Optoelectronic module using spatial light modulators and optical computer comprising same

The optoelectronic module using spatial light modulators in an optical computer addresses the speed limitations of electronic computers by enabling efficient parallel processing of large data sets, overcoming noise and complexity issues through light-based convolution operations.

WO2024228539A1PCT designated stage expired Publication Date: 2024-11-07KYUNGPOOK NAT UNIV IND ACADEMIC COOP FOUND
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
PCT/KR2024/005879
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-09-11
Filing Date
2024-04-30
Publication Date
2024-11-07

AI Technical Summary

Technical Problem

Conventional electronic computers are limited by delay time and stray capacitance, restricting high-speed computational processing, especially for 2D and 3D information, whereas optical computers leveraging spatial light modulators can overcome these limitations by processing larger input/output pixel arrays using light instead of complex electronic circuits.

Method used

An optoelectronic module comprising multiple light sources, lenses, spatial light modulators, and detectors that adjust and converge light rays to perform convolution operations, allowing for parallel processing of large pixel arrays without the complexity and noise of electronic circuits.

Benefits of technology

This solution enables faster and more efficient parallel processing of large data sets, exceeding the performance limits of conventional electronic computers by performing numerous multiplications and additions simultaneously without electromagnetic noise or complicated wiring issues.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure KR2024005879_07112024_PF_FP_ABST
    Figure KR2024005879_07112024_PF_FP_ABST
Patent Text Reader

Abstract

The present invention relates to an optoelectronic module using spatial light modulators and an optical computer comprising same. The optoelectronic module using the spatial light modulators comprises: a plurality of first light sources; a plurality of first lenses that reduce the divergence angles of light rays emitted from the first light sources; a plurality of spatial light modulators that adjust the intensity of light rays that pass through the first lenses; a plurality of light detectors that acquire a current value dependent on light intensity; and a lens unit that focuses the light rays, emitted from the first different light sources, onto one of the light detectors. The lens unit carries out the function of calculating the convolution of a light source array and a pixel array of the spatial light modulators by using the function of focusing the light rays, emitted from the first different light sources, onto the one light detector.
Need to check novelty before this filing date? Find Prior Art

Description

Optoelectronic module using spatial light modulator and optical computer including same

[0001] The present invention relates to an optoelectronic module using a spatial light modulator capable of redundant communication that can replace a telecommunication line, and an optical computer including the same.

[0002] Recently, in fields such as image processing, the demand for high-speed computational processing technology targeting two-dimensional and three-dimensional information has been rapidly increasing.

[0003] An optical computer is a computer that uses optical integrated circuits (Optical ICs) that utilize the properties of light in its computational circuits, allowing for more redundant communication than telecommunication lines.

[0004] Current electronic calculators are comprised of electronic circuits. Consequently, computational speeds are limited by factors such as component delays and stray capacitance, limiting their potential for high-speed computing. Optical computers are expected to overcome these limitations.

[0005] Figure 1 is a conceptual diagram of a typical convolutional neural network. As illustrated in Figure 1, a conventional CNN (Convolutional Neural Network) consists of an input image and a series of layers. In each layer, a series of kernels perform convolution with the input image, and the output of the convolution is pooled and passed to the next layer after a pooling operation.

[0006] In the conventional method of implementing optical CNNs to implement optically implemented convolution operations, the function of the optical convolution layer is implemented using a method of implementing a complex-valued Fourier filter.

[0007] Figure 2 is a conceptual diagram of a conventional 4f correlation optical convolutional neural network. Referring to Figure 2, the optical CNN architecture utilizes a 4f correlator design.

[0008] A traditional 4f correlation consists of two lenses, which have the same focal length, and the input and output planes can be located at the focal planes of the first and second lenses, respectively.

[0009] Meanwhile, using the Fresnel approximation, the lens provides a Fourier transform relationship between the electric field distributions located in the front focal plane and the back focal plane, and when two lenses are placed in series, it provides a Fourier transform that reverses the sign of the input electric field.

[0010] At this time, if a mask is placed on the plane between the two lenses, arbitrary Fourier domain filtering operations can be implemented, including low pass, high pass, band stop, band reject, and filtering operations, and more complex filtering is possible if a complex-valued transmittance mask is used on the filtering plane.

[0011] Here, the configuration of the 4f correlation array is configured so that the optical CNN architecture computes convolution operations on multiple input values ​​in parallel with one mask layer.

[0012] Figure 3 is a conceptual diagram of a conventional parallel optical convolutional neural network module. Referring to Figure 3, a convolution operation calculated in parallel is implemented using a system that adds two aligned lenslet array stacks and a filter mask array.

[0013] To implement a parallel photosynthetic neural network module like that shown in Figure 3, planar optics based on diffractive optical elements or metasurfaces are used. Each lens has a separate mask, allowing for parallel processing of different filtering functions.

[0014] To transmit and post-process images in an optical CNN system, an array of lenslets, an aligned source array, and a sense array that collects the output from the 4f correlations of each lenslet stack are required.

[0015] In the 4f correlation, the space-bandwidth product is can be given as, where D is the aperture size, f is the focal length, represents the optical wavelength.

[0016] For example, D is 0.6 mm, f is 3 mm, For a system with a resolution of 500 nm, the spatial bandwidth product is approximately 240х240, which means that the system can reliably compute approximately 240х240 input or output pixels.

[0017] Fig. 4 shows the input values Weighted in Bias by multiplying Then, add the sigmoid function This is a conceptual diagram of a conventional photosynthetic neural network module, which is a structure that calculates the applied value as an output value and uses it as an input value for the next layer.

[0018] From the input node's perspective, this is a method of calculating a linear combination by multiplying the weight corresponding to the kernel from one input node and sending it to multiple output nodes in the vicinity, and from the output node's perspective, it is a method of receiving values ​​from multiple input nodes in the vicinity, multiplying them by the weight, and then calculating a linear combination.

[0019] Although Figure 4 shows four inputs and four outputs as an example, the number of inputs and outputs is not limited to this.

[0020] The conventional method of calculating a neural network uses a digital computer, and can be expressed as in mathematical equation 1.

[0021] [Mathematical Formula 1]

[0022]

[0023]

[0024] Here, N represents the number of elements in the input value, and Nm represents the number of elements in the convolution kernel or filter.

[0025] This shows the case where the input / output and kernel are one-dimensional arrays, but it can be extended and used in the same way for two-dimensional arrays.

[0026] When implementing an electrical analog neural network and calculating simultaneously in this way, problems arise such as electromagnetic noise due to overlapping wires or the layout of circuit wiring becoming complicated.

[0027] The purpose of the present invention is to provide an optoelectronic module using a spatial light modulator capable of processing an input / output pixel array of a much larger size than conventional systems by utilizing the characteristics of light instead of complex electronic circuits, and an optical computer including the same.

[0028] In order to achieve the above object, an optoelectronic module using a spatial light modulator according to one embodiment of the present invention comprises: a plurality of first light sources formed to be spaced apart from each other by a predetermined distance; a plurality of first lenses formed to correspond to the plurality of first light sources, respectively, and to reduce a divergence angle of light emitted from the corresponding first light sources; a plurality of spatial light modulators formed to correspond to the plurality of first lenses, respectively, and to adjust the intensity of light passing through the corresponding first lenses through respective pixels having preset weights; a plurality of photodetectors formed to be spaced apart from the plurality of spatial light modulators and to obtain a current value according to the intensity of light; and a lens unit formed between the plurality of spatial light modulators and the plurality of photodetectors to collect light emitted from different first light sources passing through pixels at different relative positions of each spatial light modulator onto a single photodetector; wherein the lens unit comprises: a plurality of second lenses formed to correspond to the plurality of spatial light modulators, respectively, and to collect light onto a single point; And a plurality of third lenses corresponding to each of the second lenses, which pass the light rays that have passed through the second lens.

[0029] In addition, each spatial light modulator and the second lens corresponding to each spatial light modulator are formed to be spaced apart by a first distance, and each photodetector is formed to be spaced apart by a second distance from the third lens, wherein the second distance can satisfy a relationship in which the image of the pixel of the spatial light modulator is formed on the photodetector does not deviate by more than a certain error value.

[0030] Meanwhile, the lens unit is formed between a plurality of spatial light modulators and a plurality of photodetectors, and can implement a function of calculating a convolution of an array of light sources and an array of pixels of the spatial light modulator by using a function of gathering light rays irradiated from different first light sources that pass through pixels at different relative positions of each spatial light modulator into a single photodetector. In addition, a plurality of photodetectors corresponding to the third lens or arranged within the aperture of the third lens are provided, and each photodetector can perform a different convolution operation on light rays that pass through pixels of the spatial light modulator at different relative positions.

[0031] Another embodiment of the present invention may be an optical computer including an optoelectronic module using a spatial light modulator.

[0032] According to one aspect of the present invention described above, by providing an optoelectronic module using a spatial light modulator and an optical computer including the same, it is possible to process an input / output pixel array of a much larger size than a conventional system by utilizing the characteristics of light instead of a complex electronic circuit.

[0033] Figure 1 is a conceptual diagram of a general convolutional neural network.

[0034] Figure 2 is a conceptual diagram of a conventional 4f correlation optical convolutional neural network.

[0035] Figure 3 is a conceptual diagram of a conventional parallel photosynthetic neural network module.

[0036] Figure 4 is a conceptual diagram of a conventional photosynthetic neural network module.

[0037] Figure 5 is a conceptual diagram of an optoelectronic module using a spatial light modulator according to one embodiment of the present invention.

[0038] FIG. 6 is a three-dimensional example of an optoelectronic module using a spatial light modulator according to one embodiment of the present invention.

[0039] Fig. 7 is an example diagram of the pixel spacing of the spatial light modulator of Fig. 6.

[0040] Figure 8 is a conceptual diagram of a multi-output optoelectronic module using a spatial light modulator according to one embodiment of the present invention.

[0041] The following detailed description of the present invention refers to the accompanying drawings, which illustrate specific embodiments in which the present invention may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the present invention. It should be understood that the various embodiments of the present invention, while different from each other, are not necessarily mutually exclusive. For example, specific shapes, structures, and characteristics described herein may be implemented in other embodiments without departing from the spirit and scope of the present invention. Furthermore, it should be understood that the positions or arrangements of individual components within each disclosed embodiment may be modified without departing from the spirit and scope of the present invention. Accordingly, the following detailed description is not intended to be limiting, and the scope of the present invention is defined only by the appended claims, along with the full scope of equivalents to which such claims are entitled, if properly described. Like reference numerals in the drawings designate the same or similar functionality throughout the several aspects.

[0042] Preferred embodiments of the present invention will be described in more detail with reference to the drawings.

[0043] Below, the concept of an optoelectronic module (10) using a spatial light modulator (400) will be described through FIGS. 5 to 8.

[0044] FIG. 5 is a conceptual diagram of an optoelectronic module (10) using a spatial light modulator (400) according to one embodiment of the present invention.

[0045] An optoelectronic module (10) using a spatial light modulator (400) according to the present invention may include a substrate (100), a first light source (200), a lens unit (300), a spatial light modulator (400), a photodetector (500), and an electronic processing unit (600).

[0046] The substrate (100) may be formed with a plurality of first light sources (200), a plurality of light detectors (500), and a plurality of electronic processing units (600).

[0047] This means that one or more elements may be formed on the same substrate, but alternatively, they may be formed separately on different substrates.

[0048] Additionally, the substrate (100) may be formed as a semiconductor substrate or a PCB (Printed Circuit Board).

[0049] Meanwhile, the light source (200) may be a light emitting diode (LED), a semiconductor laser, or a vertical-cavity surface-emitting laser.

[0050] The light source (200) can control the amount of light by controlling the applied current, and the input values ​​of the current flowing to multiple light sources can be individually controlled.

[0051] Light from a light source (200) can pass through a lens unit (300) corresponding to the light source and then pass through a corresponding spatial light modulator (400).

[0052] The lens unit (300) is formed between a plurality of spatial light modulators (400) and a plurality of light detectors (500), and uses the function of collecting light emitted from different first light sources (200) passing through pixels at different relative positions of each spatial light modulator (400) into a single light detector (500), thereby implementing a function of calculating a convolution of the light source array and the pixel array of the spatial light modulator (400).

[0053] Additionally, the lens unit (300) may include a first lens (310), a second lens (320), and a third lens (330).

[0054] The first lens (310) is formed to correspond to each of the plurality of light sources (200) and can reduce the divergence angle of light emitted from the corresponding light source (200).

[0055] The second lens (320) may be provided in multiple units at positions corresponding to each of the multiple spatial light modulators (400) to focus the light rays into one point.

[0056] The third lens (330) may be provided in multiple numbers to correspond to each second lens (320) by passing the light rays that passed through the second lens (320).

[0057] A light ray passing through the pixel of the same space modulator (400) after passing through the first lens (310) can have the same angle after passing through the second lens (320).

[0058] Additionally, the lens unit (300) can be implemented using a refractive lens, a diffractive optical element, a meta lens, etc.

[0059] In addition, the light ray passing through the spatial light modulator (400) can pass through the corresponding second lens (320), be focused, and then spread out again to pass through the third lens (330). More specifically, when passing through the second lenses (320, 321), the angle of the light ray varies depending on the height of the light ray away from the optical axis of the second lenses (320, 321). The first light ray (350) passing through the highest position on the optical axis of the second lens (320) and the second light ray (351) passing through the highest position on the optical axis of the second lens (321) can have the same angle with respect to the horizontal axis.

[0060] Accordingly, the first light source (200) and the corresponding first lens (310) may be formed to be spaced apart by the focal length of the first lens (310), and the separation distance may include a certain error that deviates from the focal length. In addition, each spatial light modulator (400) and the corresponding second lens (320) may be formed to be spaced apart by a first distance, and each photodetector (500) may be formed to be spaced apart by a second distance from the third lens. Here, the second distance may satisfy a relationship in which the distance at which the image of the pixel of the spatial light modulator (400) is formed on the photodetector (500) does not deviate by more than a certain error value.

[0061] The spatial light modulator (400) is formed in a plurality of units corresponding to each of the plurality of first lenses (310), and can control the intensity of light passing through the first lens (310) through each pixel having a preset weight.

[0062] In addition, the relative position of the spatial light modulator (400) may mean the coordinates or position of the spatial light modulator pixel on the spatial light modulator plane based on the central axis of the first lens (310), second lens (320), or third lens (330).

[0063] The spatial light modulator (400) has spatial light modulator pixel arrays of a size corresponding to the lens diameter, and the relative positions of the pixels can be determined based on the corresponding lens central axis. For example, pixels located on the central axis of each lens can be said to be at the same relative position.

[0064] The spatial light modulator (400) plays a role in controlling the intensity of incoming light to pass through with different transmittance according to voltage, so it can be used as a liquid crystal display panel (LCD) that can control the intensity of each light beam to a predetermined value.

[0065] Meanwhile, even if the light originates from the same light source (200), the intensity of the light passing through the pixels may vary depending on the transmittance of each pixel in the LCD panel of the spatial light modulator (400).

[0066] For example, when the illuminance of the light source (200) is 1.0 W / ㎡, if the transmittance of the LCD pixel of the spatial light modulator (400) is 0.10, the illuminance of the output light can be 0.10 W / ㎡, and if the transmittance is 0.05, the illuminance of the output light can be 0.05 W / ㎡.

[0067] In addition, the transmittance of the LCD panel of the spatial light modulator (400) can be adjusted in 256 or more steps depending on the voltage, and the product of the weight and input value in the artificial neural network can be implemented.

[0068] Here, the input value is the intensity of the light source (200), the transmittance of the spatial light modulator (400) corresponds to the weight, the transmittance of the spatial light modulator (400) corresponds to the weight, and the number of pixels corresponding to the first lens (310) and the second lens (320) can be adjusted to be equal to the number of output nodes connected to one input.

[0069] The photodetector (500) may be composed of an optical semiconductor element such as a photodiode.

[0070] The photodetector (500) is formed spaced apart from a plurality of spatial light modulators (400) and can obtain a current value according to the intensity of light.

[0071] At this time, the photodetector (500) may be provided in multiple numbers corresponding to the third lens (330) or positioned within the aperture of the third lens (330), and each photodetector (500) may perform different convolution operations on light passing through the pixels of the spatial light modulator (400) at different relative positions.

[0072] At this time, the current flowing to the photodetector (500) is the sum of the intensities of light arriving at each photodetector. , can have a value proportional to .

[0073] In addition, the photodetector (500) can be arranged in multiple numbers in the optoelectronic module (10) by adjusting the ratio of the focal length f2 of the second lens (320) and the focal length f3 of the third lens (330) within the diameter of the lens corresponding to one input node.

[0074] For example, the number of photodetectors (500) is N p 2 When given as (N), the lens part (300) is placed within the diameter D, and the sub-array of the spatial light modulator (400) is (N p × N m ) 2 is given by D = N p × N m × d, so f3 / f2= N m If satisfies , N is within the area corresponding to the lens. p × N p A pixel image of the array is formed, and a photodetector (500) can be installed therein.

[0075] This is applicable when the array of input nodes is N × N by implementing an optical convolutional neural network structure using optical connections instead of pure electronic circuits, and the size of the convolution kernel array (N m × N m ) has some limitations, but the size of the array of input / output nodes is theoretically unlimited or can be very large in practice, so the amount of parallel data that can be optically processed in one instruction cycle is N. p If = 1, N 2 × N m 2 When using L layers of convolution, the parallel processing capacity is N 2 × N m 2 × It has the advantage of being able to overcome the performance limits of conventional electronic parallel computers by increasing L.

[0076] Also, N=10000, Nm = 165, L = 20, the parallel processing capacity is 5.4 × 10 13 With MAC (multiply accumulate operation) / cycle, even if the number of input terminals increases, the physical structure of the connection line does not become complicated or noise is generated like an electric circuit, and approximately N 2 × N m 2 × It can provide the effect of speeding up calculations by performing L multiplications and additions at once.

[0077] In addition, since the light rays arriving at the photodetector (500) do not interfere with photons of different paths even though the optical paths from the light source to the photodetector (500) overlap when the pixel values ​​of the spatial light modulator (400) are already determined, parallel calculation is possible, so the calculation time can be shortened.

[0078] The electronic processing unit (600) is an electronic direct circuit formed on a semiconductor substrate and can be implemented through an analog circuit or a digital circuit.

[0079] The electronic processing unit (600) adds the current value or its amplified value obtained from each corresponding photodetector (500) and the bias value stored in each electronic processing unit (600), and inputs the value into a sigmoid function. After calculating, we can obtain the neural network output value.

[0080] At this time, each output value of the electronic processing unit (600) can be transmitted to another connected light source (not shown).

[0081] Therefore, the output values ​​of the electronic processing unit (600) are , can be adjusted proportionally.

[0082] In addition, the electronic processing unit (600) can use other activation functions such as the RELU function in addition to the sigmoid function, and in addition to the function operation, it can store the result value or external input value and use it for the operation, so that the memory function of the electronic processing unit (600) can compare the calculated value in the next sequence with the stored value or perform various calculations using the stored value.

[0083] Additionally, the electronic processing unit (600) can exchange data with a surrounding electronic processing unit (600) or an external processor through a locally connected communication line, including a communication function.

[0084] This electronic processing unit (600) can electronically perform functions such as pooling and local response normalization (LRN) required in CNN through communication and calculation with electronic processing units (600) existing in the vicinity.

[0085] Additionally, the electronic processing unit (600) can store a bias value and can also perform a role of amplifying the current input from the photodetector (500) or converting the result of the operation into a current value proportional to the result when sending the result to the input of the second light source.

[0086] In addition, the second light source (not shown) located on the substrate (100) of FIG. 5 is a neural network , and can be used as input values ​​for the next layer.

[0087] Light rays in other directions starting from the first light source (200) can also be collected at other light detectors located on the second substrate (800) and connected as input values ​​of the next layer through the electronic processing unit (600) located on the second substrate (800).

[0088] In addition, the spatial light modulator (400) and the corresponding second lens (320) may be formed to be spaced apart by the focal length of the second lens (320) within a certain error, and each light detector (500) may be formed to be spaced apart from the third lens (330) within a certain error.

[0089] At this time, the spatial light modulator (400) and the light detector (500) can satisfy the positional relationship between the object and the image within a certain error by the imaging optical system composed of the second lens and the third lens.

[0090] Here, when the spatial light modulator (400) is located at a distance from the second lens (320) by the focal length of the second lens (320) and the third lens (330) is located at a distance from the photodetector (500) by the focal length of the third lens (330), the image of the pixel of the spatial light modulator (400) is generated on the photodetector (500), so that cross-talk that may occur between neighboring photodetectors (500) can be reduced.

[0091] Meanwhile, even if the distance between the spatial light modulator (400) and the second lens (320) is not exactly the focal length but slightly off, a tolerance can be allowed so that the image of the pixel of the spatial light modulator (400) is formed on the photodetector (500) by adjusting the distance between the third lens (330) and the photodetector (500).

[0092] In addition, the distance between the first light source (200) and the first lens (310) corresponding thereto can be adjusted to maintain performance even when the distance between other lenses or elements deviates from the focal length.

[0093] For example, a plurality of first lenses (310) corresponding to a plurality of light sources may be formed on another substrate, a plurality of spatial light modulators (400) may also be formed on another substrate, and a plurality of second lenses (320) and third lenses (330) may also be formed on another substrate.

[0094] This structure can have the effect of facilitating the volume or optical alignment of the entire system.

[0095] Below, a three-dimensional example of an optoelectronic module (10) using a spatial light modulator (400) and the pixel spacing of the spatial light modulator (400) will be described through FIGS. 6 and 7.

[0096] FIG. 6 is a three-dimensional example of an optoelectronic module (10) using a spatial light modulator (400) according to one embodiment of the present invention, and FIG. 7 is an example of the spacing between pixels of the spatial light modulator (400) of FIG. 6.

[0097] f1, f2, and f3 in FIG. 6 may be the focal lengths of the first lens (310), the second lens (320), and the third lens (330).

[0098] The spatial light modulator (400) pixels of FIGS. 6 and 7 are composed of blocks of a 3x3 array, and the spacing between blocks is indicated by a.

[0099] The spacing a is equal to the spacing or diameter between the lenses of the lens unit (300) of FIG. 5, and may be equal to the spacing of the input nodes (first light source) or output nodes (photodetectors), so the size of the block array may be equal to the size of the input node array.

[0100] Here, inside the block, it is again composed of a 3Х3 array in the form of a sub-array, where the size and spacing of the pixels are respectively and can be expressed as d.

[0101] These sub-arrays correspond to kernels or filter masks in convolution operations, and implement weight values ​​through the transmittance of the spatial light modulator (400), and when performing convolution based on the relative position of the pixel from the center of the sub-array, it is possible to determine to which surrounding output node the light passing through the pixel is directed after passing through the second lens (320) and the third lens (330).

[0102] Also, since the size of the sub-array is the same as the array size of one input node and the output node where convolution is performed, if the size of the sub-array is 5Х5, it can be connected to the output nodes of the surrounding 5Х5 arrays to perform a convolution operation.

[0103] Meanwhile, if all pixels are spaced apart by a distance d, they are positioned to satisfy a=5d, and since the second lens (320) and the third lens (330) form a relay optical system and focus the image of the pixels on the surface of the photodetector (500), the focal length f3 of the third lens (330) can be five times the focal length f2 of the second lens (320).

[0104] In general, if the array size of the convolution kernel is Nm×Nm, , and the aperture of the lens is And It could be.

[0105] Here is a fixed value as the f-number of the second lens (320).

[0106] Also, the f / # of the third lens (330) is Therefore, the spot diameter due to the diffraction limit of the third lens is It can be ㎛, and since both the spot diameter and the lens diameter due to the diffraction limit are proportional to Nm, there may be no noise due to the diffraction limit even if Nm is increased.

[0107] For example, if the cell spacing is 20 μm and N m If this is 5, then D = 100㎛ When it is 2, can be 10 days.

[0108] Therefore, when the diffraction limit spot diameter of the third lens (330) is 10㎛, since D is 100㎛, there may be no crosstalk between neighboring photodetectors (500).

[0109] On the other hand, the geometric optical aberration becomes important between the second lens (320) and the pixels of the spatial light modulator (400), and in the case of a system with f / # of 2, if three lenses are used, the geometric optical angle aberration is 3 m㎭, so when f2 is 3.3 mm, the geometric optical spot diameter is 10 ㎛, so it may exceed 50% of the input node spacing, causing serious crosstalk.

[0110] In addition, the geometric optical limit of the array of kernels is about 165 × 165, but kernels usually do not require a large array size and do not place restrictions on the size of the input or output arrays, so the parallel processing capacity of the optoelectronic module (10) can be greatly increased.

[0111] Figure 8 is a multi-output conceptual diagram of an optoelectronic module (10) using a spatial light modulator (400) according to one embodiment of the present invention.

[0112] The optoelectronic module (10) according to the present invention can be applied in cascade, and can further include a plurality of first photodetectors (500, 521, 522, 523) and a plurality of second photodetectors (511, 512, 513, 514) whose output values ​​from the preceding layer are connected to the inputs of the subsequent layer, a plurality of first electron processing (600, 621, 622, 623) and a plurality of second electron processing units (611, 612, 613, 614), and a third light source (721, 723, 725, 727) and a fourth light source (722, 724, 726, 728).

[0113] For example, the output values ​​of the first photodetector (521) and the second photodetector (512) are respectively connected to the first electronic processing unit (621) and the second electronic processing unit (612), and the first electronic processing unit (621) and the second electronic processing unit (612) can be connected.

[0114] Additionally, the output values ​​of the first photodetector (521) and the second photodetector (512) can be connected together to either the first electronic processing unit (621) or the second electronic processing unit (612).

[0115] The structure of this embodiment provides the advantage of being able to perform various operations using the values ​​of two photodetectors, so that light corresponding to the positive (+) term of the weight can be collected in the second photodetector (511, 512, 513, 514), light corresponding to the negative (-) term can be collected in the first photodetector (500, 521, 522, 523), and these two results can be subtracted to perform calculations.

[0116] for example, class is positive Assuming this is negative, , , , , , The weights of the spatial light modulator (400) can be set as follows, and the outputs of the two detectors are each and It could be.

[0117] The result of subtraction in the second electronic processing unit (612) is as shown in mathematical expression 2.

[0118] [Equation 2]

[0119]

[0120]

[0121] Additionally, the second electronic processing unit (612) calculates the bias and sigmoid function together with the subtraction. can be obtained.

[0122] In this way, the light rays corresponding to the (+) term have their weights set to the original model values, the light rays corresponding to the (-) term have their weights set to 0, and then only the light rays corresponding to the (+) term can be collected and added in the second light detector (512).

[0123] Additionally, the rays corresponding to the (-) term can be collected and sent to the first photodetector (521) to be added, thereby performing general neural network-related operations including negative weights.

[0124] At this time, each electronic processing unit (600, 621, 622, 623, 611, 612, 613, 614) can use the output of a plurality of adjacent photodetectors (500, 521, 522, 523, 511, 512, 513, 514) as input, and can implement various functions other than subtraction through electronic circuits.

[0125] The components according to the present invention are defined by functional distinctions rather than physical distinctions, and can be defined by the functions each component performs. Each component may be implemented as hardware or as program code and processing units that perform each function, and the functions of two or more components may be implemented by including them in a single component. Therefore, it should be noted that the names given to the components in the following embodiments are not intended to physically distinguish each component, but rather to suggest the representative functions performed by each component, and that the technical spirit of the present invention is not limited by the names of the components.

[0126] An optoelectronic module using a spatial light modulator according to one embodiment of the present invention described above is included in an optical computer.

[0127] Although various embodiments of the present invention have been illustrated and described above, the present invention is not limited to the specific embodiments described above, and various modifications can be made by a person having ordinary skill in the art to which the present invention pertains without departing from the gist of the present invention as claimed in the claims, and such modifications should not be understood individually from the technical idea or prospect of the present invention.

[0128] [Explanation of symbols]

[0129] 10: Optoelectronic module

[0130] 100: Substrate

[0131] 200: First light source

[0132] 300: Lens section

[0133] 310: First lens

[0134] 320, 321: Second lens

[0135] 330: Third lens

[0136] 350, 351: Rays

[0137] 400: Spatial Light Modulator

[0138] 500: Photodetector

[0139] 600: Electronic Processing Unit

[0140] 700: Second light source

[0141] 800: Second board

Claims

1. A plurality of first light sources formed at a certain distance from each other; A plurality of first lenses formed to correspond to each of the plurality of first light sources and reducing the divergence angle of light emitted from the corresponding first light sources; A plurality of spatial light modulators each formed corresponding to the plurality of first lenses and controlling the intensity of light passing through the corresponding first lenses through respective pixels having preset weights; A plurality of photodetectors formed spaced apart from the plurality of spatial light modulators and obtaining a current value according to the intensity of light; and A lens unit formed between the plurality of spatial light modulators and the plurality of photodetectors, and collecting light rays irradiated from different first light sources passing through pixels at different relative positions of each spatial light modulator onto one photodetector; The above lens part, A plurality of second lenses formed corresponding to each of the plurality of spatial light modulators and concentrating light rays into one point; and An optoelectronic module using a spatial light modulator, which includes a plurality of third lenses corresponding to each of the second lenses and which pass light passing through the second lens.

2. In paragraph 1, Each spatial light modulator and the second lens corresponding to each spatial light modulator are formed to be spaced apart by a first distance, and each photodetector is formed to be spaced apart by a second distance from the third lens. An optoelectronic module using a spatial light modulator, wherein the second distance satisfies a relationship in which the distance at which the image of the pixel of the spatial light modulator is formed on the photodetector does not deviate by more than a certain error value.

3. In paragraph 1, The above lens part, An optoelectronic module using a spatial light modulator, which implements a function of calculating a convolution of an array of light sources and an array of pixels of a spatial light modulator by using a function of collecting light emitted from different first light sources, which passes through pixels at different relative positions of each spatial light modulator, onto a single light detector, formed between a plurality of spatial light modulators and a plurality of photodetectors.

4. In paragraph 1, An optoelectronic module using a spatial light modulator, wherein a plurality of photodetectors corresponding to a third lens or positioned within the third lens aperture are provided, and each photodetector performs a different convolution operation on light passing through a spatial light modulator pixel at a different relative position.

5. An optical computer including an optoelectronic module using a spatial light modulator according to paragraph 1.

Citation Information

Patent Citations

  • Transition unit providing an optical output signal

    EP1256905A1

  • Thermoelectric element

    KR1020220013223A

  • Movable Marker System for Drone Delivery, and Operation Method Thereof

    KR1020240129348A

  • Device for point of care testing

    KR1020240150933A

  • Optical processing system

    US11200487B1