Computing Method and System Applied to Large-Scale Intelligent Computing and Optical Sensing Chip Architecture

Abstract

The present disclosure relates to the field of optical computing technologies, and in particular, to a computing method and system applied to a large-scale intelligent computing optical-sensing optical chip architecture. Among them, the architecture includes: a pre-sensing computing unit and a sensing computing chip cluster, and the pre-sensing computing sub-units in the pre-sensing computing unit correspond one-to-one with the sensing computing chips in the sensing computing chip cluster; among them, the pre-sensing computing sub-unit is configured to receive an input optical field and perform parallel processing on multi-dimensional optical field information in the input optical field according to a task target to obtain a parallel-processed optical field; the sensing computing chip is configured to load the parallel-processed optical field onto a signal light based on a resonant ring resonance mechanism to obtain a loaded signal light, wherein each dimension of the loaded signal light carries scene information of the input optical field. The present disclosure adopting the above solution can improve the processing speed and efficiency of optical computing in natural scenes.

Description

Technical Field

[0001] The present disclosure relates to the field of optical computing technologies, and in particular, to a computing method and system applied to a large-scale intelligent computing optical chip architecture for computing sense of light. Background Art

[0002] With the rapid development of the fields of artificial intelligence and scientific computing, the complexity and scale of computing requirements are also increasing continuously. However, existing electronic computing technologies are limited by Moore's Law, and their performance is gradually approaching a saturation state, making it difficult to effectively meet the increasingly stringent requirements for computing power and power consumption of large-scale complex algorithms. Light has natural advantages such as high throughput and low latency during propagation, and optical computing technology using photons instead of electrons as the computing carrier is regarded as the key to breaking the existing computing bottleneck.

[0003] However, the currently prevalent optical computing paradigm often assumes that natural scenes have been recorded by sensors, and loads this information onto coherent light such as lasers through means such as phase modulators. In this process, the limitations of the sensor's optoelectronic conversion and analog-to-digital conversion speeds severely restrict the processing speed and efficiency of optical computing in natural scenes. Summary of the Invention

[0004] The present disclosure aims to solve at least one of the technical problems in the related art to some extent.

[0005] To this end, the first object of the present disclosure is to propose a large-scale intelligent computing optical chip architecture for computing sense of light to improve the processing speed and efficiency of optical computing in natural scenes.

[0006] The second object of the present disclosure is to propose a large-scale intelligent computing optical chip system for computing sense of light.

[0007] To achieve the above object, a first aspect embodiment of the present disclosure proposes a large-scale intelligent computing optical chip architecture for computing sense of light, including: a pre-sensing computing unit and a computing optical chip cluster for computing sense of light, where the pre-sensing computing sub-units in the pre-sensing computing unit and the computing optical chips for computing sense of light in the computing optical chip cluster for computing sense of light correspond one by one; wherein,

[0008] The pre-sensing computing sub-unit is configured to receive an input optical field, and perform parallel processing on multi-dimensional optical field information in the input optical field according to a task target to obtain a parallel processed optical field;

[0009] The computing optical chip for computing sense of light is configured to load the parallel processed optical field onto a signal light based on a resonant ring resonance mechanism to obtain a loaded signal light, wherein each dimension of the loaded signal light carries the scene information of the input optical field.

[0010] Optionally, the architecture further includes:

[0011] A microlens array for inputting the optically processed light field in parallel into the sensing and computing chip.

[0012] Optionally, the microlens array includes at least one of the following:

[0013] Metasurface;

[0014] Liquid crystal array.

[0015] Optionally, when the sensing and computing chip is used to load the optically processed light field in parallel onto the signal light based on the resonant ring resonance mechanism, it is specifically configured to:

[0016] Determine the bias voltage and modulation method according to the pre-trained network parameters;

[0017] Modulate the optically processed light field in parallel according to the modulation method to load the optically processed light field in parallel onto the signal light.

[0018] Optionally, when the sensing and computing chip is used to modulate the optically processed light field in parallel according to the modulation method, it is specifically configured to:

[0019] Determine the scene preprocessing task according to the bias voltage;

[0020] Preprocess the optically processed light field in parallel according to the scene preprocessing task to obtain a preprocessed light field;

[0021] Modulate the preprocessed light field according to the modulation method.

[0022] Optionally, the scene preprocessing task includes at least one of the following:

[0023] Feature extraction;

[0024] Noise suppression.

[0025] Optionally, when the pre-sensing computing sub-unit is used to process the multi-dimensional light field information in the input light field in parallel, it adopts at least one of the following processing methods:

[0026] For the amplitude dimension light field information, by utilizing the change in the refractive index of the metasurface and reflection and scattering at different spatial positions of the input light field, the regulation of the amplitude at different spatial positions is completed;

[0027] For the phase dimension light field information, by changing the geometric parameters of the metasurface unit, the regulation of the phase delay of the input light field is achieved;

[0028] For the polarization dimension light field information, by adjusting the anisotropy of the permittivity in the metasurface, the regulation of the polarization state of the input light field is achieved, so that light with different polarizations has specific propagation and transmission characteristics;

[0029] For the optical field information in the frequency dimension, by changing the geometric parameters of the metasurface units, the frequency selection of the input optical field is realized to achieve the regulation of the optical field information in the frequency dimension.

[0030] For the optical field information in the angular dimension, by adjusting the spatial arrangement of the metasurface units, the regulation of the light propagation angle of the input optical field is realized.

[0031] To achieve the above object, the second aspect of the present disclosure provides a large-scale intelligent computing and sensing optical chip system, including: the large-scale intelligent computing and sensing optical chip architecture shown in any one of the foregoing first aspects.

[0032] Optionally, the system further includes:

[0033] An optical computing unit, configured to receive the loaded signal light and perform optical computing on the loaded signal light to obtain an optical computing result corresponding to the input optical field.

[0034] Optionally, the optical computing unit uses an optical computing chip.

[0035] In summary, the large-scale intelligent computing and sensing optical chip architecture and system provided by the present disclosure, through the reconstruction and optimization of the sensors in the traditional paradigm, gradually eliminate the functions of the sensors and replace them with an intelligent computing and sensing optical chip architecture of pre-sensing computing - sensing chip. This architecture is an all-optical computing and sensing architecture that can directly load the scene information of the input optical field onto the signal light, can change the dependence on sensors in the traditional paradigm, effectively eliminate various delays introduced by sensors, and achieve true optical sensing and optical processing. Compared with the traditional method, the advantage of this architecture is that it can achieve seamless connection between sensing and computing on the same physical layer, can minimize the delay and energy consumption in the information processing process, and can significantly improve the processing efficiency of the input optical field.

[0036] Some of the additional aspects and advantages of the present disclosure will be given in the following description, some will become obvious from the following description, or will be understood through the practice of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The above and / or additional aspects and advantages of the present disclosure will become obvious and easy to understand from the following description of the embodiments in conjunction with the drawings, where:

[0038] Figure 1 is a schematic structural diagram of a large-scale intelligent computing and sensing optical chip architecture provided by an embodiment of the present disclosure;

[0039] Figure 2Schematic diagram of the structure of a large-scale intelligent computing and sensing optical chip architecture provided by another embodiment of the present disclosure;

[0040] Figure 3 Schematic diagram of the structure of a large-scale intelligent computing and sensing optical chip system provided by an embodiment of the present disclosure. Detailed implementation manners

[0041] The embodiments of the present disclosure will be described in detail below. The examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals denote 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 explain the present disclosure, but should not be construed as limiting the present disclosure.

[0042] With the rapid development of artificial neural network technology, the performance and complexity of machine vision algorithms have shown a significant upward trend. Along with this progress, the demand for computing power of new artificial intelligence technologies such as large models has been increasing day by day, driving the urgent pursuit of high computing power. However, with the gradual slowdown of Moore's law in existing electronic computing technologies, the improvement of processor performance is approaching saturation, which makes it unable to cope with the computing power and power consumption requirements of large-scale complex algorithms. The main bottleneck of traditional electronic computing architectures lies in their high power consumption and limited computing speed, making it difficult to effectively meet the high parallel processing capabilities necessary for dynamic machine vision processing.

[0043] In this context, optical computing technology has attracted wide attention due to its excellent high-speed and high-throughput characteristics, leading to the emergence of a variety of new optical computing paradigms. However, the traditional optical computing paradigm is a separated sensing and computing paradigm, that is, a paradigm in which sensing and computing are processed separately. It separates the sensor and the processor, resulting in the analog-to-digital conversion becoming a bottleneck in the system operation, and the time delay caused by it has become a significant obstacle to the wide application of optical computing technology.

[0044] The present disclosure will be described in detail below in conjunction with specific embodiments.

[0045] Figure 1 Schematic diagram of the structure of a large-scale intelligent computing and sensing optical chip architecture provided by an embodiment of the present disclosure. As Figure 1 shown, the large-scale intelligent computing and sensing optical chip architecture includes: a pre-sensing computing unit and a sensing and computing chip cluster, and the pre-sensing computing sub-units in the pre-sensing computing unit correspond to the sensing and computing chips in the sensing and computing chip cluster one by one.

[0046] According to some embodiments, the pre-sensing computing sub-unit is configured to receive an input optical field and perform parallel processing on multi-dimensional optical field information in the input optical field according to a task target to obtain a parallel processed optical field.

[0047] In some embodiments, the input optical field refers to the optical field that needs to be processed, and this optical field can be, for example, a natural scene.

[0048] In some embodiments, according to the task objective, the pre-sensing calculation subunit can respond to and process the optical field information in dimensions such as phase, amplitude, polarization, and frequency in the input optical field. Among them, the relevant parameters involved in the processing can be determined according to the task objective.

[0049] For example, if the task objective indicates static classification of the scene corresponding to the input optical field, the pre-sensing calculation subunit will complete the modulation of the optical field information in the spatial dimension, and compress the optical field information in dimensions such as polarization and frequency other than the spatial dimension optical field information;

[0050] For example, if the task objective indicates analysis of the scene spectrum of the input optical field, the pre-sensing calculation subunit can use a wavelength-sensitive dispersive medium to modulate the spectral information of the scene and retain as much information as possible.

[0051] It should be noted that the pre-sensing calculation subunit can retain more original scene information in the optically field after parallel processing by performing parallel processing on different dimensional information of the input optical field.

[0052] According to some embodiments, a sensing and computing chip is used to load the optically field after parallel processing onto the signal light based on the resonant ring resonance mechanism to obtain the loaded signal light.

[0053] In some embodiments, the resonant ring resonance mechanism means that for light with a wavelength equal to the resonance wavelength, the optical path it travels in the ring is exactly equal to an integer number of wavelengths. At this time, the laser in the ring will undergo constructive interference with the input laser. As the light corresponding to the wavelength continues to be input, the laser energy will continuously accumulate in the resonant ring, and there is almost no light intensity at the output end of the resonant ring. For light with a wavelength not equal to the resonance wavelength, the above-mentioned constructive interference effect is almost non-existent, and almost all the light is output through the output end. In summary, from the perspective of the input and output of the resonant ring, the output will drop sharply at a specific wavelength and be almost constant for other wavelengths. The sensing and computing chip precisely uses this mechanism to modulate light of different wavelengths.

[0054] In some embodiments, the signal light can be, for example, a coherent light wave such as a laser suitable for computing.

[0055] In some embodiments, each dimension of the loaded signal light carries the scene information of the input optical field.

[0056] Among them, the loaded signal light includes but is not limited to at least one of the following dimensions:

[0057] Space;

[0058] Time;

[0059] Polarization;

[0060] Phase;

[0061] Amplitude;

[0062] Frequency.

[0063] It should be noted that the sensing and computing chip can successfully convert the intensity information of the scene into the spectral dimension information of the laser through the resonance ring resonance mechanism. This process can effectively realize the real-time feeding of natural information, enabling the sensing and computing chip to obtain richer and more diverse scene data.

[0064] According to some embodiments, during the process of coordinated work between the pre-sensing calculation subunit and the sensing and computing chip, information can be directly transmitted in the architecture in the form of optical signals, avoiding delays and losses that may be caused during the conversion of traditional electrical signals.

[0065] In some embodiments, by realizing the direct mapping between the scene information of the input optical field and the signal light, this architecture can not only greatly improve the computing speed and data processing ability, but also help reduce power consumption and meet the computing requirements in resource-constrained scenarios. This architecture is expected to be applied to intelligent task scenarios such as unmanned driving and drones that require high-efficiency real-time processing capabilities and low-power solutions.

[0066] In summary, the architecture provided in this embodiment, through the reconstruction and optimization of sensors in the traditional paradigm, gradually eliminates the functions of sensors and replaces them with an intelligent sensing and computing optical chip architecture of pre-sensing calculation - sensing and computing chip. This architecture is an all-optical sensing and computing architecture that can directly load the scene information of the input optical field onto the signal light, can change the dependence on sensors in the traditional paradigm, effectively eliminate various delays introduced by sensors, and achieve true on-light sensing and on-light processing. Compared with traditional methods, the advantage of this architecture is that it can achieve seamless connection between sensing and computing on the same physical layer, can minimize delays and energy consumption during the information processing process, and can significantly improve the processing efficiency of the input optical field.

[0067] Optionally, the large-scale intelligent sensing and computing optical chip architecture further includes:

[0068] A microlens array for inputting the optically field after parallel processing into the sensing and computing chip.

[0069] According to some embodiments, through the microlens array, the optically field after parallel processing can be accurately input onto the sensing and computing chip. The design of this microlens array can not only improve the efficiency of optical input, but also effectively focus and distribute light rays, thereby optimizing the subsequent signal processing process.

[0070] In some embodiments, the microlens array may include at least one of the following:

[0071] Metasurface;

[0072] Liquid crystal array.

[0073] Among them, the metasurface includes a plurality of metasurface units. A metasurface unit refers to the basic unit for the metasurface to regulate the light field, which is composed of sub-wavelength scale micro-nano structures (such as columnar, stepped, etc.) and the substrate material and environmental medium where they are located. Based on the design of the rich degrees of freedom of the micro-nano structures in terms of shape, height, material, etc., diverse modulations of various dimensions (spectrum, polarization, phase, amplitude) of the light field can be achieved.

[0074] Optionally, when the pre-sensing computing sub-unit is used to perform parallel processing on the multi-dimensional light field information in the input light field, at least one of the following processing methods can be adopted:

[0075] For the light field information in the amplitude dimension, by utilizing the change in the refractive index of the metasurface and reflection and scattering at different spatial positions of the input light field, the regulation of the amplitude at different spatial positions is completed; among them, the amplitude target value to which the amplitude needs to be regulated can be determined according to the task target;

[0076] For the light field information in the phase dimension, by changing the geometric parameters (including but not limited to size, shape, etc.) of the metasurface unit, the regulation of the phase delay of the input light field is achieved; among them, the phase delay target value to which the phase delay needs to be regulated can be determined according to the task target;

[0077] For the light field information in the polarization dimension, by adjusting the anisotropy of the permittivity in the metasurface, the regulation of the polarization state of the input light field is achieved, so that light with different polarizations has specific propagation and transmission characteristics; among them, the polarization state target value to which the polarization state needs to be regulated can be determined according to the task target;

[0078] For the light field information in the frequency dimension, by changing the geometric parameters (including but not limited to size, shape, etc.) of the metasurface unit, the frequency selection of the input light field is achieved to realize the regulation of the light field information in the frequency dimension; among them, the frequency to be selected can be determined according to the task target;

[0079] For the light field information in the angle dimension, by adjusting the spatial arrangement of the metasurface units, the selective regulation of the light wave propagation direction and light wave propagation mode is achieved to realize the regulation of the light propagation angle of the input light field; among them, the light propagation angle target value to which the light propagation angle needs to be regulated can be determined according to the task target. Optionally, when the sensing and computing chip is used to load the parallel processed light field onto the signal light based on the resonant ring resonance mechanism, it is specifically used for:

[0080] Determine the bias voltage and modulation method according to the pre-trained network parameters;

[0081] Modulate the optically processed field after parallel processing according to the modulation method to load the optically processed field after parallel processing onto the signal light.

[0082] According to some embodiments, the bias voltage can be a pre-set bias voltage, which does not specifically refer to a certain fixed voltage and can be determined according to the actual application scenario and task. When the pre-trained network structure changes, the corresponding bias voltage and modulation method will also change.

[0083] It should be noted that through reasonable setting of the bias voltage, not only can the modulation of the input signal be achieved, but also a series of scene preprocessing tasks can be completed. This adjustable operation method can ensure the flexibility and adaptability of the architecture in different application scenarios and lay a solid foundation for subsequent complex task processing.

[0084] That is to say, the sense-compute chip can also determine the scene preprocessing task according to the bias voltage; preprocess the optically processed field after parallel processing according to the scene preprocessing task to obtain the preprocessed optically processed field; modulate the preprocessed optically processed field according to the modulation method.

[0085] In some embodiments, the scene preprocessing task includes but is not limited to at least one of the following:

[0086] Feature extraction;

[0087] Noise suppression.

[0088] It is easy to understand that this architecture directly modulates the information of the natural scene onto coherent light waves such as lasers suitable for calculation through the "sense-compute chip cluster", and at the same time completes operations such as basic feature extraction. Through coordinated work, the "sense-compute chip cluster" can perform fast feature extraction and preliminary analysis while capturing natural scene information in real time, not only supporting diverse perception tasks, but also showing excellent performance when processing complex scene information.

[0089] In summary, the architecture provided in this embodiment not only breaks through the methodological limitations of the traditional separation of sensing and computing, but also creates a new solution for on-light sensing and processing. At the same time, by effectively integrating the advantages of the sense-compute chip and the large-scale optical computing chip, it can provide a solid foundation for the intelligent sense-compute processing of future complex natural scenes and has important practical significance and broad market prospects in multiple application fields.

[0090] To implement the above embodiment, the present disclosure also proposes a large-scale intelligent sense-compute optical chip system, including: the large-scale intelligent sense-compute optical chip architecture provided in the foregoing embodiment.

[0091] Optionally, the large-scale intelligent computing and optical sensing chip system further includes:

[0092] An optical computing unit, configured to receive the loaded signal light and perform optical computing on the loaded signal light to obtain an optical computing result corresponding to the input optical field.

[0093] According to some embodiments, Figure 2 is a schematic structural diagram of a large-scale intelligent computing and optical sensing chip system provided by an embodiment of the present disclosure. As Figure 2 shown, the optical computing unit may adopt an optical computing chip, and the optical computing chip may be a large-scale optical computing chip.

[0094] In some embodiments, by setting a large-scale optical computing chip at the backend, the information transmitted in real time by the sensing and computing chip can be processed. This large-scale optical computing design improves the computing power, enabling the architecture to meet the requirements of large-scale natural scene intelligent computing and optical sensing processing.

[0095] It should be noted that during the operation of the architecture provided in this embodiment, the information processed by the sensing and computing architecture is coherent light carrying the original scene information, which can directly enter various proposed large-scale optical computing chips for more complex and advanced intelligent task computing. At this time, the advantages of the sensing and computing architecture are manifested. The multi-dimensional optical field information of the natural scene loads the information onto each dimension of the output coherent light (such as intensity, phase, and spectrum, etc.) through the proposed architecture, without going through traditional sensor perception, data storage, and other links. Therefore, the bottlenecks of optoelectronic conversion and analog-to-digital conversion are broken through. In addition, by designing the output features, this architecture can efficiently implement various intelligent task processing operations in the optical domain, thus promoting the further development of optical information processing technology.

[0096] Exemplarily, Figure 3 is a schematic structural diagram of a large-scale intelligent computing and optical sensing chip system provided by an embodiment of the present disclosure. As Figure 3 shown, the multi-dimensional optical field information of the natural scene is fused through pre-sensing computing, and the feature extraction and loading are completed by the sensing and computing chip. Finally, each dimension of the coherent light carries the original scene information, such as intensity and spectrum, etc. The processed information is optimized and extracted, and finally will be transmitted to the large-scale optical computing chip for more complex and advanced intelligent task computing. This important feature enables various intelligent task processing operations to be efficiently implemented in the optical domain through design, thus promoting the further development of optical information processing technology.

[0097] In summary, the system provided in this embodiment is applicable not only to fields such as image recognition and machine learning, but also demonstrates broad prospects and potential in various applications such as autonomous driving, intelligent monitoring, and virtual reality. With the ability to handle large-scale tasks, this system can simultaneously process input information from multiple sensors, thereby achieving highly integrated and intelligent scene understanding.

[0098] The collection, storage, use, processing, transmission, provision, and disclosure of user personal information involved in this disclosure all comply with the provisions of relevant laws and regulations and do not violate public order and good customs.

[0099] It should be noted that personal information from users should be collected for legal and reasonable purposes and should not be shared or sold outside of these legal uses. In addition, such collection / sharing should be carried out after obtaining the informed consent of the user, including but not limited to notifying the user to read the user agreement / user notice and sign an agreement / authorization including authorizing relevant user information before the user uses the function. In addition, any necessary steps should be taken to safeguard and protect access to such personal information data and ensure that others with access to personal information data comply with their privacy policies and procedures.

[0100] This disclosure anticipates providing embodiments that allow users to selectively block the use or access of personal information data. That is, this disclosure anticipates providing hardware and / or software to prevent or block access to such personal information data. Once personal information data is no longer needed, the risk can be minimized by restricting data collection and deleting the data. In addition, when applicable, personal identifiers are removed from such personal information to protect the privacy of the user.

[0101] The acquisition, transmission, storage, use, processing, etc. of data in the technical solution of this disclosure all comply with the relevant provisions of national laws and regulations.

[0102] It should be noted that in the embodiments of this disclosure, certain industry-existing solutions such as software, components, models, etc. may be mentioned. They should be considered exemplary, and their purpose is only to illustrate the feasibility in the implementation of the technical solution of this application, but it does not mean that the applicant has already or necessarily used this solution.

[0103] In the descriptions of the foregoing embodiments, the descriptions referring to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples" mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, the schematic representations of the above terms are not necessarily directed to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, without contradiction, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0104] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present disclosure, "a plurality of" means at least two, such as two, three, etc., unless otherwise specifically defined.

[0105] Any process or method description in a flowchart or described in other ways herein may be understood to represent a module, segment, or portion of code including one or more executable instructions for implementing a customized logic function or process, and the scope of the preferred embodiments of the present disclosure includes additional implementations, where the functions may be performed in a substantially simultaneous manner or in a reverse order according to the functions involved, rather than in the order shown or discussed, which should be understood by those skilled in the art to which the embodiments of the present disclosure pertain.

[0106] The logic and / or steps represented in the flowchart or otherwise described herein can, for example, be considered as a definable sequence list of executable instructions for implementing logical functions, which can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device), or used in conjunction with these instruction execution systems, apparatus, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device. More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection portion with one or more wirings (electronic device), a portable computer diskette (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or other suitable media on which a program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or other appropriate processing as necessary, and then stored in a computer memory.

[0107] It should be understood that various parts of the present disclosure can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application-specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0108] Those of ordinary skill in the art of this technology can understand that all or part of the steps carried by the methods of the above embodiments can be completed by a program instructing relevant hardware. The program can be stored in a computer-readable storage medium, and when the program is executed, it includes one or a combination of the steps of the method embodiments.

[0109] In addition, each functional unit in various embodiments of the present disclosure may be integrated into one processing module, may exist physically alone for each unit, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. When the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0110] The above-mentioned storage medium may be a read-only memory, a magnetic disk, an optical disc, etc. Although the embodiments of the present disclosure have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as a limitation to the present disclosure. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present disclosure.

Claims

1. A calculation method applied to a large-scale intelligent computing and optical sensing chip architecture, characterized in that, The large-scale intelligent computing and sensing optical chip architecture includes a pre-sensing computing unit and a sensing computing chip cluster, and the pre-sensing computing sub-units in the pre-sensing computing unit correspond one-to-one with the sensing computing chips in the sensing computing chip cluster; the computing method includes: Controlling the pre-sensing computing sub-unit to receive the input optical field, and performing parallel processing on the multi-dimensional optical field information in the input optical field according to the task objective to obtain the optically field after parallel processing; Controlling the sensing computing chip to load the optically field after parallel processing onto the signal light based on the resonant ring resonance mechanism to obtain the loaded signal light, wherein each dimension of the loaded signal light carries the scene information of the input optical field; Among them, when controlling the pre-sensing computing sub-unit to perform parallel processing on the multi-dimensional optical field information in the input optical field, at least one of the following processing methods is adopted: For the optical field information in the amplitude dimension, by utilizing the change in the refractive index of the metasurface and reflection and scattering at different spatial positions of the input optical field, the regulation of the amplitude at different spatial positions is completed; For the optical field information in the phase dimension, by changing the geometric parameters of the metasurface unit, the regulation of the phase delay of the input optical field is realized; For the optical field information in the polarization dimension, by adjusting the anisotropy of the permittivity of the metasurface, the regulation of the polarization state of the input optical field is realized, so that light with different polarizations has specific propagation and transmission characteristics; For the optical field information in the frequency dimension, by changing the geometric parameters of the metasurface unit, the frequency selection of the input optical field is realized to realize the regulation of the optical field information in the frequency dimension; For the optical field information in the angle dimension, by adjusting the spatial arrangement of the metasurface units, the regulation of the light propagation angle of the input optical field is realized.

2. The calculation method according to claim 1, wherein The architecture further includes a microlens array, and the computing method further includes: Controlling the microlens array to input the optically field after parallel processing into the sensing computing chip.

3. The calculation method according to claim 2, characterized in that The microlens array includes at least one of the following: Metasurface; Liquid crystal array.

4. The calculation method according to claim 1, characterized in that, The loading of the optically field after parallel processing onto the signal light based on the resonant ring resonance mechanism includes: Determining the bias voltage and modulation method according to the pre-trained network parameters; Modulating the optically field after parallel processing according to the modulation method to load the optically field after parallel processing onto the signal light.

5. The calculation method according to claim 4, characterized in that The modulating the optically field after parallel processing according to the modulation method includes: Determining the scene preprocessing task according to the bias voltage; Preprocessing the optically field after parallel processing according to the scene preprocessing task to obtain the preprocessed optically field; Modulating the preprocessed optically field according to the modulation method.

6. The calculation method according to claim 5, characterized in that, The scene preprocessing task includes at least one of the following: Feature extraction; Noise suppression.

7. A large-scale intelligent computing and optical sensing chip system, characterized in that, Including: The computing method applied to the large-scale intelligent computing and sensing optical chip architecture according to any one of claims 1 to 6.

8. The system according to claim 7, characterized in that The system further includes: An optical computing unit for receiving the loaded signal light and performing optical computing on the loaded signal light to obtain the optical computing result corresponding to the input optical field.

9. The system according to claim 8, wherein The optical computing unit uses an optical computing chip.

Images