CN115021826A - Intelligent coding and decoding computing system and method for optical computing communication - Google Patents

Abstract

The present application relates to the technical field of electrical digital data processing, in particular to an optical computing communication intelligent encoding and decoding computing system and method, wherein the system includes: an optical computing encoding module for unsupervised encoding and encryption of input communication information, An optical communication signal is generated; an optical fiber communication module is used to send an optical communication signal based on the optical fiber; and an optical calculation decoding module is used to receive the optical communication signal, and use optical modulation to decode and reconstruct the optical communication signal to restore the communication information. Therefore, the technical problem that the optical computing neural network in the related art is not suitable for the encoding and decoding functions of optical communication, resulting in the inability to meet the speed, quality and confidentiality requirements of optical communication at the same time, is solved.

Description

Optical computing communication intelligent codec computing system and method

technical field

The present application relates to the technical field of electrical digital data processing, and in particular, to an intelligent encoding and decoding computing system and method for optical computing communication.

Background technique

Communication is an important part of the modern information society. Optical fiber communication, with its advantages of low loss, high speed and high throughput, undertakes more than 95% of the load of modern information communication. At present, a large number of encoding, decoding and digital signal processing involved in optical fiber communication are all converting preprocessed electrical signals into optical signals and then transmitting them in optical fibers, or converting analog optical signals into electrical signal processing after receiving them. Therefore, optical fiber communication is a serious problem. Depends on the precision and speed of photoelectric conversion.

Therefore, if the complex processing in the electrical signal domain can be completed in the optical signal domain through optical computing, not only the power consumption and time cost of a large number of photoelectric conversion devices can be saved, but also the extra cost caused by photoelectric or electro-optical conversion can be effectively reduced. error, and improve the signal-to-noise ratio of the communication signal.

The related technology realizes complex intelligent computing in the optical signal domain through optical diffraction deep neural network, including fully connected network, convolutional neural network, reservoir computing, impulse neural network, etc., which can realize image classification, audio recognition, saliency detection and many other tasks. At the same time, optical diffraction deep neural networks can often achieve faster processing speeds and orders of magnitude lower energy consumption than equivalent electronic networks.

However, the optical computing neural network that can be implemented in the related art is not suitable for the encoding and decoding functions of optical communication. Optical communication has high requirements for encoding and decoding, not only to meet the speed of communication frequency, but also to meet high throughput and have certain encryption and anti-noise performance to prevent eavesdropping and bit errors in the communication process, and it is urgent to improve.

SUMMARY OF THE INVENTION

The present application provides an optical computing communication intelligent encoding and decoding computing system and method to solve the problem that the optical computing neural network in the related art is not suitable for the encoding and decoding functions of optical communication, resulting in the inability to meet the speed, quality and confidentiality of optical communication at the same time. technical issues required.

An embodiment of the first aspect of the present application provides an optical computing communication intelligent encoding and decoding computing system, including: an optical computing encoding module, used for unsupervised encoding and encryption of input communication information to generate an optical communication signal; an optical fiber communication module, The optical communication signal is sent based on the optical fiber; and the optical calculation and decoding module is used for receiving the optical communication signal, and uses optical modulation to decode and reconstruct the optical communication signal to restore the communication information.

Optionally, in an embodiment of the present application, the optical calculation and encoding module includes: a light source, used to convert a digital electrical signal composed of the communication information into a coherent optical signal; an optical diffraction phase modulation layer, used to convert the The coherent optical signal is compressed and modulated into the optical communication signal.

Optionally, in an embodiment of the present application, the optical computing and coding module is further configured to use a pre-established unsupervised learning variational autoencoder neural network based on the optical diffraction phase modulation layer to convert the coherent The optical signal is compressed and encoded into a preset low-dimensional space, and the distribution of the signal in the preset low-dimensional space is constrained by normal distribution to obtain a compressed optical communication signal.

Optionally, in an embodiment of the present application, the optical calculation and decoding module includes: an optical phase diffraction modulation layer for reconstructing the optical communication signal to restore the digital electrical signal.

Optionally, in an embodiment of the present application, the optical phase diffraction modulation layer is programmable by lithography processing or spatial light modulator, and the parameters of the optical phase diffraction modulation layer are optimized by a machine learning method until meet the optimization conditions.

Optionally, in an embodiment of the present application, the optical phase diffraction modulation layer is obtained by stacking a plurality of diffraction layers, wherein each diffraction layer is obtained by simulation using phase modulation and Fresnel propagation of a preset spatial distance. .

Optionally, in an embodiment of the present application, the optical fiber communication module includes: a coupling mirror array; an optical fiber bundle for performing space division multiplexing on optical communication signals coupled by the coupling mirror array; a collimating mirror array , which is used to restore the signal of the fiber exit end of the fiber bundle to space light to obtain the optical communication signal.

An embodiment of the second aspect of the present application provides an intelligent encoding and decoding method for optical computing communication, including: unsupervised encoding and encryption of input communication information to generate an optical communication signal; sending the optical communication signal based on an optical fiber; The optical communication signal is decoded and reconstructed using optical modulation to restore the communication information.

The embodiments of the present application can perform the function of unsupervised data encoding and decoding based on machine learning through the optical computing encoding module, and create an optical fiber communication signal encoding and decoding function that can be realized without additional photoelectric conversion and digital signal processing in the optical signal domain. The optical computing decoding module uses optical modulation to decode and reconstruct the optical communication signal, which can significantly improve the speed and energy efficiency of optical fiber communication encoding and decoding, and can realize the encryption function at the same time of encoding. potential. Therefore, the technical problem that the optical computing neural network in the related art is not suitable for the encoding and decoding functions of optical communication, resulting in the inability to meet the speed, quality and confidentiality requirements of optical communication at the same time, is solved.

Additional aspects and advantages of the present application will be set forth, in part, in the following description, and in part will be apparent from the following description, or learned by practice of the present application.

Description of drawings

The above and/or additional aspects and advantages of the present application will become apparent and readily understood from the following description of embodiments taken in conjunction with the accompanying drawings, wherein:

1 is a schematic structural diagram of an optical computing communication intelligent encoding and decoding computing system provided according to an embodiment of the present application;

2 is a schematic diagram of the principle of an optical computing communication intelligent encoding and decoding computing system according to an embodiment of the present application;

3 is a flowchart of an optical computing communication intelligent encoding and decoding computing system according to an embodiment of the present application;

FIG. 4 is a flowchart of an optical computing communication intelligent encoding and decoding computing method provided according to an embodiment of the present application.

Detailed ways

The following describes in detail the embodiments of the present application, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals refer to the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary, and are intended to be used to explain the present application, but should not be construed as a limitation to the present application.

The following describes the optical computing communication intelligent encoding and decoding computing system and method according to the embodiments of the present application with reference to the accompanying drawings. In view of the technical problem that the optical computing neural network in the related art mentioned by the above-mentioned background technology center is not suitable for the encoding and decoding functions of optical communication, resulting in the inability to meet the requirements of speed, quality and confidentiality of optical communication at the same time, the present application provides a A kind of optical computing communication intelligent encoding and decoding computing system, in which the optical computing encoding module can perform unsupervised data encoding and decoding functions based on machine learning, creating an optical signal domain without additional photoelectric conversion and digital signal processing. The achievable encoding and decoding of optical fiber communication signals, combined with the optical calculation and decoding module, uses optical modulation to decode and reconstruct optical communication signals, which can significantly improve the speed and energy efficiency of optical fiber communication encoding and decoding. It has the potential to be further improved by integrating with a variety of existing coding methods. Therefore, the technical problem that the optical computing neural network in the related art is not suitable for the encoding and decoding functions of optical communication, resulting in the inability to meet the speed, quality and confidentiality requirements of optical communication at the same time, is solved.

Specifically, FIG. 1 is a schematic structural diagram of an optical computing communication intelligent encoding and decoding computing system provided by an embodiment of the application.

As shown in FIG. 1 , the optical computing communication intelligent encoding and decoding computing system 10 includes: an optical computing encoding module 100 , an optical fiber communication module 200 and an optical computing decoding module 300 .

Specifically, the optical computing and encoding module 100 is used for unsupervised encoding and encryption of the input communication information to generate an optical communication signal.

In the actual execution process, the optical computing coding module 100 in this embodiment of the present application can be used to encode the input raw signal at the speed of light or near the speed of light through a pre-trained unsupervised neural network, and perform feature representation in a low-dimensional space , generates an optical communication signal to compress the input signal, thereby increasing the optical communication throughput.

Optionally, in an embodiment of the present application, the optical calculation and encoding module 100 includes: a light source and an optical diffraction phase modulation layer.

Among them, the light source is used to convert digital electrical signals composed of communication information into coherent optical signals.

The optical diffraction phase modulation layer compresses and modulates the coherent optical signal into an optical communication signal.

It can be understood that the light source can convert the signal to be transmitted, that is, a digital electrical signal composed of communication information, into a coherent optical signal. The optical diffractive phase modulation layer can compress the original optical signal into an encrypted signal for transmission, that is, an optical communication signal.

Optionally, in an embodiment of the present application, the optical computing encoding module 100 is further configured to compress and encode the coherent optical signal to a pre-established unsupervised learning variational autoencoder neural network based on the optical diffraction phase modulation layer. A low-dimensional space is preset, and the distribution of the signal in the preset low-dimensional space is constrained by normal distribution to obtain a compressed optical communication signal.

Specifically, the optical diffraction phase modulation layer can update parameters in the process of co-training with the reconstruction module through backpropagation, so as to achieve the purpose of unsupervised encoding and compression of the input image information into a low-dimensional space, and output low-dimensional feature Optical communication signals, where noise during fiber propagation can be added as a prior for neural network training.

Further, the establishment of the optical computational coding network model can be as follows:

The embodiment of the present application can establish an unsupervised learning variational autoencoder neural network according to the complexity of the communication data and the optical fiber link model, compress and encode the input signal into a low-dimensional space, and constrain it in the low-dimensional space through a normal distribution The distribution in space, using the following KL divergence as part of the loss function:

D[Q(z)||P(z|X)]=E _z～Q [logQ(z)-logP(z|X)],

Among them, z is the parameter in the coding space, X is the parameter in the original space, Q is the probability distribution in the coding space, P(z|X) is the data distribution to be learned and estimated, and Q(z) is the estimated positive state distribution, D is the required divergence, D[Q(z)||P(z|X)] is the distance between the two distributions P(z|X) and Q(z), E _z～Q is the The expectation of the distribution Q with respect to the variable z.

Due to the particularity of the optical computing coding module 100, there are certain constraints on the value of the low-dimensional space. Therefore, the embodiment of the present application can impose certain constraints on the learned mean value of the Gaussian distribution during training. At the same time, the variance in the learned Gaussian distribution can be understood as modeling the noise amplitude in the optical fiber. Therefore, the embodiment of the present application can be based on The physical real link specifies this variance training.

The optical fiber communication module 200 is used for transmitting optical communication signals based on the optical fiber.

As a possible implementation manner, the embodiment of the present application can transmit the compressed optical communication signal through the optical fiber bundle in a spatial division multiplexing manner through the optical fiber communication module 200 to realize the transmission of the optical communication signal.

Optionally, in an embodiment of the present application, the optical fiber communication module 200 includes: a coupling mirror array, an optical fiber bundle, and a collimating mirror array.

The optical fiber bundle is used for space division multiplexing of optical communication signals coupled by the coupling mirror array.

The collimating mirror array is used to restore the signal at the fiber exit end of the fiber bundle to space light to obtain the optical communication signal.

In the actual implementation process, the coupling mirror array can be used to couple the encrypted signal for transmission into the optical fiber bundle to realize space division multiplexing, and then complete the long-distance, low-loss optical signal transmission through the optical fiber bundle, and through the microlens The array converts the signal transmitted in the fiber bundle into space light to obtain an optical communication signal.

The optical computing and decoding module 300 is used for receiving the optical communication signal, and using optical modulation to decode and reconstruct the optical communication signal to restore the communication information.

As a possible implementation method, the optical computing and decoding module 300 in the embodiment of the present application can decode the spatial optical signal output by the optical fiber communication module 200, restore and reconstruct the original signal, and restore the original data for detection or next Phase transfer.

Optionally, in an embodiment of the present application, the optical calculation and decoding module 300 includes: an optical phase diffraction modulation layer.

Among them, the optical phase diffraction modulation layer is used to reconstruct the optical communication signal to restore it to a digital electrical signal.

Specifically, the optical phase diffraction modulation layer of the embodiment of the present application can reconstruct and restore the received compressed optical signal to the original signal, and then the restored original signal, that is, the digital electrical signal, can be detected by the detector or transmitted to the lower-level communication chain It is convenient for the normal use of optical communication.

Optionally, in an embodiment of the present application, the optical phase diffraction modulation layer is obtained by stacking a plurality of diffraction layers, wherein each diffraction layer is obtained by simulation using phase modulation and Fresnel propagation of a preset spatial distance.

Further, both the optical phase diffraction modulation layer and the optical diffraction phase modulation layer can be composed of multiple diffraction layers stacked, wherein each diffraction layer of the optical phase diffraction modulation layer can be performed by using phase modulation and Fresnel propagation at a certain spatial distance. simulation.

It should be noted that, in the embodiments of the present application, the propagation of light in all free spaces and homogeneous media can be simulated by Fresnel propagation, while the propagation of light in a single-mode fiber can be simulated by fundamental mode propagation.

In addition, the preset spatial distance can be set by those skilled in the art according to the actual situation, which is not specifically limited here.

Optionally, in an embodiment of the present application, the optical phase diffraction modulation layer is programmable by lithography processing or spatial light modulator, and the parameters of the optical phase diffraction modulation layer are optimized by a machine learning method until the optimization conditions are satisfied.

Specifically, in this embodiment of the present application, the parameters of the model can be optimized by using reverse gradient propagation, a machine learning network can be established according to the above-mentioned simulation model, the image to be processed is used as the input, and the image to be input is used as the reference for calculating the loss function to construct a suitable training set. , validation set, test set, use the minimum mean square error as the loss function, use the error back propagation algorithm to iteratively adjust the parameters of the optical diffraction phase modulation layer in the optical computing coding module 100, and adjust the compression ratio, noise modeling amplitude, Hyperparameters such as the optical diffraction phase modulation layer and/or the diffraction layer scale of the optical phase diffraction modulation layer, to obtain the best optimization results.

Further, in the embodiment of the present application, according to various parameters obtained by simulation optimization, the optical diffraction phase modulation layer and/or the optical phase diffraction modulation layer can be manufactured by using lithography technology or the spatial light modulator can be programmed to control, and the optical diffraction phase modulation layer and/or the optical phase diffraction modulation layer can be controlled by means of machine learning. The parameters of the optical diffractive phase modulation layer and/or the diffractive layer of the optical phase diffractive modulation layer are optimized.

To sum up, the embodiments of the present application can implement signal encoding and decoding in the optical signal domain through light-speed or near-light-speed devices based on an existing general optical fiber communication link. The optical computing encoding module 100 and the optical computing decoding module 300 support unsupervised conditions. High-speed and large-throughput reconstruction of new data can significantly improve channel throughput, and achieve encryption while compressing and reducing dimensionality, enabling communication links to achieve better transmission effects.

The working principle of the optical computing communication intelligent encoding and decoding computing system 10 of the present application will be described in detail below with reference to FIG. 2 and FIG. 3 , with a specific embodiment.

As shown in FIG. 2, the optical computing communication intelligent codec computing system 10 may include: an optical computing coding module 100, a light source 110, an optical diffraction phase modulation layer 120, a diffraction layer 121, an optical fiber communication module 200, a coupling mirror array 210, an optical fiber bundle 220 , the collimating mirror array 230 , the optical calculation and decoding module 300 , the optical phase diffraction modulation layer 310 , the diffraction layer 311 , the detector 320 and the lower-level transmission system 330 .

Among them, the optical computing coding module 100 can be used to encode the input raw signal at the speed of light or near the speed of light through a pre-trained unsupervised neural network, and perform feature representation in a low-dimensional space to generate an optical communication signal to convert the input Signal compression, thereby increasing optical communication throughput.

The optical computing coding module 100 may include: a light source 110 and an optical diffraction phase modulation layer 120 .

The light source 110 can convert the signal to be transmitted, that is, a digital electrical signal composed of communication information, into a coherent optical signal.

The optical diffractive phase modulation layer 120 is composed of a plurality of diffractive layers 121, which can compress the original optical signal into an encrypted signal for transmission, that is, an optical communication signal. During the training process, the parameters are updated to achieve the purpose of unsupervised coding and compression of the input image information into a low-dimensional space, and output light signals with low-dimensional features.

The optical fiber communication module 200 can transmit the compressed optical communication signal through the optical fiber bundle 220 by means of space division multiplexing, so as to realize the transmission of the optical communication signal.

The fiber optic communication module 200 may include: a coupling mirror array 210 , a fiber bundle 220 and a collimating mirror array 230 . In the actual implementation process, the coupling mirror array 210 can be used to couple the encrypted signal for transmission into the optical fiber bundle 220 to realize space division multiplexing, and then complete the long-distance, low-loss optical signal transmission through the optical fiber bundle 220 , and The signal transmitted from the optical fiber bundle 220 is converted into space light by the microlens array 230 to obtain an optical communication signal.

The optical calculation and decoding module 300 can decode the spatial optical signal output by the optical fiber communication module 200, restore and reconstruct it into the original signal, and restore the original data for detection or transmission in the next stage.

The optical computing decoding module 300 may include: an optical phase diffraction modulation layer 310 , a detector 320 and a lower-level transmission system 330 . Specifically, the optical phase diffraction modulation layer 310 in the embodiment of the present application is composed of a plurality of diffraction layers 311, and the received compressed optical signal can be reconstructed and restored to the original signal, that is, the digital electrical signal, and then the restored original signal can be composed of The detector 320 performs detection or is transmitted to the lower-level communication link by the lower-level transmission system 330, so as to facilitate the normal use of optical communication.

As shown in FIG. 3 , this embodiment of the present application may include the following steps:

Step S301: Establish a mathematical model of the optical computing codec network. The optical computing coding module 100 mainly realizes the unsupervised coding and encryption of the input communication information by establishing a mathematical model of the optical computing coding and decoding network, and generates an optical communication signal. Specifically, the embodiment of the present application can establish an unsupervised learning variational autoencoder neural network according to the communication data complexity and the optical fiber link model, compress and encode the input signal into a low-dimensional space, and constrain it through a normal distribution. Distribution in low-dimensional space, using the following KL divergence as part of the loss function:

D[Q(z)||P(z|X)]=E _z～Q [logQ(z)-logP(z|X)],

Among them, z is the parameter in the coding space, X is the parameter in the original space, Q is the probability distribution in the coding space, P(z|X) is the data distribution to be learned and estimated, and Q(z) is the estimated positive state distribution, D is the required divergence, D[Q(z)||P(z|X)] is the distance between the two distributions P(z|X) and Q(z), E _z～Q is the The expectation of the distribution Q with respect to the variable z.

Step S302: Establish a physical simulation model of the optical computing codec network. In the actual execution process, due to the particularity of the optical computing coding module 100, there are certain constraints on the value of the low-dimensional space. Therefore, the embodiment of the present application can impose certain constraints on the learned mean value of the Gaussian distribution during training. At the same time, the variance in the learned Gaussian distribution can be understood as modeling the noise amplitude in the optical fiber. Therefore, the embodiment of the present application can be based on The physical real link specifies this variance training.

In the optical computing decoding module 300, the optical phase diffraction modulation layer 310 is a stack of a plurality of diffraction layers 311, each of which is simulated with phase modulation and Fresnel propagation over a spatial distance.

It should be noted that, in the embodiments of the present application, the propagation of light in all free spaces and homogeneous media is simulated by Fresnel propagation, and in a single-mode fiber, it is simulated by fundamental mode propagation.

Step S303: Construct and select appropriate training sets, validation sets, and test sets. Specifically, in this embodiment of the present application, a machine learning network can be established according to the above-mentioned simulation model, an image to be processed is used as an input, and an image to be input is used as a reference for calculating the loss function to construct appropriate training sets, validation sets, and test sets, and use The minimum mean square error is used as the loss function, and the error back propagation algorithm is used to iteratively adjust the parameters of the optical diffraction phase modulation layer 120 in the optical computing coding module 100, and the parameters of the optical diffraction phase modulation layer 120 are adjusted by adjusting the compression ratio, noise modeling amplitude, and the optical diffraction phase modulation layer 120. Hyperparameters such as the scale of the diffractive layer 121 and/or the diffractive layer 311 of the optical phase diffractive modulation layer 310 can obtain the best optimization result.

Step S304: Use the error back propagation algorithm to optimize the model to determine the physical parameters of the optical diffraction phase modulation layer 120 and/or the optical phase diffraction modulation layer 310. In the embodiment of the present application, the diffraction layer 121 of the optical diffraction phase modulation layer 120 and/or the diffraction layer 311 of the optical phase diffraction modulation layer 310 can be controlled by using the lithography technology to manufacture or programmatically control the diffraction layer 311 of the optical phase diffraction modulation layer 310 according to the parameters obtained by the simulation optimization. .

Step S305: Manufacture the system model, install the hardware system, and test the communication quality. In this embodiment of the present application, an appropriate light source 110, a microlens array 230, an optical fiber bundle 220, etc. can be selected, and the hardware system can be correctly installed according to the simulation model, so as to realize the function of intelligent encoding and decoding of optical computing communication.

According to the optical computing communication intelligent encoding and decoding computing system proposed in the embodiment of the present application, the optical computing encoding module can perform the unsupervised data encoding and decoding function based on machine learning, creating an optical signal domain that does not require additional photoelectric conversion and digital signal processing. The optical fiber communication signal encoding and decoding can be realized, and combined with the optical calculation decoding module, the optical modulation is used to decode and reconstruct the optical communication signal, which can significantly improve the speed and energy efficiency of the optical fiber communication encoding and decoding, and can realize the encryption function at the same time of encoding, and It has the potential to be further improved by integrating with a variety of existing coding methods. Therefore, the technical problem that the optical computing neural network in the related art is not suitable for the encoding and decoding functions of optical communication, resulting in the inability to meet the speed, quality and confidentiality requirements of optical communication at the same time, is solved.

Next, the intelligent encoding and decoding computing method for optical computing communication proposed according to the embodiment of the present application will be described with reference to the accompanying drawings.

FIG. 4 is a flowchart of an optical computing communication intelligent encoding and decoding computing method according to an embodiment of the present application.

As shown in FIG. 4 , the optical computing and communication intelligent encoding and decoding computing method utilizes the above-mentioned optical computing and communication intelligent encoding and decoding computing system, which includes the following steps:

In step S401, the input communication information is encoded and encrypted unsupervised to generate an optical communication signal.

In step S402, an optical communication signal is sent based on the optical fiber.

In step S403, the optical communication signal is received, and the optical communication signal is decoded and reconstructed by optical modulation to restore the communication information.

It should be noted that the foregoing explanations on the embodiment of the optical computing communication intelligent encoding and decoding computing system are also applicable to the optical computing communication intelligent encoding and decoding computing method of this embodiment, and are not repeated here.

According to the optical computing communication intelligent encoding and decoding calculation method proposed in the embodiment of the present application, the optical computing encoding module can perform the unsupervised data encoding and decoding function based on machine learning, and create an optical signal domain that does not require additional photoelectric conversion and digital signal processing. The optical fiber communication signal encoding and decoding can be realized, and combined with the optical calculation decoding module, the optical modulation is used to decode and reconstruct the optical communication signal, which can significantly improve the speed and energy efficiency of the optical fiber communication encoding and decoding, and can realize the encryption function at the same time of encoding, and It has the potential to be further improved by integrating with a variety of existing coding methods. Therefore, the technical problem that the optical computing neural network in the related art is not suitable for the encoding and decoding functions of optical communication, resulting in the inability to meet the speed, quality and confidentiality requirements of optical communication at the same time, is solved.

In the description of this specification, description with reference to the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples", etc., mean specific features described in connection with the embodiment or example , structure, material or feature is included in at least one embodiment or example of the present application. In this specification, schematic representations of the above terms are not necessarily directed to the same embodiment or example. Furthermore, the particular features, structures, materials or characteristics described may be combined in any suitable manner in any one or N of the embodiments or examples. Furthermore, those skilled in the art may combine and combine the different embodiments or examples described in this specification, as well as the features of the different embodiments or examples, without conflicting each other.

In addition, the terms "first" and "second" are only used for descriptive purposes, and should not be construed as indicating or implying relative importance or implying the number of indicated technical features. Thus, a feature delimited with "first", "second" may expressly or implicitly include at least one of that feature. In the description of the present application, "N" means at least two, such as two, three, etc., unless otherwise expressly and specifically defined.

Claims (8)

1. An intelligent codec computing system for optical computing communications, comprising:

the optical calculation coding module is used for coding and encrypting the input communication information in an unsupervised manner to generate an optical communication signal;

an optical fiber communication module for transmitting the optical communication signal based on an optical fiber; and

and the optical calculation decoding module is used for receiving the optical communication signal, decoding and reconstructing the optical communication signal by utilizing optical modulation, and restoring the communication information.

2. The system of claim 1, wherein the light computing encoding module comprises:

the light source is used for converting the digital electric signal formed by the communication information into a coherent light signal;

and an optical diffraction phase modulation layer compressing and modulating the coherent optical signal into the optical communication signal.

3. The system of claim 2, wherein the optical computational encoding module is further configured to, based on the optical diffraction phase modulation layer, utilize a pre-established unsupervised learning variational self-encoder neural network to compressively encode the coherent optical signal into a preset low-dimensional space, and constrain distribution of the signal in the preset low-dimensional space by normal distribution to obtain a compressed optical communication signal.

4. The system of claim 2 or 3, wherein the optical computing decoding module comprises:

an optical phase diffraction modulation layer for reconstructing the optical communication signal to restore the digital electrical signal.

5. The system of claim 4, wherein the optical phase diffraction modulation layer is programmably controlled by a lithographic process or a spatial light modulator, and parameters of the optical phase diffraction modulation layer are optimized by a machine learning method until an optimization condition is satisfied.

6. The system according to claim 4 or 5, wherein the optical phase diffraction modulation layer is obtained by stacking a plurality of diffraction layers, wherein each diffraction layer is simulated by phase modulation and Fresnel propagation at a preset spatial distance.

7. The system of claim 1, wherein the fiber optic telecommunications module comprises:

an array of coupling mirrors;

the optical fiber bundle is used for performing space division multiplexing on the optical communication signals coupled by the coupling mirror array;

and the collimating mirror array is used for recovering the signal at the fiber outlet end of the optical fiber bundle into space light to obtain the optical communication signal.

8. A light computing communication smart codec calculation method, characterized in that it uses a light computing communication smart codec calculation system according to claims 1-7, wherein the method comprises the following steps:

encoding and encrypting input communication information in an unsupervised manner to generate an optical communication signal;

transmitting the optical communication signal based on an optical fiber;

and receiving the optical communication signal, decoding and reconstructing the optical communication signal by utilizing optical modulation, and restoring the communication information.

Images