Parallel optical computing system
By programming and modulating light through liquid crystal cells, a parallel optical computing system is constructed, which solves the problems of insufficient interaction and computational complexity in existing systems, and achieves efficient data processing and computing capabilities, making it suitable for industrial applications.
Patent Information
- Application Number
- CN202080099561.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2020-04-09
- Publication Date
- 2026-02-27
- Estimated Expiration
- 2040-04-09
AI Technical Summary
Existing parallel optical computing systems have shortcomings in terms of interaction with external digital systems and internal computational complexity, making it difficult to achieve efficient data processing and computation. Furthermore, their hardware configurations are not compact enough, limiting their widespread application in industry.
A parallel optical computing system is constructed by programming liquid crystal cells to modulate and process light. The system includes an optical modulator and an optical processor module. It uses polarizing filters and liquid crystal cells to perform light polarization and calculation, and combines prisms to reflect light, thereby realizing parallel operation of multiple sets of input data.
It enables efficient data exchange and internal computing operations between optical computing systems and external digital systems, has highly versatile hardware and software configurations, improves computing speed and system flexibility, and is suitable for highly complex computing tasks such as neurons and artificial intelligence networks.
Smart Images

Figure CN115398373B_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present invention relates to a parallel optical computing system. The present invention also relates to a parallel optical computing method. BACKGROUND
[0002] As is known, in the field of parallel optical computing, efforts are being made to develop systems capable of processing light instead of electricity.
[0003] The aim of developing such systems is not only to speed up the transmission interfaces to and from the telecommunication networks already transmitted through optical fibres, but also to increase the computing speed within the system, thus making it possible to use the system for high complexity computer operations, such as applications in general neural and artificial intelligence networks.
[0004] However, the currently known systems still have many drawbacks, mainly related to the complexity of the interface between the optical computing system and the external digital system (usually a binary coded system) and to the complexity of implementing the calculations within the optical system itself. In fact, for such optical systems to be used industrially and not only for research purposes, it is required that they can be reprogrammed and that their effective widespread use can be achieved only through reasonably compact hardware.
[0005] The article "Digital optical architecture for multi-matrix multiplication" by YU FT S et al. (Optics Communications, Elsevier, Amsterdam, XP024491185, ISSN: 0030-4018) is an example of an optical computing system.
[0006] In view of the above, the technical problem of the present invention is to conceive a parallel optical computing system which exhibits optimal performance features both in terms of inputting and outputting data to and from external systems (including telecommunication networks) and in terms of the information processing and related calculation operations performed within the system itself, while also adopting a highly versatile hardware and software configuration, substantially overcoming all the limitations and drawbacks which have so far affected the known optical computing systems. SUMMARY
[0007] The basic idea of the present invention is to program a liquid crystal cell to perform parallel optical computing, in particular to program the cell to modulate light representative of the input of the calculation to be performed, both with the aim of performing the calculation on the basis of the modulated light representative of the input and of outputting the result of the calculation in the form of light detectable by a light sensor (photodetector).
[0008] Advantageously, since each of these units can be reprogrammed for modulation and computation, based on the above concept, multiple sets of input data sets can be represented and different multiple operations performed on these data sets, each of said multiple representations and operations corresponding to a predefined programming of the modulation and computation units, hereinafter also referred to as part of a so-called first module (designed for modulation) and a second module (designed for operation as a processor of the computing system) of the parallel optical computing system.
[0009] Based on the above concept, the Applicant envisages various embodiments of the parallel optical computing system which differ from each other with respect to the extension of the number of modules of the computing system used for modulation and the number of modules used for computation and the relative spatial configuration of the modules, but all of them are based on the principle of the above concept.
[0010] In each of the embodiments, the parallel optical computing system comprises:
[0011] The parallel optical computing system comprises at least one first module comprising at least one polarizing filter and at least one liquid crystal unit, the first module being configured as an optical modulator for receiving light from a light source and encoding light output from the liquid crystal unit as optical data to be processed;
[0012] at least one second module comprising at least one polarizing filter and at least one liquid crystal unit, the second module being configurable as an optical processor for receiving the optical data to be processed and outputting processed optical results;
[0013] at least one optical detector designed to receive the processed optical results and convert them into corresponding electrical results.
[0014] The first module and / or the second module can comprise a plurality of liquid crystal units, thereby enhancing the parallel computing capability of the computing system. For the sake of simplicity only, the following description at least initially refers to a minimum module configuration, but it is clear that the same modules are not limited in terms of the number of liquid crystal units.
[0015] at least one prism positioned to receive the optical data from the first module and reflect it to the second module, wherein the first module is vertically aligned with the second module. Specific details of the above configuration are provided in the detailed description of the disclosure, in conjunction with the drawings. Basically, the prism comprises a base and a slanted wall connected to the base, wherein the base is positioned parallel to and facing the units of the first module and the second module, the slanted wall being used to reflect the optical data from the first module to the second module.
[0016] According to an aspect, the second module is configured as an optical processor and the optical detector is configured to receive the optical results from the second module.
[0017] According to another aspect, the optical computing system comprises at least one third module vertically aligned with the first and second modules below the second module and configured to receive optical data from the second module. According to this aspect, the second module is configured as a second optical modulator (and thus not necessarily as a processor) or as an optical processor, the third module being correspondingly (i.e. depending on the configuration of the second module) configured as an optical processor or as a second optical processor. At least one second prism is configured to receive optical data from the second module and to reflect the optical data to the third module. The second prism comprises a base and a slanted wall connected to the base, wherein the base is positioned parallel to the cells of the second and third modules and along a surface of the second module opposite to the surface of the second module where the first prism is located, the slanted wall being configured to reflect the optical data from the second module to the third module. At least in this possible configuration, the optical detector is configured to receive optical results from the third module.
[0018] However, the Applicant envisages other spatial configurations, the specific details of which are only briefly mentioned hereinafter and fully described in the detailed description of the present disclosure, in conjunction with the figures, where the reference numerals facilitate the understanding of possible example embodiments of the computing system.
[0019] For example, the first module of the optical computing system can be positioned facing the second module (and thus not necessarily vertically aligned).
[0020] Also according to this variant, the optical computing system can comprise a third module (in this example positioned facing the second module), which is configured as an optical processor, and the optical detector is configured to receive optical results from the third module.
[0021] As mentioned above, also in further embodiments of the optical computing system, each of the first, second or third modules can comprise a plurality of liquid crystal cells arranged laterally and / or vertically adjacent in the plane of the first, second or third module.
[0022] Further features and advantages of the parallel optical computing method and computing system according to the present application are provided in conjunction with the following description and the figures, which are merely non-limiting examples of the present application. BRIEF DESCRIPTION OF DRAWINGS
[0023] Figure 1 is a schematic view of an optical computing system according to the present disclosure.
[0024] Figure 2 is a schematic view of an optical computing system according to a variant of the present disclosure.
[0025] Figure 3 is a schematic view of an optical computing system according to another variant of the present disclosure.
[0026] Figure 4 is a schematic view of an optical computing system according to yet another variant of the present disclosure.
[0027] Figure 5 is a schematic view illustrating the basic principles of an optical computing system of the present disclosure as conceived by the Applicant.
[0028] Figure 6 is a schematic view of an optical computing system according to another variant of the present disclosure.
[0029] Figure 7 is a schematic view of an optical computing system according to yet another variant of the present disclosure.
[0030] Figure 8 is a schematic view of an optical computing system according to still another variant of the present disclosure.
[0031] Figure 9 is a schematic view of an optical computing system used by a possible encoding system in a non-limiting example of the present disclosure.
[0032] Figure 10 is a detail (a) of Figure 9
[0033] Figure 11 is a view showing a possible system for encoding light in a computing system according to the present disclosure. DETAILED DESCRIPTION
[0034] With reference to the drawings, reference numeral 1 represents and schematically denotes a parallel optical computing system according to the present disclosure, hereinafter also simply referred to as "computing system".
[0035] The computing system 1 comprises a plurality of liquid crystal cells combined together in modules, in particular at least a first module 10 and a second module 20 (for example Figure 1 and Figure 3 ) and optionally a third module 30 (for example Figure 2 and Figure 4 ) or other additional modules. As will be made clearer in the following, the specific modules are designed to implement specific functions, in particular light modulation functions 100, 101 and computing functions 200, 201.
[0036] The operation principle of the computing system 1 is independent of the number of liquid crystal cells comprised in the first module 10 and / or the second module 20 (and / or the third module 30 and / or the additional modules), and for the sake of easier description, the computing system 1 can be considered as having only one liquid crystal cell 13, 23 (optionally a third liquid crystal cell 33 for the third module 30) for each of the first module 10 and the second module 20. Despite the above, it is to be noted that, as envisaged by the Applicant, the industrialization of the computing system 1 is particularly suitable for using a plurality of liquid crystal cells for each of the first module 10 and the second module 20 (and optionally the third module 30), and as will be clearer in the following, in some embodiments, each function (in particular the light modulation function and the light computation function) has a plurality of modules, in order to enhance the parallelism of the data processing. Therefore, with respect to the following, the features described in the following in connection with each module are independent of the number of liquid crystal cells.
[0037] The liquid crystal cells 13, 23, 33 are located between two glass walls 95, through which the light rays emitted by the non-polarized white light source 70 are intended to pass. The white light source 70 is located as a light source at the input of the computing system 1. For example, the white light source 70 comprises a plurality of LEDs.
[0038] The light emitted by the light source 70 is intended to be modulated by the first module 10 according to the configuration or state of the liquid crystal cells 13, in order to represent the information or data to be processed. The light thus modulated (i.e. the light output from the first module 10) is then input to the second module 20 (and optionally to the third module 30 or to the additional modules), where it is processed according to the configuration or state of the liquid crystal cells 23, 33 of the second module and / or of the third module. According to different embodiments, the second module and / or the third module and / or the additional modules can be associated with a modulation function or a computation function.
[0039] In the following, an overview of the method for processing light in the modules is provided.
[0040] The light is constituted by a plurality of light rays, which are in fact superimposed by a large number of atoms randomly emitting light rays. In order to perform a parallel light computation process, the computing system 1 is configured to polarize the light rays, substantially by ordering the light rays through the liquid crystal cells of the first module 10 and of the second module 20 (and optionally of the third module 30 and of any additional modules), for the subsequent processing of the light rays.
[0041] This control operation is performed, in particular, by means of at least one polarizing filter 12 comprised in the first module 10 and at least one second polarizing filter 22 comprised in the second module 20 (or additional filters in the third module and / or in the additional modules). The light rays are intended to be further controlled by the modules 10, 20 (30), in particular by the liquid crystal cells 13, 23, 33 of the modules, which can be activated by means of a plurality of electrical impulses applied to a plurality of electrical contacts of the liquid crystal cells.
[0042] In modules 10, 20 (30), an electric field is applied by a predetermined program such that the crystals of liquid crystal cells 13, 23 (33) are arranged to allow or disallow light previously polarized by filters 12, 22 (32) to be phase-shifted according to the program to be performed, so as to modulate the light (in the first module 10) and perform calculations (in the second module 20), for example, a phase shift of 0° or 90°.
[0043] Figure 5 The schematic diagram illustrates the operating logic of liquid crystal cells 13, 23 (33), which can be used to form the first module 10 and the second module 20.
[0044] Figure 5 The upper section displays the OFF state of a single image element (pixel) of the twisted nematic light modulator LCD operating in "normal white" mode, i.e., a mode where light is transmitted through the liquid crystal when no electric field is applied. In the OFF state, i.e., without an applied electric field, in... Figure 5 A twisted configuration (spiral or helical structure) of nematic liquid crystal molecules is formed between two glass layers G. These two glass layers G are separated by various spacers and lined with transparent electrodes (not shown). Identical electrodes are lined with an alignment layer (…). Figure 5 (not shown in the image), when there is no external field, such as Figure 5 As shown above, the arrangement of layers precisely twists the liquid crystal by 90°.
[0045] If the light source is polarized ( Figure 5 The left side (indicated by the horizontal incident arrow) illuminates the front of the screen. The light passes through the first polarizer P2 and is transmitted to the liquid crystal, where the spiral structure rotates the light. The light is then appropriately polarized to pass through the second polarizer P1, which is positioned at 90° relative to the first polarizer. Therefore, in… Figure 5 ( Figure 5 In the example shown above, light passes through the back of the cell, and the image appears transparent (white rectangle on the right). This appearance of light (transparency) can be associated with light information.
[0046] Conversely, in the ON state, when a field is applied between the two electrodes, the crystal rearranges with the external field. Figure 1 (Lower part). This configuration prevents the crystal from rotating, thus preventing the polarized light passing through the liquid crystal from being redirected. In this case, the light is blocked by the rear polarizer P1, and the image appears opaque (the black rectangle on the right). This appearance of light (opaqueness) can be associated with light information that is different from that associated with the first appearance of light (transparency).
[0047] Based on the above operational logic, filters P1 and P2 can be used to form each module (i.e., the first module 10 and the second module 20 of computing system 1, as well as any other module 30, etc.). These filters are labeled, for example, 12 and 22 in the figure.Figure 2 ) and 32( Figure 1 ), the liquid crystal cells are indicated with 13, 23( Figure 1 ) and 33 (figure 33). As mentioned above, each of said cells allows controlling light.
[0048] The computing system 1 can advantageously interface with a serial input interface at the input and / or with a serial output interface at the output, respectively, the serial input interface providing data to be processed in the form of electrical signals, the serial output interface being configured to receive electrical data processed by the computing system 1 after it has been parallel modulated and then processed in parallel a number of mathematical operations. The electrical input data are converted into light, which is subjected to a corresponding modulation by the first module 10 of the computing system 1. Similarly, the light data processed by the computing system 1, in particular the data processed by the second module 20, are reconverted into data or electrical signals by the light detector 40 of the computing system 1.
[0049] The hardware of the first module 10 and of the second module 20 comprises electro-optical modulation means; in principle, the hardware of the first module 10 and of the second module 20 can be very similar or even identical, but the programming of the hardware determines the modulation function or the computing function in the system 1, at least in one of the possible embodiments of the computing system 1, substantially the first module 10 having the function of "light modulator" and the second module 20 having the function of "light processor".
[0050] The conversion of the information into light data allows performing parallel processing by applying different mathematical logic variants, such as multi-valued mathematical logic. The final result of the processing performed by the computing system 1 can be redirected to one or more other successive computing systems 1, i.e. other parallel light computing systems, or can be reconverted into serial data by the light detector 40 as mentioned above.
[0051] With reference to Figure 1 , the liquid crystal cells are grouped to form the first light modulation module 10 and the second computing module 20, and the liquid crystal cells are vertically aligned to each other in a plane, in turn coplanar, with the aim of forming a device or a display having predetermined dimensions (in inches). The vertically aligned cells face the light prism 50, which is designed to deflect by 180 degrees the light rays output from the cells of the first module 10 towards the corresponding cells of the second module 20, by making the amount of displacement on the display equal to the size of said cells. In a particularly compact and precise spatial configuration of the modules and of the corresponding cells, the light detector 40 faces the second module 20 along a surface of the second module 20 opposite to the surface of the second module 20 designed to receive at its input the light coming from the first module 10 and suitably deflected at the first light prism 50.
[0052] Figure 2The computing system 1 of the type shown can be advantageously used to perform computing operations using binary algebra. The light rays emitted by the non-polarized light source 70 are controlled by the first polarizing filter 12 and then enter the cell 13 of the first module 10. The polarizing filter 12 and the cell 13 of the first module 10 form the light modulator 10 of the computing system 1, which has the function of modulating the light so that it represents the serial data to be processed.
[0053] The light rays corresponding to the information to be processed exit from the light modulator 10 and are deflected by 180° by the prism 50 so as to enter the second module 20, which comprises a second liquid crystal cell 23 arranged vertically next to the first module, i.e. substantially coplanar with the respective cells of the modules 20, 10. The light rays that pass through the cell 23 of the second module 20 assume a polarization depending on the state of the cell 23, which can be pre-programmed depending on the computation performed using the second module 20 and pass through the polarizing filter 22. Finally, the light rays that exit from the polarizing filter 22 are detected by the photodetector 40 and represent the result of the computing system 1.
[0054] As mentioned above, the computing system 1 can be configured to perform multi-valued logic computing operations. For example, with reference to Figure 2 (not limited to the use of multi-valued computing logic), the computing system 1 is configured to comprise a third module 30. The computing system 1 comprises in particular three modules 10, 20, 30, each having at least one liquid crystal cell 13, 23, 33, a light source 70, a polarizing filter 12, 22, 32, two optical prisms 50 and 80 and a photodetector 40. The non-polarized light source 70 emits light rays, which are controlled by the first filter 12 and then enter the cell 13 of the first module 10 and pass through the second polarizing filter 42. The two polarizing filters 12, 42 and the cell 13 of the first module 10 form the light modulator 10.
[0055] The light rays, which are suitably modulated in correspondence with the data to be processed in the form of light, exit from the light modulator 10 and enter the prism 50, from which they are deflected (by 180° in the example given) towards the polarizing filter 82 of the second module 20, from which they are directed towards the liquid crystal cell 23 of the second module 20. The liquid crystal cell 23 of the second module 20 is vertically aligned with the liquid crystal cell 13 of the first module 10, i.e. coplanar, the surfaces of the cells 13, 23 being parallel to the surface or base of the prism 50, i.e. configured to input the light rays modulated by the first module 10 to the prism 50 and to output the same light rays again, re-directing them to the input of the second module 20, in particular to the polarizing filter 82, based on a very compact hardware structure.
[0056] The light rays that pass through the second cell 23 continue to assume the given polarization depending on the programming of the cell 23, as described above, and pass through the polarizing filter 22 of the second module 20, from which they are directed towards the second prism 80.
[0057] The second prism 80 has the same function of deflecting the light rays of the first prism 50, but is positioned between the second module 20 and the third module 30, i.e. the surface of the prism 80 is oriented (i.e. parallel to) the surfaces of the cells 23 and 33 of the second and third modules 20 and 30 and of the respective filters 22, 32, in particular with a suitable spacing, so that the light rays emitted by the cells 23 of the second module 20 exactly enter the cells 33 of the third module 30. The cells 33 of the third module 30 thus entered are controlled by the filter 32 of the module 30 positioned between the prism 80 and the cells 33.
[0058] Here it is not necessary to go further into the explanation, the flow of light rays through the third module 30 reaches the photodetector 40 via the filter 62 of the third module 30.
[0059] The light rays that have passed through the cells 33 of the third module 30 are polarized according to the state of the cells 33, which can be pre-programmed according to the calculation to be performed with the third module 30 and pass through the polarizing filter 62. Finally, the light rays output from the polarizing filter 62 are detected by the photodetector 40 and represent the result of the calculation system 1.
[0060] In combination Figure 1 , the second module 20 can be configured as a light modulation module or as a calculation module, i.e. as an additional modulator with respect to the first modulation module 10 or as an additional processor with respect to the second calculation module formed by the third module 30.
[0061] In the case where the second module 20 is configured as a second modulation module, the third module 30 can in fact be configured as a first calculation module, or in the case where the second module 20 is configured as a first calculation module, the third module 30 can be configured as a second calculation module.
[0062] Reference is now made to Figure 2 and Figure 1 The description provided relates to a calculation system 1 formed with only two liquid crystal cells or with three liquid crystal cells (i.e. one cell for each of the modules) in the so-called minimal configuration. But as mentioned above, this structure can be enlarged so that Figure 2 or Figure 6 each module comprises a plurality of cells and / or a plurality of modulation modules and / or calculation modules are used to form a screen of the calculation system 1 in vertical and / or horizontal adjacent arrangement.
[0063] Figure 7 and Figure 1 show respectively Figure 2 and Figure 6 an enlarged version of the systems shown in Figure 7 and Figure 3In this system, the structure of the modules, filters, and light sources can be repeated vertically and horizontally, thereby extending the computing system along two directions X and Y of the plane, corresponding to a basic bidirectional extension in the form of a light screen in optical system 1. To further enhance computing power, this disclosure also envisions a so-called three-dimensional extension of the computing system, in which multiple screens are arranged side by side along a direction Z perpendicular to directions X and Y.
[0064] According to different embodiments of this disclosure, the spatial configuration of the components of computing system 1 (i.e., the mutual arrangement of modules 10-30, light source 70, and detector 40) varies and does not need to be positioned next to the prism. For example, referring to Figure 4 and Figure 3 The publicly available examples are precisely this type of situation.
[0065] exist Figure 1 In this configuration, the first module 10 and the second module 20 face each other and are arranged between the light source 70 and the detector 40 along an axis perpendicular to the surface of the first filter 12, the surface of the unit 13 of the first module 10, the surface of the unit 23 of the second module 20, and the surface of the second filter 22. (Similar to reference...) Figure 3 The way it is described, in this Figure 4 In the example case, the first module 1 is configured as modulator 100 and the second module 2 is configured as processor 200.
[0066] exist Figure 2 In this configuration, the first module 10 and the second module 20 face each other and are arranged between the light source 70 and the detector 40 along an axis perpendicular to the surface of the first filter 12, the surface of the unit 13 of the first module 10, the surface of the unit 23 of the second module 20, and the surface of the second filter 22. However, in this example, a third module 30 is further inserted between the second module 20 and the detector 40 and includes a unit 33 and two filters 32A and 32B with surfaces perpendicular to the axis. In this case, similar to reference... Figure 8 In the described manner, the first module 1 is configured as modulator 100, the second module 2 can be configured as a second modulator 101 or a first processor 200, and the third module can be configured as a first processor 200 or a second processor 202.
[0067] Figure 3 Schematic illustration based on Figure 4 or Figure 4 The system is an extended form that includes multiple liquid crystal cells in each module. In particular, the unpolarized white light source (labeled 70) is configured to emit multiple rays of light, which are intended to be controlled at least by a first polarizing filter 12 and multiple cells 13 of the first module 10 (e.g., in the form of a first liquid crystal screen 10).
[0068] The first polarizing filter 12 and the cells 13 of the first module 10 form an optical modulator 100 of the computing system 1, having the function of converting serial data into optical data to be processed.
[0069] The first module 10 generates, at its output, a plurality of light rays, corresponding to the plurality of cells 13 of the module 10, each light ray assuming a polarization according to the state of the corresponding cell 13. The light rays are input into the respective cells 23 of the second module 20. These light rays are processed by the second module 20, in particular each light ray is processed by the corresponding cell 23 of the second module 20, depending on the state of the cell 23. The light rays thus processed represent the result of the computing system 1, intended to be detected by a plurality of sensors of the probe 40, each sensor being associated with a light ray output by the second polarizing filter 22 located between the probe 40 and the second module 20.
[0070] Similarly, an extended computing system 1 can be obtained from the basic structure shown. Figure 11
[0071] Figure 11 Examples are shown of possible ways of encoding the light in the computing system 1 based on binary, ternary, quaternary, double-track or n-track systems. During modulation, the encoding system serves to attribute to and correspond to the light characteristics representative of the data to be processed.
[0072] In the encoding system, not only is the information represented by the opacity (black) or transparency (white) of the light, as described above, but also by further characteristics of the light, i.e. of the light rays, in particular the white light direction (indicated by the Figure 11 corresponding arrow in white square in Figure 11 ) or other colors such as blue, green and red (indicated by the Figure 11 corresponding colored squares in Figure 9 ) or chromatic light direction (indicated by the Figure 9 corresponding arrow and colored square in .
[0073] Figure 2 Based on a logic system, i.e. substantially associated with the hardware arrangement of the components, the operation of the computing system 1 is schematically shown, for example based on the hardware architecture of Figure 11 .
[0074] In the first module 10 and in the second module 20, the serial data input to the computing system are encoded by one of the encoding systems shown in Figure 9 , so as to attribute to the light, in optical form, a predetermined modulation representative of the input data. This representation is shown as the output of the second module 20 in the logic diagram of Figure 9 .
[0075] The third module 30 performs a computation on the light rays output from the second module 20. This computation is predefined on a computation mask which transforms the light data (i.e. the light rays) input to the third module 30. In Figure 10 the computation mask is not clearly visible, therefore in Figure 11 the computation mask is shown in an enlarged scale. The computation mask is predefined according to the computation, i.e. the processing performed on the light rays within the third module 30. Thus the third module can be configured by redefining a different computation mask.
[0076] Again based on the encoding system used to modulate the light, for example one of the systems shown in the results of the optical computation can be reconverted into corresponding digital data for an external system in the detector 40.
Claims
1. A parallel optical computing system, comprising: A first module (10) includes at least one polarizing filter (12) and at least one liquid crystal cell (13). The first module (10) is configured as a light modulator (100) for receiving light from a light source (70) and encoding light output from the liquid crystal cell (13) into light data to be processed. The second module (20) includes at least one polarizing filter (22) and at least one liquid crystal unit (23). The second module (20) is configured as an optical processor (200) for receiving the optical data to be processed and outputting the processed optical result. At least one photodetector (40) is designed to receive the optical result of the processing and convert the optical result into a corresponding electrical result. The first module includes a plurality of liquid crystal cells (13), each of the plurality of liquid crystal cells (13) of the first module (10) being configured to emit modulated light toward a corresponding liquid crystal cell (23) of a plurality of liquid crystal cells (23) of the second module (20), the corresponding liquid crystal cell (23) of the second module being positioned in a plane XY at a predetermined distance from the liquid crystal cells (13) of the first module (10), wherein light emitted from the liquid crystal cells (13) of the first module (10) in the Z direction perpendicular to the plane is reflected in the Z direction, wherein, The optical modulator (100) is configured to modulate the incident light based on a predetermined binary, ternary, quaternary, dual-track, or n-track logic coding system. The optical processor (200) is configured to perform calculations on the light emitted by the optical modulator (100), the calculations corresponding to a predefined computation mask in the optical processor (200) based on the calculations. The optical processor (200) can be programmed to redefine different computational masks within the optical processor (200), the computational masks being loaded into the optical processor (200) as predetermined states of a plurality of liquid crystal cells of the optical processor. Its features are, The optical computing system includes at least one first prism (50) associated with a plurality of liquid crystal cells (13) of the first module (10) and a plurality of liquid crystal cells (23) of the second module (20) to receive optical data from the first module (10) and reflect the optical data to the second module (20). The first module (10) and the second module (20) are vertically aligned. The first prism (50) includes a base (51) and a sloped wall (52) connected to the base (51). The base (51) is parallel to and facing the plurality of liquid crystal cells (13) of the first module (10) and the plurality of liquid crystal cells (23) of the second module (20). The sloped wall (52) is used to reflect the optical data from the first module (10) to the second module (20).
2. The photonic computing system of claim 1, wherein, The second module (20) is configured as an optical processor (200), and the photodetector (40) is configured to receive the optical results from the second module (20).
3. The photonic computing system of claim 1, wherein, The optical computing system includes a third module (30) and at least one second prism (80). The third module (30) is vertically aligned below the second module (20) with the first module (10) and the second module (20) and is configured to receive optical data from the second module (20). The at least one second prism (80) is used to receive the optical data from the second module (20) and reflect the optical data back to the third module (30). The second prism (80) includes a base (81) and an inclined wall (86) connected to the base (81). The substrate (81) is parallel to the plurality of liquid crystal cells (23) of the second module (20) and the plurality of liquid crystal cells (33) of the third module (30) and is positioned along the surface of the second module (20) facing them, the surface of the second module (20) being opposite to the surface of the first prism (50) in the second module (20), the inclined wall (86) being used to reflect light data from the second module (20) to the third module (30), and the photodetector (40) being configured to receive light results from the third module (30).
4. The photonic computing system of claim 1, wherein, Each of the plurality of liquid crystal units (13) of the first module (10) and each of the plurality of liquid crystal units (23) of the second module (20) are arranged horizontally and / or vertically adjacent to each other along two vertical axes X and Y of the plane of the first module (10) and the second module (20), and the adjacent arrangement forms the screen of the computing system.
5. The optical computing system according to claim 4, characterized in that, The optical computing system includes multiple aligned screens or an array of aligned screens.
6. The optical computing system according to claim 1, characterized in that, The light source (70) includes multiple unpolarized light sources or LEDs, each of which is aligned with a corresponding liquid crystal unit of the first module (10).
Citation Information
Patent Citations
Single-lens local-gating imaging detector based on liquid crystal
CN101576410A
STED parallel microscopic imaging system based on structural light illumination
CN108121059A