ under input conditions A , Δ B .

[0328] Uncertainty of the measuring device Δ M .

[0329] · Coupling coefficient α(ω): This is a systematic source of uncertainty (Type B of the NIST guidelines [1]). To evaluate α(ω), the user must measure the absolute value of the spectrum and normalize it. To do this, the user must use another measuring instrument that introduces an unknown coupling coefficient. This parameter can be iterated infinitely without the user knowing the first coefficient. It is the same problem as measuring the exact temperature of a body: the temperature of the thermometer needs to be known in advance, because it reads the equilibrium temperature between the thermometer and the coupling environment of the body. Typical light sources have statistical fluctuations in power amplitude of the order of a few percent. Just to make a concrete example, the CL-band tunable source T100S-HP from EXFO ensures an absolute wavelength accuracy of better than ±20 pm and a power repeatability sweep performance of better than ±0.05 dB, which implies a difference of as little as ±1% per wavelength. Currently available optics / optomechanics are stable to fluctuations below 10 nm, and the changes in the measured optical observables are small if compared to the source fluctuations. Thermal fluctuations in communication channels are also negligible, typically having an order of magnitude smaller than the source amplitude. Currently available measurement instruments have dynamic ranges in the range of tens of dB and spectral accuracy in the range of tens of picometers, providing very precise readings with standard deviations much smaller than the source amplitude variation. In a typical experimental implementation, we are therefore in the range of Δ A,B,c,M <<Δ S under the circumstances.

[0330] IV. Multi-bit AHB conversion

[0331]

[0332] Supplementary Figure 2. Block scheme of multi-bit AHB transformation.

[0333] Supplementary Figure 2d provides a block diagram of the process. For each single point x in the spectrum obtained for a user n First, it is decomposed into its mantissa m n and the unsigned exponent e n Then two nonlinear filter functions m n +1 = F 12 (m n , e n )n times modified generated physical data (m n , e n ), generate two independent data sequences, which are then converted into binary through Gabor transform with a threshold th=0.5.

[0334] By using the exponent e nThe two sequences generated by continuous random permutation in RP1 and RP2 are finally merged together through a bit-by-bit OR operation to become a single output sequence of N bits. The nonlinear filter we use is a chaotic one-dimensional tent map: n ,F(m n , e n )=μ, And for F(m n , e n )μ(1-m n ), the variable is μ. We use μ=μ0+(2-μ0)β x e n , where 1<μ0≤2 and x=1,2 represents each mapping. In the chosen range of values ​​of μ0 and for any irrational number m n , the transform maps the interval (0, 1) onto itself, generating white noise data with exponentially different trajectories for infinitely different input conditions. The advantage of this mapping is that it has a Lyapunov coefficient λ = logμ, which can be controlled from λ ≈ 0 (chaos after an infinite number of steps) to log 2 = 0.69 (chaos after a single iteration), which allows fine control over the dynamics of exponential separation of nearby trajectories.

[0335]

[0336] Replenish Figure 3 .In σ AB = 8% for the training of the nonlinear AHB transform to generate the optimized bit sequence. (a) Norm of the cost function |C| and (b) Optimized AHB coefficients (μ0, β1, β2) vs. the number of bit sequences N constituting the training dataset.

[0337] The values ​​of coefficients μ0, β1 and β2 that guarantee the minimum bit error rate (BER) between users are obtained by combining the states P Alice and P Bob The PDS spectra are calculated using a dataset where:

[0338] ·P Alice is the spectrum measured by Alice.

[0339] ·P Bob is the spectrum measured by Bob.

[0340] We consider different communication scenarios where Alice and Bob measure the spectral difference P Alice -P Bob , standard deviation

[0341] For each value of σAB, we generate different datasets of different chaotic spectra by the following procedure. The chaotic spectrum P measured by Alice Alice Generated by a random field following equation (1) of the main text, with 1000 different frequencies. The spectrum measured by Bob is generated as P Bob =P Alice +Δ, with a spectral difference Δ representing the uncertainty in Bob's measurement.

[0342] In our analysis, we consider the uncertainty Δ, where <Δ> = 0 and This represents the statistical fluctuation of the source.

[0343] We found optimal values ​​for μ0, β1, and β2 for the following desired quantities:

[0344] The minimum BER between the bit sequences generated by Alice and Bob at their remote ends.

[0345] The average number of zeros contained in the keys generated by each user, N z is 50% of the key length. This is one of various statistical requirements that keys be as random as possible [1].

[0346] By minimizing the following vector cost function C: C(B a , B b ,μ0,β1,β2)=(BER,0.5-N za , 0.5-N zb ) norm to impose these requirements, (3)B x is the bit sequence generated from the chaotic spectrum of each user: Alice (x = a), Bob (x = b), BER is B a and B b The bit error rate between zx is the number of zeros in each user’s key. For N and For each value of , C is minimized using the Nelder-Mead algorithm [2].

[0347]

[0348] Figure 4 The probability P of observing a repeated state at the same position in Alice and Bob's sequences is (ov) , as the probability P of a user changing state after each repeated communication (Alice) and P (Bob) The calculations are performed on a statistical sample of 1000 sequences, each with 100000 states.

[0349] Replenish Figure 3It shows that at different N and σ AB = 8% of the typical training. Figure 3 a Plots the minimization of the cost function versus the number of bit sequences N considered in the dataset for N = 10 (green), N = 12 (red), and N = 16 (blue). ch . The function converges quickly after 2 bit sequences, indicating that very few spectra are needed to train the AHB algorithm to generate an optimized bit sequence. This result depends on the fact that we are generating the bit sequence with a chaotic spectrum with a universal probability density (see Fig. 2c of the main text). This means that each spectrum has the same statistics and that even a single spectrum is statistically representative when a sufficiently large number of frequencies are considered. Figure 3 b shows the corresponding optimized values ​​of μ0, β1 and β2 for the three cases where N = 10, 12, and 16 are considered. They all converge to the same value independently of the considered N. By fixing N and changing σ AB The same result can be observed: the optimized μ0, β1 and β2 converge to the same value (not shown here for simplicity). This means that the required cost function C has a global minimum that allows the generation of an optimized bit sequence regardless of the input conditions N and σ AB how.

[0350] V Billiards Dynamics Simulation: Computational Details

[0351]

[0352] Replenish Figure 5 . Geometric computation of the reflection trajectory for the same input condition hitting two scatterers shifted ∈(a). Equivalent computation in the configuration where the scatterers remain fixed.

[0353] Light ball collisions are mathematical models for the chaotic study of classical and quantum waves propagating into a microresonator. They study the particle dynamics of light following the motion of a ray of light inside a resonator, elastically reflected at the boundaries due to Snell's law [3, 4]. Figure 3 In the fingerprint resonator, the dynamics starts from the initial position x0 = (x0, y0) and the initial velocity vector n = (n x , n y )=(cosα, sinα), the particle position x at t+Δt is evolved by the following formula:

[0354]

[0355] As long as the particle does not collide with a scatterer. When such an event occurs at x1 = (x1, y1), the particle is reflected on the scatterer surface along a new velocity vector n1, and the motion (4) starts again until the next collision.

[0356] The trajectory of the billiard ball in evolution consists of a map M = (x0, x1, ..., x n ) until the particle escapes the billiard ball. As the particle motion is strongly chaotic, it is important not to use any numerical iterative solution to evaluate the scattering collision points (x0, x1, ..., x n ). In our study, we employ an analytical solution to compute the intersection points between the trajectories of (4) and the scatterers in order to estimate the properties of the chaotic dynamics as accurately as possible.

[0357] The calculation of the Lyapunov exponents follows the algorithm detailed in [3]. The phase space of the input conditions is represented by all possible input vectors starting from an initial position x0 = (0, y0) shifted along the y-axis and pointing in all possible directions n = (cosα, sinα), where For each input condition (x0,n), we add the smallest spatial displacement representable on a computer and evolve the dynamics (x0,n) and (x'0 = x0+∈,n), continuously iteratively computing the collision map M = (x0,x1, ..., x n ) and M'=(x'0,x'1,...,x' n ). Then, we pass to evaluate the maximum Lyapunov exponent.

[0358] The average Lyapunov exponent is computed by averaging(μ>) over the space of different input positions y0∈[−L,L] and angular orientations n=(cosα,sinα).

[0359]

[0360] Supplementary Fig. 6. Block scheme for integrated ultrafast optical modulation with on-chip optical coupling from end to end of the communication line: (EO: electro-optic modulator, SM: switch matrix).

[0361]

[0362] Replenish Figure 7 The power density spectra Ps(ω) of the sources used by Alice (red line) and Bob (blue line), measured with the spectrometer used by each user.

[0363] VI. Dynamic exponential separation of identical input conditions in transformed fingerprint billiards

[0364] We consider an infinitesimal displacement s' at the position of the scatterer i →s iThe propagation of the same input condition in the billiard structure of +∈, ε is an infinitesimal quantity along any direction in space. The collision map generated by the dynamics of the same input condition starting from the original and displaced billiards is denoted as M = (x0, x1, ..., x n ) and M' = (x'0, x'1, ..., x' n ).

[0365] To show that the evolution represented by M and M′ undergoes exponential separation, we use mathematical induction. We first consider the same input condition (x in , n) starts with the dynamics of the first collision points x0 and x'0. Figure 5 a shows the geometric calculation of the reflection trajectory, which hits the continuous scatterer at the new collision point x1, x'1. Figure 5 It is shown how the same construction can be obtained in an equivalent configuration by fixing the position of the scatterer to the original unmodified billiard ball and replacing the input condition by the same infinitesimal transformation ∈ applied to the scatterer.

[0366]

[0367] Figure 8. Statistical tests of generated keys. Proportional results of NIST statistical tests for cryptographic applications, from a statistical population of 1000 keys, 1000 bits each. The red lines indicate the minimum threshold to pass each test.

[0368] Then, the final distance d between the new collision points 11 ' can be expressed as:

[0369] d 11’ =d0+∈, (5)

[0370] d0 is the distance between the collision points, as if they originated from an unmodified billiard ball. When the latter is completely chaotic, the distance d0 = ∈ e μ It grows exponentially with respect to the original displacement ∈. This means that the distance d 11’ =∈e μ +~∈e μ It is also growing exponentially.

[0371] We now consider the n+1th collision (x n+1 , ..., x' n+1 ), assuming that after the nth collision the distance d n,n’ =|x n -x' n |~e μn Exponential growth. Through repeated replenishment Figure 5The same construction of , we obtain an exponential growth of the distance at the n+1th collision. By induction, this means that the distance of the trajectories caused by the same input conditions in the original billiards and the modified billiards grows exponentially at the beginning and increases exponentially with each collision. For this series of transformations, all trajectories are affected and no old trajectory is observed in the modified billiards, because there will be at least one different reflection, which will cause an exponentially divergent motion.

[0372] Similar results can be obtained by adding different scatterers to the fingerprint pattern, in the limit when each new scatterer hits at least once during the dynamics of one of the input conditions in the unmodified structure. The proof is as follows. Let us first consider the case of adding a single scatterer. Two identical input conditions in the original resonator and the modified resonator observe identical trajectories until one ray hits the new scatterer. From this point, a trajectory is reflected at a different angle. In a completely chaotic structure, this means the creation of trajectories that are first exponentially divergent. The addition of a scatterer generates a new family of exponentially divergent trajectories for all input conditions that, in the unmodified resonator, generated trajectories that hit the new scatterer. When a sufficiently large number of scatterers are added and all input conditions hit at least one of the new scatterers, all trajectories are exponentially modified.

[0373] VII. Integrated On-Chip Implementation with Ultrafast Light Modulation

[0374] The integrated nature of the fingerprint chip enables ultrafast modulation using integrated optical coupling. While the implementation of this type of high-density device is beyond the scope of this paper, we discuss possible implementations with currently available technology.

[0375] Figure 4 The results in c show that the fingerprint chip generates uncorrelated data sequences from TE and TM polarized light at each spatial point. The use of different input positions and light polarizations can be combined into the block scheme of Supplementary Figure 6. The initial stage consists of a modified Mach-Zender, excited by an unpolarized source or equivalently a combination of TE and TM polarized light sources. The Mach-Zender interferometer consists of a y-junction fiber polarization beam splitter, followed by two ultrafast electro-optic (EO) modulators and a final y-junction for recombining the two polarization arms. The output fiber is then connected to a 1×N optical switch matrix [5].

[0376] Both the EO modulator and the switch matrix can provide modulation up to hundreds of GHz and hundreds of channels using currently available technology [5], thus providing ultrafast selection of input locations on the fingerprint chip without using any mechanical components. The output light from the optical switch is then fed directly on the fingerprint chip through a multi-taper channel coupler [6-8], which provides on-chip coupling from end to end to the communication line. Using this integrated structure, 4·N 2 Generation of different spectra.

[0377] [1] L. E. Bassham, III, A. L. Rukhin, J. Soto, J. R. Nechvatal, M. E. S. Mid, E. B. Barker, S. D. Leigh, M. Levenson, M. Vangel, D. L. Banks, N. A. Heckert, J. F. Ray, and S. Vo, SP 800-22 Rev. 1a. A Statistical Test Suite for Random and Pseudorandom Number Generators for Cryptographic Applications, Tech. Rep. (Gaithersburg, MD, USA, 2010).

[0378] [2] F. Gao and L. Han, Computational Optimization and Applications 51, 259 (2012).

[0379] [3] E. Ott, Chaos in Dynamical Systems, 2nd ed. (Cambridge University Press, 2002).

[0380] [4] N. Chernov and R. Markarian, Chaotic billiards (American Mathematical Society, 2006).

[0381] [5] R. Stabile, A. Albores-Mejia, A. Rohit and K. A. Williams, Microsystems and Nanoengineering 2, 15042EP (2016), literature review.

[0382] [6] FE Doany, BG Lee, S. Assefa, WM Green, M. Yang, CL Schow, CV Jahnes, S. Zhang, J. Singer, VI Kopp, et al., Journal of Lightwave Technology 29, 475 (2010).

[0383] [7] VR Almeida, RR Panepucci and M. Lipson, Optics Letters 28, 1302 (2003).

[0384] [8] Y. Vlasov, W. M. Green and F. Xia, Nature Photonics 2, 242 (2008).

Claims

1. A cryptographic key distribution system, include: A first optical encryption terminal, configured to transform an optical input signal into an optical output signal according to a first optical encryption pattern set at the first optical encryption terminal, wherein the first optical encryption pattern is formed in a first photonic nanostructure of the first chip and adopts a random and / or chaotic first scattering component, the first scattering component is obtained by drilling a hole in a silicon top layer of the first chip, the hole having a size between 0.1λ and λ, wherein λ is the wavelength of light in the first photonic nanostructure; a second optical encryption terminal, configured to transform an optical input signal into an optical output signal according to a second optical encryption pattern set at the second optical encryption terminal, the first optical encryption terminal and the second optical encryption terminal being optically coupled so as to exchange optical output signals with each other, wherein the second optical encryption pattern is formed in a second photonic nanostructure of the second chip and adopts a random and / or chaotic second scattering component, the second scattering component is obtained by drilling a hole in the silicon top layer of the second chip, the hole having a size between 0.1λ and λ, wherein λ is the wavelength of light in the second photonic nanostructure; a first electromagnetic radiation source configured to provide a first input signal of electromagnetic radiation to the first optical encryption terminal; a second electromagnetic radiation source configured to provide a second input signal of electromagnetic radiation to the second optical encryption terminal; a first detection element configured to detect at least one first amount of radiation and / or light associated with a first optical output signal of the first optical encryption terminal, the first optical output signal being based on a second input signal continuously transformed by the second optical encryption terminal and the first optical encryption terminal; a second detection element configured to detect at least one second amount of radiation and / or light associated with a second optical output signal of a second optical encryption terminal, the second optical output signal being based on a first input signal continuously transformed by the first and second optical encryption terminals; a first electronic processing unit coupled to the first detection element, the first electronic processing unit being configured to determine a first cryptographic key signal based on at least one first amount of radiation and / or light detected by the first detection element; as well as A second electronic processing unit is coupled to the second detection element, the second electronic processing unit being configured to determine a second cryptographic key signal based on at least one second detected amount of radiation and / or light detected by the second detection element.

2. The cryptographic key distribution system according to claim 1, in, Each of the first and second optical encryption patterns includes a plurality of regions of a first material randomly arranged in a second material, wherein the first and second materials differ from each other in at least one optical characteristic.

3. The cryptographic key distribution system according to claim 2, in, The first material has greater absorption and / or transmission and / or diffraction properties with respect to electromagnetic radiation than the second material.

4. A cryptographic key distribution system according to claim 2 or 3, in, The plurality of regions of the first material are arranged according to a random and / or chaotic and / or predefined scattering pattern of a respective optical encryption terminal.

5. The cryptographic key distribution system according to one of claims 1 to 3, in, Each of the first optical encryption pattern and the second optical encryption pattern includes an additional scatterer.

6. The cryptographic key distribution system according to claim 5, wherein the additional scatterer comprises one of the following materials: a material containing a colloid-retaining liquid, a material that can be continuously deformed using temperature, pressure, light or an electrical signal, an elastic metamaterial and an optical device based on a metamaterial.

7. A method for generating a cryptographic key in a cryptographic key distribution system, the cryptographic key distribution system having a first optical encryption terminal and a second optical encryption terminal, the method include: providing electromagnetic radiation from a first electromagnetic radiation source to the first optical encryption terminal as a first optical input signal; transforming the first optical input signal into a first optical output signal according to a first optical encryption pattern formed at the first optical encryption terminal, wherein the first optical encryption pattern is formed in a first photonic nanostructure of the first chip and adopts a random and / or chaotic first scattering component, the first scattering component is obtained by drilling a hole in the silicon top layer of the first chip, the hole having a size between 0.1λ and λ, wherein λ is the wavelength of light in the first photonic nanostructure; providing the first optical output signal as a second optical input signal to a second optical encryption terminal; transforming the second optical input signal into a second optical output signal according to a second optical encryption pattern formed at the second optical encryption terminal, wherein the second optical encryption pattern is formed in a second photonic nanostructure of the second chip and adopts a random and / or chaotic second scattering component, the second scattering component is obtained by drilling a hole in the silicon top layer of the second chip, the hole having a size between 0.1λ and λ, wherein λ is the wavelength of light in the second photonic nanostructure; detecting at least one first amount of radiation and / or light associated with the second optical output signal; determining a first cryptographic key based on the detected at least one first amount of radiation and / or light; providing electromagnetic radiation from a second electromagnetic radiation source as a third optical input signal to the second optical encryption terminal; transforming the third optical input signal into a third optical output signal according to the second optical encryption pattern; providing the third optical output signal as a fourth optical input signal to the first optical encryption terminal; transforming the fourth optical input signal into a fourth optical output signal according to the first optical encryption pattern; detecting at least one second amount of radiation and / or light associated with said fourth optical output signal; and A second cryptographic key is determined based on the at least one second detected amount of radiation and / or light.

8. The method according to claim 7, in, The first cryptographic key and the second cryptographic key are identical.

9. The method according to claim 7, in, The first optical encryption pattern and / or the second optical encryption pattern is modified before and after each communication.

10. The method according to claim 9, in, The first optical encryption pattern and / or the second optical encryption pattern is modified by a reversible or irreversible mechanical deformation applied to the first optical encryption pattern and / or the second optical encryption pattern.

11. The method according to one of claims 7 to 10, in, Each of the first optical encryption terminal and the second optical encryption terminal has at least two input terminals configured to receive an optical input signal generated by an electromagnetic radiation source, the method further comprising: For at least one of the first optical encryption terminal and the second optical encryption terminal, an input port of the selected optical encryption terminal is selected and an optical input signal is provided to the selected optical encryption terminal via the selected optical input port.

12. The method according to claim 11, in, In response to measuring an optical signal according to an optical input signal input via a selected optical input terminal, repeating the selection of the input terminal and the measurement of the optical output signal until all input terminals are selected, recording all instances of the measured optical signal and providing the recorded instances at each of the first optical encryption terminal and the second optical encryption terminal, extracting at least one optical signal of the measured optical signal that is the same at the first optical encryption terminal and the second optical encryption terminal, and generating a key based on the optical signal extracted at the first optical encryption terminal and the second optical encryption terminal.

13. A communication system for providing communication between a third party device and a controller employing a cryptographic key distribution system according to one of claims 1 to 4.

14. The communication system according to claim 13, in, At least one of the first electromagnetic radiation source and the second electromagnetic radiation source is an optical-to-electrical converter or modulator integrated in the third-party device.