Distributed optical fiber sensing event identification method and device based on optical calculation

Abstract

The invention relates to a distributed optical fiber sensing event identification method and device based on optical calculation, and relates to the technical field of optical fiber sensing event identification. Two-dimensional time-frequency signals of distributed sensing information are collected through a coherent detection type phi-OTDR system, and accurate recognition of sensing events is achieved on an optical path through optical calculation. The optical calculation comprises an optical convolution operation and an optical full connection operation, a two-dimensional time-frequency signal is processed into a one-dimensional tensor, and the neural network weight is loaded to the light through the cooperation of a light intensity amplitude modulator and a programmable multi-wavelength laser array, so that the optical neural network calculation is completed, and finally the corresponding light intensity of four wavelengths is obtained. Accurate identification of sensing events is realized. Sensing signals are processed through optical calculation, dependence on electronic calculation resources is avoided, and the response speed and the anti-interference capacity of a sensing system are improved; meanwhile, through distributed processing, the fault tolerance and reliability of the system are enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of optical fiber sensing event recognition, and in particular to a distributed optical fiber sensing event recognition method and device based on optical computing. Background Art

[0002] With the accelerated development of industrialization and urbanization, higher requirements are put forward for environmental safety, public safety and infrastructure health monitoring. Distributed fiber optic sensors, with their high sensitivity, strong anti-electromagnetic interference ability and suitability for harsh environments, have shown unique advantages in structural health monitoring, temperature sensing, pressure measurement and boundary intrusion detection. Especially in long-distance, large-scale or high-requirement electromagnetic compatibility application scenarios, their unique performance is irreplaceable. This type of system uses the principles of Rayleigh scattering and Brillouin scattering of optical fibers to monitor changes in physical parameters such as temperature, strain or vibration along the optical fiber path. Its highly sensitive sensing characteristics, coupled with good adaptability to harsh environmental conditions (such as high electromagnetic interference and corrosive environments), make distributed fiber optic sensors widely used in key areas such as oil and gas pipeline monitoring, geological disaster warning, and building health monitoring.

[0003] However, although traditional distributed fiber optic sensing systems can provide continuous monitoring information, their data processing capabilities are limited by the photoelectric conversion efficiency and electronic signal processing speed. In the complex signal analysis and event identification process, this bottleneck may lead to response delays and fail to meet the needs of real-time monitoring and rapid response.

[0004] The emergence of optical computing technology provides a possible solution to this challenge. Optical computing uses photons instead of electrons to perform computing operations. Its processing speed is much faster than electronic computing, and it can directly process optical signals without converting them into electrical signals. This technology can greatly improve data processing speed, reduce energy consumption, and reduce signal loss during the conversion process. More importantly, optical computing can perform parallel processing on the signal path, allowing simultaneous identification and location of multiple events, improving the system's multi-tasking capabilities. Distributed fiber optic sensing systems based on optical computing combine the high sensitivity of fiber optic sensors with the high-speed data processing capabilities of optical computing, enabling rapid acquisition, processing and identification of optical signals, and providing real-time and efficient monitoring results. This emerging technology is expected to overcome the limitations of traditional systems and bring revolutionary progress to real-time security monitoring and accurate event location. Therefore, researching and developing event recognition methods based on optical computing is an inevitable trend in the development of distributed fiber optic sensing technology. In addition, since traditional fiber optic sensing technology relies on photoelectric converters and electrical signal processing equipment, the transmission and processing of electrical signals will inevitably introduce noise, reduce system stability, and increase system energy consumption. In order to overcome these shortcomings and improve the monitoring performance and adaptability of fiber optic sensing systems, a new event recognition technology is urgently needed. Summary of the invention

[0005] The purpose of the present invention is to provide a distributed optical fiber sensing event identification method and device based on optical computing, aiming to provide an efficient, reliable and highly applicable optical fiber sensing event identification method that can achieve high-sensitivity monitoring and has high-speed signal processing capabilities to adapt to complex and changeable monitoring environments and respond to various emergencies, thereby ensuring the safety of people's lives and property and the stability of social operations.

[0006] To achieve the above object, the present invention provides the following solutions:

[0007] A distributed optical fiber sensing event recognition method based on optical computing, the method includes a distributed sensing information processing part, an optical convolution computing part and an optical full-connection computing part:

[0008] The distributed sensor information processing part includes:

[0009] (1) Obtain phase information along the length of the optical fiber through a fiber sensing system (such as φ-OTDR) and demodulate it into a two-dimensional time-frequency signal. The amplitude of the phase change can be mapped to the brightness of the grayscale image, where the grayscale level represents the phase change. Then, the image signal of the C channel is output and the image signal containing the sensing event information is cropped into multiple C-channel H×W images;

[0010] (2) An improved YOLOv8 micro-detection framework with an attention mechanism (Swin-Transformer) and an efficient detection head (Shape-IoU) is used to locate and crop sensitive areas containing sensor events in image signals. The model input image size is [C1, H1, W1], and the output image size is [n, C2, H2, W2], where n is the number of cropped events.

[0011] (3) Convert the image of size [C2, H2, W2] into a single-channel grayscale image, and reduce the size of each grayscale image to 36×36 through bilinear interpolation.

[0012] The optical convolution calculation part includes:

[0013] (1) According to the convolution kernel size and step size requirements, the 36×36 two-dimensional image signal is expanded into a one-dimensional tensor;

[0014] (2) The one-dimensional tensor with an effective bit of 8 bits is sent to the microprocessor. The raw data is converted into a corresponding voltage waveform through an arbitrary waveform generator and drives the light intensity amplitude modulator.

[0015] (3) sending the value of the convolution kernel and the corresponding weight to the microprocessor, which modulates the programmable multi-wavelength laser array by generating corresponding electrical signals;

[0016] (4) Multiplexing light of different wavelengths into a single optical fiber and outputting it to an optical intensity amplitude modulator. The optical signal loaded with the convolution kernel factor enters the modulator and is modulated;

[0017] (5) The optical switch selects the optical convolution channel and outputs the multiplied optical signal to the single-mode optical fiber. The single-mode optical fiber causes a one-bit time delay between the time domain waveforms of each wavelength. The intensity of light in different time domains is detected by the second photodetector, thereby completing the addition of the optical convolution.

[0018] (6) A multi-channel parallel data acquisition card collects the signal after optical convolution and outputs it to a computer to restore it into a feature map.

[0019] The optical full-connection computing part includes:

[0020] (1) sending the activated one-dimensional tensor to a microprocessor, which generates a corresponding time-domain waveform to control a light intensity amplitude modulator;

[0021] (2) sending the fully connected weights to a microprocessor, which generates a corresponding electrical signal to modulate the programmable multi-wavelength laser array, and controls the microprocessor to synchronously send the flattened one-dimensional tensor to an arbitrary waveform generator, which generates a corresponding time domain waveform to an intensity amplitude modulator;

[0022] (3) Multiplexing light of different wavelengths into a single optical fiber and outputting it to an optical intensity amplitude modulator, which modulates the multiplexed light and loads a one-dimensional tensor signal onto light of different wavelengths, thus achieving optical full-connection weight loading.

[0023] (4) The optical switch selects the optical fully connected channel, outputs the weighted optical signal to the wavelength demultiplexer, separates the light of multiple wavelengths, detects the intensity of light of different wavelengths, and converts it into electrical signals and outputs it to the computer. By accumulating over time, the intensity of different wavelengths of light is obtained.

[0024] (5) According to the different light wavelength intensities, the sensing events are classified. The sensing events are: background, knocking, walking, and fence. The wavelength of the programmable multi-wavelength laser array is 4 bands. In the fully connected terminal, the photodetector receives the light intensity at different times. The light of different wavelengths is accumulated in the time domain and classified to obtain the corresponding four light intensities. The four light intensities are normalized and uniformly mapped to the range of 0% to 100%. The higher the light intensity, the higher the probability of the corresponding sensing event. The band with the highest light intensity corresponds to the prediction result.

[0025] A distributed optical fiber sensing event recognition device based on optical computing can implement a distributed optical fiber sensing event recognition method based on optical computing; the device specifically comprises: a narrow linewidth laser, a first optical coupler, an acousto-optic modulator, an erbium-doped optical fiber amplifier, a circulator, a sensing optical fiber, a second optical coupler, a first photodetector, an acousto-optic modulation driver, a data acquisition card, a computer, a programmable multi-wavelength laser array, a wavelength multiplexer, an arbitrary waveform generator, an optical intensity amplitude modulator, an optical switch, a first semiconductor optical amplifier, a wavelength demultiplexer, a photodetector array, a second semiconductor optical amplifier, a second photodetector, a multi-channel parallel data acquisition card and a microprocessor.

[0026] Narrow linewidth lasers are used to emit narrow linewidth, high-correlation continuous lasers;

[0027] The first optical coupler is used to split the continuous laser light input by the narrow line width laser into two paths, one path is a detection light signal, and the other path is a reference light signal;

[0028] The acousto-optic modulator is modulated by the acousto-optic modulation driver and is used to modulate the continuous laser light input by the first optical coupler into a pulsed detection light;

[0029] The erbium-doped fiber amplifier is used to amplify the pulsed detection light input by the acousto-optic modulator;

[0030] The loop is used to input the pulsed detection light amplified by the erbium-doped fiber amplifier into the sensing fiber and receive the back-scattered Rayleigh light reflected by the sensing fiber;

[0031] The second optical coupler is used for coupling the reference optical signal input by the first optical coupler and the backscattered Rayleigh light input by the loop;

[0032] The first photodetector is used for detecting the coupled light input by the second optical coupler and converting the coupled light into an interference electrical signal;

[0033] The data acquisition card is used to collect the interference electrical signal of the first photodetector, and convert the interference electrical signal into a digital signal, and output it to the computer; the data acquisition card is also used to output a clock carrier signal and a pulse trigger modulation signal to the acousto-optic modulation driver, so that the acousto-optic modulator outputs a high-level pulse modulation signal to the acousto-optic modulator;

[0034] The computer is used to control the working state of the data acquisition card and analyze and process the digital signals collected by the data acquisition card;

[0035] A microprocessor is controlled by a computer and is used to send clock signals and weights to the programmable multi-wavelength laser array;

[0036] The programmable multi-wavelength laser array is used to output lasers with different wavelength weights;

[0037] The wavelength multiplexer is used to multiplex multiple lasers of different wavelengths into a single optical fiber and output them to the optical intensity amplitude modulator;

[0038] The arbitrary waveform generator is controlled by a microprocessor and is used to convert the time domain waveform signal generated by the microprocessor into a high-speed waveform;

[0039] The optical intensity amplitude modulator is used to load the high-speed waveform generated by the arbitrary waveform generator onto each wavelength in a single optical fiber, complete the loading of the time domain signal, and output it to the optical switch;

[0040] The optical switch switches the optical convolution calculation and the optical full connection calculation according to the clock signal sent by the microprocessor, and outputs to the first semiconductor optical amplifier or the single-mode optical fiber under different electrical signals;

[0041] The single-mode optical fiber is used to generate a time delay of one bit between the time domain waveforms of each wavelength, and output to the second semiconductor amplifier;

[0042] The second semiconductor amplifier is used to amplify the optical signal that needs to be subjected to optical convolution calculation;

[0043] The second photodetector is used to detect the time domain intensity of the optical signal amplified by the second semiconductor amplifier, and output the total power of all wavelengths of light in each time interval, and convert the total power of all wavelengths of light in each time interval into an electrical signal;

[0044] The first semiconductor optical amplifier is used to amplify the optical signal that needs to perform optical full-connection computing;

[0045] The wavelength demultiplexer is used to demultiplex the optical signal that needs to be fully connected optically, separate the light of multiple different wavelengths, and convert the light of multiple different wavelengths into electrical signals;

[0046] The photodetector array is used to detect the intensity of light of multiple different wavelengths and convert the intensity of light of multiple different wavelengths into electrical signals;

[0047] When performing optical convolution calculation, the multi-channel parallel data acquisition card is used to receive the electrical signal of the second photodetector; when performing optical full connection calculation, the multi-channel parallel data acquisition card is used to receive the electrical signal of the photodetector array and output it to the computer;

[0048] The computer is also used to receive the electrical signals collected by the multi-channel parallel data acquisition card, convert the electrical signals into two-dimensional time-frequency signals, convert the two-dimensional time-frequency signals into one-dimensional tensors, and restore the one-dimensional signals after optical convolution calculation into two-dimensional time-frequency signals, output the judgment results of distributed optical fiber sensing events: background, knocking, walking, fence, output clock signals, weights and one-dimensional tensors to the microprocessor;

[0049] The microprocessor is also used to convert the one-dimensional tensor into a time domain waveform signal and output it to the arbitrary waveform generator.

[0050] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects: by collecting the two-dimensional time-frequency signal of distributed sensor information, optical computing is used on the optical path to realize accurate identification of sensor events. By processing sensor signals through optical computing, the dependence on electronic computing resources is avoided, the response speed and anti-interference ability of the sensor system are improved, and it has significant performance advantages when processing large-scale sensor data; through distributed processing, the fault tolerance and reliability of the system are enhanced; at the same time, the tensor acceleration technology is introduced to further improve the calculation speed of distributed sensor signals, and it has the advantages of high recognition accuracy and large data throughput. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0052] Figure 1 This is a system structure diagram provided by the present invention.

[0053] Figure 2 This is the classification event involved in the present invention.

[0054] Figure 3Schematic diagram of the present invention converting a two-dimensional signal of a distributed optical fiber sensing event into a one-dimensional tensor.

[0055] Figure 4 It is a schematic diagram of optical full-connection calculation of the present invention.

[0056] Figure 5 It is a specific Dropout implementation schematic diagram provided by the present invention. DETAILED DESCRIPTION

[0057] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0058] The purpose of the present invention is to provide a distributed optical fiber sensing event recognition method and device based on optical computing, which can perform optical computing on the optical path to achieve accurate recognition of distributed optical fiber sensing events.

[0059] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0060] Figure 1 The present invention provides a system schematic diagram of a distributed optical fiber sensing event identification device based on optical computing. The system of the device is divided into a coherent detection type φ-OTDR subsystem and an optical computing subsystem. The coherent detection type φ-OTDR subsystem specifically includes: a narrow linewidth laser, a first optical coupler, an acousto-optic modulator, an erbium-doped fiber amplifier, a circulator, a sensing fiber, a second optical coupler, a first photodetector, an acousto-optic modulation driver, a data acquisition card and a computer. The optical computing subsystem specifically includes: a programmable multi-wavelength laser array, a wavelength multiplexer, an arbitrary waveform generator, an optical intensity amplitude modulator, an optical switch, a first semiconductor optical amplifier, a wavelength demultiplexer, a photodetector array, a second semiconductor optical amplifier, a second photodetector, a multi-channel parallel data acquisition card, a microprocessor and a computer.

[0061] Narrow linewidth lasers are used to emit narrow linewidth, high-correlation continuous lasers.

[0062] The first optical coupler is used to split the continuous laser light input by the narrow line width laser into two paths, one path is a detection light signal, and the other path is a reference light signal.

[0063] The acousto-optic modulator is modulated by the acousto-optic modulation driver and is used to modulate the continuous laser light input by the first optical coupler into a pulsed detection light;

[0064] The erbium-doped fiber amplifier is used to amplify the pulsed detection light input by the acousto-optic modulator.

[0065] The loop is used to input the pulse detection light amplified by the erbium-doped fiber amplifier into the sensing fiber and receive the back Rayleigh scattered light reflected by the sensing fiber.

[0066] The second optical coupler is used for coupling the reference optical signal input by the first optical coupler and the backscattered Rayleigh light input by the loop.

[0067] The first photodetector is used for detecting the coupled light input by the second optical coupler and converting the coupled light into an interference electrical signal.

[0068] The data acquisition card is used to collect the interference electrical signal of the first photodetector, and convert the interference electrical signal into a digital signal and output it to the computer. The data acquisition card is also used to output a clock carrier signal and a pulse trigger modulation signal to the acousto-optic modulation driver, so that the acousto-optic modulator outputs a high-level pulse modulation signal to the acousto-optic modulator.

[0069] The computer is used to control the working state of the data acquisition card and analyze and process the digital signals collected by the data acquisition card.

[0070] A microprocessor is controlled by a computer and is used to send clock signals and weights to the programmable multi-wavelength laser array.

[0071] The programmable multi-wavelength laser array is used to output multi-wavelength lasers with different weights.

[0072] The wavelength multiplexer is used to multiplex multiple lasers of different wavelengths into a single optical fiber and output them to the optical intensity amplitude modulator.

[0073] The arbitrary waveform generator is controlled by a microprocessor and is used to convert the time domain waveform signal generated by the microprocessor into a high-speed waveform.

[0074] The optical intensity amplitude modulator is used to load the high-speed waveform generated by the arbitrary waveform generator onto each wavelength in a single optical fiber, complete the loading of the time domain signal, and output it to the optical switch.

[0075] The optical switch switches the optical convolution calculation and the optical full connection calculation according to the clock signal sent by the microprocessor, and outputs to the first semiconductor optical amplifier or the single-mode optical fiber under different electrical signals.

[0076] The single-mode optical fiber is used to generate a time delay of one bit between the time domain waveforms of each wavelength, which are output to the second semiconductor amplifier.

[0077] The second semiconductor amplifier is used to amplify the optical signal that needs to be subjected to optical convolution calculation.

[0078] The second photodetector is used to detect the time domain intensity of the optical signal amplified by the second semiconductor amplifier, and output the total power of all wavelengths of light in each time interval, and convert the total power of all wavelengths of light in each time interval into an electrical signal.

[0079] The first semiconductor optical amplifier is used to amplify the optical signal that needs to perform optical full-connection computing.

[0080] The wavelength demultiplexer is used to demultiplex the optical signals that need to be fully connected for optical computing, separate the multiple lights of different wavelengths, and convert the multiple lights of different wavelengths into electrical signals.

[0081] The photodetector array is used to detect the intensity of light of multiple different wavelengths and convert the intensity of light of multiple different wavelengths into electrical signals.

[0082] When performing optical convolution calculation, the multi-channel parallel data acquisition card is used to receive the electrical signal of the second photodetector. When performing optical full connection calculation, the multi-channel parallel data acquisition card is used to receive the electrical signal of the photodetector array and output it to the computer.

[0083] The computer is also used to receive electrical signals collected by multi-channel parallel data acquisition cards, convert the electrical signals into two-dimensional time-frequency signals, convert the two-dimensional time-frequency signals into one-dimensional tensors, and restore the one-dimensional signals after optical convolution calculation into two-dimensional time-frequency signals, and output the judgment results of distributed optical fiber sensing events: background, knocking, walking, fence, output clock signals, weights and one-dimensional tensors to the microprocessor.

[0084] The microprocessor is also used to convert the one-dimensional tensor into a time domain waveform signal and output it to the arbitrary waveform generator.

[0085] Based on the above-mentioned distributed optical fiber sensing event recognition device based on optical computing, the corresponding recognition method includes the following steps:

[0086] Step 1: The narrow linewidth laser emits narrow linewidth, high correlation continuous light to the first optical coupler, and the optical signal is divided into two parts (the splitting ratio is 90:10); 90% of the light is used as sensing light and output to the acousto-optic modulator, and 10% of the light is used as reference light and output to the second optical coupler.

[0087] Step 2: The data acquisition card sends a signal to the acousto-optic modulation driver, which sends a high-level pulse modulation signal to control the acousto-optic modulator. The acousto-optic modulator modulates the sensing light, and the modulated light is power-amplified by the erbium-doped fiber amplifier and output to the loop. The sensing light is sensed and reflected to the second optical coupler (with a split ratio of 50:50) to couple with 10% of the reference light.

[0088] Step 3: The light coupled by the second optical coupler contains sensor information, which is output to the first photodetector and converted into an electrical signal. After being processed by the data acquisition card, it is sent to the computer. The computer converts the sensor signal into a 3-channel two-dimensional time-frequency signal. The event categories include background, knocking, walking, and fence. The two-dimensional time-frequency signals of some events are as follows: Figure 2 shown.

[0089] Step 4: Expand the two-dimensional time-frequency signal into a one-dimensional tensor and input it into the microprocessor. The microprocessor controls the programmable multi-wavelength laser array to emit optical signals of different wavelengths. The optical signals are multiplexed into a single optical fiber through the wavelength multiplexer. The microprocessor sends an electrical signal to control the arbitrary waveform transmitter to generate a time domain waveform corresponding to the one-dimensional tensor to modulate the light intensity amplitude modulator.

[0090] Step 5: The microprocessor controls the optical switch to switch the optical convolution calculation channel. The optical signal after optical convolution is output to the second semiconductor optical amplifier through the single-mode optical fiber, and then output to the second photodetector to be converted into an electrical signal. The microprocessor controls the optical switch to switch the optical full-connection calculation channel. The optical signal after optical full-connection calculation is output to the wavelength demultiplexer through the first semiconductor optical amplifier, and then output to the photodetector array to be converted into an electrical signal.

[0091] Step 6: The multi-channel parallel data acquisition card collects the electrical signals of the second photodetector and the photodetector array and outputs them to the computer, which processes the data.

[0092] In one embodiment, a distributed fiber optic sensing event recognition method based on optical computing includes a distributed sensing information processing part, an optical convolution computing part, and an optical full-connection computing part.

[0093] The distributed sensor information processing part includes:

[0094] (1) Obtain the sensing signal of the sensing event and demodulate the sensing signal into a two-dimensional time-frequency signal. The grayscale value of each pixel in the grayscale image corresponds to the phase change measured by the optical fiber sensor at a specific position. Areas with large phase changes can be marked by higher or lower grayscale values ​​to form a clear contrast in the image. Then, a 3-channel image signal is output and the image signal containing the sensing information is cropped into multiple 3-channel 128×128 images.

[0095] (2) The YOLOv8 neural network with improved backbone and detection head is loaded with pre-trained weights to locate and crop the areas containing sensor events in the image signal. The backbone part uses the Swin-Transformer module to replace the original traditional convolution and other operations to achieve feature extraction. At the same time, the window attention mechanism has efficiency advantages over the traditional ViT. The detection head part introduces shape-IoU as a regression loss function. By considering the influence of the shape and angle of the bounding box itself on the detection accuracy, the model input image size is [3, 750, 750], and the output image size is [n, 3, 128, 128], where n is the number of cropped events.

[0096] (3) Convert the 3-channel 128×128 image into a single-channel grayscale image, and reduce the size of each grayscale image to 36×36 by bilinear interpolation.

[0097] Specifically, in the process of distributed fiber optic sensor signal positioning and cropping, the backbone uses the Swin-Transformer module to replace the original traditional convolution and other operations to achieve feature extraction. At the same time, the window attention mechanism has an efficiency advantage over the traditional ViT. It first splits the input [H, W, 3] 3-channel 750×750 image into several equal-sized (P×P) but non-overlapping [N0, P, P, 3] sub-images (Ni represents the number of sub-images in the i-th stage, the number of original sub-images ) and flattened into a sequence of [N0, (P2×3)] as a token sequence. Then in the first stage, this token sequence tensor is projected into a new token tensor of shape [N0, C] through a linear layer and enters the Swin-Transformer module for two feature extractions while keeping the output shape unchanged. The first-level feature extraction is completed. In the second stage, the merging layer is used to splice adjacent 2×2 sub-images into new sub-images as the second-level features. The token tensor changes to Similar to the second stage, after the sub-images are reassembled, the third and fourth stages perform 6 and 2 Swin-Transformer module feature extractions respectively to complete the third and fourth level feature extractions, and the token tensor changes to [N3,4C] Through these four stages, the hierarchical extraction of image features is completed while ensuring a certain computational efficiency.

[0098] Considering that the bounding box regression is more accurate and fully includes the scope of the entire event, a bounding box regression method focusing on the shape and scale of the bounding box itself is used during cropping. First, the GT bounding box regression characteristics are analyzed, and it is concluded that the bounding box's own shape factors and scale factors will affect the regression results. Among them, the Shape-IoU method can calculate the loss by focusing on the bounding box's own shape and scale, thereby making the predicted bounding box regression more accurate and better locating the fiber sensing event. Then, with this half as the center, cropping is performed. After introducing the Center-based methods (based on the center point) of Anchor-Free, the model changes from outputting "anchor box size offset offest" to "predicting the distance between the left, top, right, and bottom bounding boxes of the target box and the target center point (ltrb = left, top, right, bottom)". In order to cooperate with Anchor-Free and improve generalization, DFL loss is added in v8. DFL optimizes the probability of the two positions closest to the label y on the left and right in the form of cross entropy, so that the network can focus on the target position and the distribution of the neighboring area faster. The model input image size is [C1, H1, W1], and the output image size is [n, C2, H2, W2], where n is the number of events.

[0099] After completing the distributed sensing information processing, the optical convolution calculation part can be started. The specific steps are as follows:

[0100] (1) According to the convolution kernel size and step size requirements (the convolution kernel size is 2×2 and the step size is 1), the 36×36 two-dimensional image signal is expanded into a one-dimensional tensor for data preparation for subsequent one-dimensional convolution calculations. The size of the expanded one-dimensional tensor is 1×5476, as shown in Figure 3 As shown. Specifically, the input image is traversed, and a local region is extracted at each position of the image with the stride size as the step size; each local region is expanded into a column vector and arranged in the columns of the output matrix. When expanding, it can be in the order of row priority or column priority; each column of the final output matrix corresponds to a local region in the input image, and the entire output matrix contains all possible local regions in the input image.

[0101] (2) The computer sends the expanded one-dimensional sensor signal and the expanded convolution kernel to the microprocessor, which sends an electrical signal to control the arbitrary waveform generator to generate the corresponding time domain waveform. The microprocessor sends an electrical signal to control the programmable multi-wavelength laser array to generate four wavelengths of laser. The power of each wavelength corresponds to the convolution kernel, and then it is multiplexed into a single optical fiber through a wavelength multiplexer.

[0102] (3) sending the value of the convolution kernel and the corresponding weight to the microprocessor, which modulates the programmable multi-wavelength laser array by generating corresponding electrical signals;

[0103] (4) The arbitrary waveform generator generates a corresponding time domain waveform to modulate the light intensity amplitude modulator to modulate the optical signal to complete the multiplication part of the optical convolution. The rate of the arbitrary waveform generator is 12.5 Gbps, the vertical resolution is 10 bits, the bandwidth is 5.3 GHz, and the corresponding time to send one bit is 80 ps.

[0104] (5) The microprocessor sends an electrical signal to control the optical switch to switch the optical convolution channel. The switching speed of the optical switch is 10 ms. The multiplied optical signal is output to the single-mode optical fiber. The single-mode optical fiber causes a time delay of one bit between the time domain waveforms of each wavelength. The dispersion coefficient of the single-mode optical fiber is 20 ps / (nm·km). The distance of the single-mode optical fiber is 2 km, and the wavelength change is 2 nm, that is, the corresponding phase delay is 80 ps, ​​corresponding to one bit of data. The intensity of light in different time domains is then detected by the second photodetector. The bandwidth of the second photodetector is 40 GHz, thereby completing the addition of the optical convolution.

[0105] (6) The multi-channel parallel data acquisition card collects the signal after optical convolution and outputs it to the computer. When the misaligned vectorized data is accumulated and calculated at the photodetector, redundant data will be generated. The effective information is extracted by the computer, and the final result is restored to the feature map. Specifically, the input column vector is traversed, each column vector is restored to the local area in the original image according to the corresponding position, and each local area is filled back to the corresponding position of the original image. The final output is the restored original image, which is then activated by the Relu layer to increase the nonlinearity of the model.

[0106] After completing the optical convolution calculation part, you can start the optical full connection calculation part. The specific steps are as follows:

[0107] (1) The activated one-dimensional tensor is sent to a microprocessor, which generates an electrical signal to control an arbitrary waveform generator, and the arbitrary waveform generator generates a corresponding time domain waveform to control a light intensity amplitude modulator.

[0108] (2) The fully connected weights are sent to the microprocessor, which generates corresponding electrical signals to modulate the programmable multi-wavelength laser array. The microprocessor is controlled to synchronously send the flattened one-dimensional tensor to the arbitrary waveform generator, and the arbitrary waveform generator generates the corresponding time domain waveform to the light intensity amplitude modulator; the modulation rate of the programmable multi-wavelength laser array is 12.5 Gbps, and the weights are loaded onto different wavelengths.

[0109] (3) Light of different wavelengths is multiplexed into a single optical fiber and output to an optical intensity amplitude modulator. The optical intensity amplitude modulator modulates the multiplexed light to load a one-dimensional tensor signal onto light of different wavelengths, thereby achieving optical full-connection weight loading.

[0110] (4) The optical switch selects the optical fully connected channel and outputs the weighted optical signal to the wavelength demultiplexer to separate multiple wavelengths of light. The bandwidth of each photodetector in the photodetector array is 40 GHz, so as to detect the intensity of light of different wavelengths and convert it into electrical signals to be output to the computer. The intensity of different wavelengths of light is obtained by temporal accumulation.

[0111] When performing optical full-connection computing, combined with the Dropout operation, the events of distributed optical fiber sensing are: background, knocking, walking, and fence, with corresponding outputs Y1, Y2, Y3, and Y4, such as Figure 4 As shown, assuming that the output is each neuron, the output can be expressed by the following formula:

[0112] Y i =WX+b

[0113] Where X is the input of the current layer, W is the pre-trained weight matrix, and b is the bias vector. When performing optical full connection calculations, the above formula is modified to obtain

[0114] Y i =W'X

[0115]

[0116] It can be concluded from the above formula that when realizing optical full-connection computing on the optical path, it is only necessary to load weights on the programmable multi-wavelength laser array, corresponding to the customized intensity output of each wavelength on the programmable multi-wavelength laser array, and finally output different light intensities for each wavelength. The Dropout operation is introduced, as shown in 5. When the electrical signal modulates the arbitrary waveform generator, the input characteristic time domain waveform can be randomly modulated and set to zero, simulating the effect of the neuron output being set to zero during the Dropout process, thereby reducing overfitting.

[0117] (5) According to the intensity of different light wavelengths, the sensing events are classified into: background, knocking, walking, and fence. The wavelength of the programmable multi-wavelength laser array is 4 bands. The higher the light intensity, the higher the probability of the corresponding sensing event. The band with the highest light intensity corresponds to the prediction result.

[0118] Specifically, in the optical fully connected computing part, it involves converting electronic signals into optical signals and using optical signals for data processing and classification, including the following steps:

[0119] S1. Convert the fully connected weights into electrical signals: The weights of the fully connected layer need to be sent to the microprocessor, which generates corresponding electrical signals based on the received weight data.

[0120] S2. Electrical signal modulated laser array: The microprocessor uses electrical signals to modulate the programmable multi-wavelength laser array. Each laser emits a light signal of a specific wavelength. At the same time, the microprocessor sends the flattened one-dimensional tensor data to the arbitrary waveform generator, which generates the corresponding time domain waveform to the light intensity amplitude modulator for further modulating the light signal.

[0121] S3. Multiplexing and modulating optical signals: Use wavelength division multiplexing technology to combine optical signals of different wavelengths and transmit them to a single optical fiber. The optical signal is further modulated when passing through the optical intensity amplitude modulator to load the one-dimensional tensor signal onto light of different wavelengths, that is, load the corresponding fully connected layer weights on each wavelength.

[0122] S4. Wavelength demultiplexing and signal detection: The weighted optical signal is transmitted to the wavelength demultiplexer through optical fiber. The demultiplexer separates the multiplexed optical signal into different wavelengths. The light intensity of each wavelength is converted into an electrical signal through a photodetector and output to a computer. The computer performs temporal accumulation processing on the light intensity of different wavelengths to obtain the total light intensity information of each wavelength.

[0123] S5. Classification and prediction: According to the light intensity of different wavelengths, the computer classifies the sensing events, including: background, knocking, walking, and fence; according to the preset mode, the higher the light intensity band, the higher the probability of the corresponding event occurring.

[0124] Specifically, in the optical fully connected computing part, for discrete systems, the intensity and routing of multiple wavelengths are controlled through a programmable multi-wavelength laser array, and the update rate of the weight on the programmable multi-wavelength laser array is consistent with the modulator rate to ensure clock synchronization; for on-chip integrated systems, an optical intensity amplitude modulator is used to load the fully connected weights onto the time domain waveform of a one-dimensional tensor signal to complete the loading of the weights of each wavelength.

[0125] The present invention processes sensor signals through optical computing, thus avoiding dependence on electronic computing resources and improving the response speed and anti-interference ability of the sensor system; through distributed processing, the fault tolerance and reliability of the system are enhanced; the tensor acceleration technology is introduced to further improve the calculation speed of distributed sensor signals, and has the advantages of high recognition accuracy and large data throughput.

[0126] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0127] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A distributed fiber optic sensing event recognition method based on optical computing, It is characterized in that The method includes a distributed sensing information processing part, an optical convolution calculation part and an optical full connection calculation part: The distributed sensor information processing part includes: (1) Obtaining phase information along the length of the optical fiber through the optical fiber sensing system and demodulating it into a two-dimensional time-frequency signal. The amplitude of the phase change is mapped to the brightness of the grayscale image, where the grayscale level represents the phase change. The image signal of the C channel is output, and the image signal containing the sensing information is divided into multiple C-channel H×W images; (2) An improved YOLOv8 micro-detection framework with an attention mechanism and an efficient detection head is used to locate and crop sensitive areas containing sensor events in image signals. The image size of the model input is [C 1 , H 1 , W 1 ], the output image size is [n, C 2 , H 2 , W 2 ], where n is the number of clipping events; (3) Its size is [C 2 , H 2 , W 2 ]The image is converted into a single-channel grayscale image, and the size of each grayscale image is reduced to 36×36 by bilinear interpolation; The optical convolution calculation part includes: (1) According to the convolution kernel size and step size requirements, the 36×36 two-dimensional image signal is expanded into a one-dimensional tensor; (2) The one-dimensional tensor is sent to the microprocessor, and the raw data is converted into a corresponding voltage waveform through an arbitrary waveform generator, and drives the light intensity amplitude modulator; (3) sending the value of the convolution kernel and the corresponding weight to the microprocessor, which modulates the programmable multi-wavelength laser array by generating corresponding electrical signals; (4) Multiplexing light of different wavelengths into a single optical fiber and outputting it to an optical intensity amplitude modulator. The optical signal loaded with the convolution kernel factor enters the modulator and is modulated; (5) The optical switch selects the optical convolution channel and outputs the multiplied optical signal to the single-mode optical fiber. The single-mode optical fiber causes a one-bit time delay between the time domain waveforms of each wavelength. The intensity of light in different time domains is detected by the second photodetector, thereby completing the addition of the optical convolution. (6) A multi-channel parallel data acquisition card collects the signal after optical convolution and outputs it to a computer to restore it into a feature map; The optical full-connection computing part includes: (1) sending the activated one-dimensional tensor to a microprocessor, which generates a corresponding time-domain waveform to control a light intensity amplitude modulator; (2) sending the fully connected weights to a microprocessor, which generates a corresponding electrical signal to modulate the programmable multi-wavelength laser array, and controls the microprocessor to synchronously send the flattened one-dimensional tensor to an arbitrary waveform generator, which generates a corresponding time domain waveform to an intensity amplitude modulator; (3) Multiplexing light of different wavelengths into a single optical fiber and outputting it to an optical intensity amplitude modulator, which modulates the multiplexed light and loads a one-dimensional tensor signal onto light of different wavelengths, thus achieving optical full-connection weight loading. (4) The optical switch selects the optical fully connected channel, outputs the weighted optical signal to the wavelength demultiplexer, separates the light of multiple wavelengths, detects the intensity of light of different wavelengths, and converts it into electrical signals and outputs it to the computer. By accumulating over time, the intensity of different wavelengths of light is obtained. (5) According to the intensity of different light wavelengths, the sensing events are classified into: background, knocking, walking and fence. The wavelength of the programmable multi-wavelength laser array is 4 bands. The higher the light intensity, the higher the probability of the corresponding sensing event. The band with the highest light intensity corresponds to the predicted result.

2. The distributed optical fiber sensing event recognition method based on optical computing according to claim 1, It is characterized in that In the optical convolution calculation part, expanding the two-dimensional image signal into a one-dimensional tensor specifically includes: S1, traverse the input image, with the stride size as the step size, and extract a local area at each position of the image; S2, expand each local area into a column vector and arrange them in columns of the output matrix; S3, output matrix; the entire output matrix contains all possible local regions in the input image, and each column of the matrix corresponds to a local region in the input image.

3. The distributed optical fiber sensing event recognition method based on optical computing according to claim 1, It is characterized in that In the optical convolution calculation part, the restoration to the feature map specifically includes: S1, traverse the input column vectors and restore each column vector to a local area in the original image according to the corresponding position; S2, filling each local area back to the corresponding position of the original image; S3. Output image, which is the restored original image.

4. The distributed optical fiber sensing event recognition method based on optical computing according to claim 1, It is characterized in that In the optical fully connected computing part, it involves converting electronic signals into optical signals and using optical signals for data processing and classification, including the following steps: S1. Convert the weight of the fully connected layer into an electrical signal: The weight of the fully connected layer needs to be sent to the microprocessor, which generates a corresponding electrical signal based on the received weight data; S2. Electrical signal modulated laser array: The microprocessor uses electrical signals to modulate the programmable multi-wavelength laser array. Each laser emits a light signal of a specific wavelength. At the same time, the microprocessor sends the flattened one-dimensional tensor data to the arbitrary waveform generator, which generates the corresponding time domain waveform to the light intensity amplitude modulator for further modulating the light signal. S3. Multiplexing and modulating optical signals: Use wavelength division multiplexing technology to combine optical signals of different wavelengths and transmit them to a single optical fiber. The optical signal is further modulated when passing through the optical intensity amplitude modulator to load the one-dimensional tensor signal onto light of different wavelengths, that is, load the corresponding fully connected layer weights on each wavelength. S4, wavelength demultiplexing and signal detection: The optical signal after weight loading is transmitted to the wavelength demultiplexer through the optical fiber. The demultiplexer separates the multiplexed optical signal into different wavelengths. The light intensity of each wavelength is converted into an electrical signal through the optical detector and output to the computer. The computer performs temporal accumulation processing on the light intensity of different wavelengths to obtain the total light intensity information of each wavelength. S5. Classification and prediction: According to the light intensity of different wavelengths, the computer classifies the sensing events, including: background, knocking, walking and fence; according to the preset mode, the higher the light intensity band, the higher the probability of the corresponding event occurring.

5. The distributed optical fiber sensing event recognition method based on optical computing according to claim 1, It is characterized in that In the optical fully connected computing part, combined with the Dropout operation, the events of distributed optical fiber sensing are: background, knocking, walking and fence, and the corresponding output Y 1 , Y 2 , Y 3 , Y 4 , assuming that the output is each neuron, the output can be expressed by the following formula: Y i =WX+b Where X is the input of the current layer, W is the pre-trained weight matrix, and b is the bias vector. When performing optical full connection calculations, Y i =W'X When implementing optical fully connected computing on the optical path, it is only necessary to load weights on the programmable multi-wavelength laser array, corresponding to the customized intensity output of each wavelength on the programmable multi-wavelength laser array, and finally output different light intensities for each wavelength.

6. The distributed optical fiber sensing event recognition method based on optical computing according to claim 1, It is characterized in that In the optical fully connected computing part, for discrete systems, the intensity and routing of multiple wavelengths are controlled through a programmable multi-wavelength laser array, and the update rate of the weight on the programmable multi-wavelength laser array is consistent with the modulator rate to ensure clock synchronization; For the on-chip integrated system, an optical intensity amplitude modulator is used to load the fully connected weights onto the time domain waveform of the one-dimensional tensor signal to complete the loading of the weights for each wavelength.

7. The distributed optical fiber sensing event recognition method based on optical computing according to claim 1, It is characterized in that The input image size of the improved YOLOv8 micro-detection framework is [3, 750, 750], and the output image size after positioning and cropping is [n, 3, 128, 128].

8. A distributed optical fiber sensing event recognition device based on optical computing, It is characterized in that The device can implement the method described in any one of claims 1 to 7; the device specifically comprises: a narrow linewidth laser, a first optical coupler, an acousto-optic modulator, an erbium-doped fiber amplifier, a circulator, a sensing fiber, a second optical coupler, a first photodetector, an acousto-optic modulation driver, a data acquisition card, a computer, a programmable multi-wavelength laser array, a wavelength multiplexer, an arbitrary waveform generator, an intensity amplitude modulator, an optical switch, a first semiconductor optical amplifier, a wavelength demultiplexer, a photodetector array, a second semiconductor optical amplifier, a second photodetector, a multi-channel parallel data acquisition card and a microprocessor; The narrow linewidth laser is used to emit narrow linewidth, high-correlation continuous laser; The first optical coupler is used to split the continuous laser light input by the narrow linewidth laser into two paths, one path is a detection light signal, and the other path is a reference light signal; The acousto-optic modulator is modulated by the acousto-optic modulation driver and is used to modulate the continuous laser light input by the first optical coupler into a pulsed detection light; The erbium-doped fiber amplifier is used to amplify the pulsed detection light input by the acousto-optic modulator; The loop is used to input the pulsed detection light amplified by the erbium-doped fiber amplifier into the sensing fiber, and receive the back Rayleigh scattered light reflected by the sensing fiber; The second optical coupler is used for coupling the reference optical signal input by the first optical coupler and the backscattered Rayleigh light input by the loop; The first photodetector is used for detecting the coupled light input by the second optical coupler and converting the coupled light into an interference electrical signal; The data acquisition card is used to collect the interference electrical signal of the first photodetector, and convert the interference electrical signal into a digital signal, and output it to a computer; the data acquisition card is also used to output a clock carrier signal and a pulse trigger modulation signal to the acousto-optic modulation driver, so that the acousto-optic modulator outputs a high-level pulse modulation signal to the acousto-optic modulator; The computer is used to control the working state of the data acquisition card and analyze and process the digital signals collected by the data acquisition card; The microprocessor is controlled by the computer and is used to send clock signals and weights to the programmable multi-wavelength laser array; The programmable multi-wavelength laser array is used to output lasers with different wavelength weights; The wavelength multiplexer is used to multiplex a plurality of lasers of different wavelengths into a single optical fiber and output them to the light intensity amplitude modulator; The arbitrary waveform generator is controlled by the microprocessor and is used to convert the time domain waveform signal generated by the microprocessor into a high-speed waveform; The optical intensity amplitude modulator is used to load the high-speed waveform generated by the arbitrary waveform generator onto each wavelength in a single optical fiber, complete the loading of the time domain signal, and output it to the optical switch; The optical switch switches the optical convolution calculation and the optical full connection calculation according to the clock signal sent by the microprocessor, and outputs to the first semiconductor optical amplifier or the single-mode optical fiber under different electrical signals; The single-mode optical fiber is used to generate a time delay of one bit between the time domain waveforms of each wavelength, and output it to the second semiconductor amplifier; The second semiconductor amplifier is used to amplify the optical signal that needs to be subjected to optical convolution calculation; The second photodetector is used to detect the time domain intensity of the optical signal amplified by the second semiconductor amplifier, and output the total power of all wavelengths of light in each time interval, and convert the total power of all wavelengths of light in each time interval into an electrical signal; The first semiconductor optical amplifier is used to amplify the optical signal that needs to be subjected to optical full-connection computing; The wavelength demultiplexer is used to demultiplex the optical signal that needs to be subjected to optical full-connection computing, separate a plurality of lights of different wavelengths, and convert the plurality of lights of different wavelengths into electrical signals; The photodetector array is used to detect the intensity of the multiple lights of different wavelengths and convert the intensity of the multiple lights of different wavelengths into electrical signals; When performing optical convolution calculation, the multi-channel parallel data acquisition card is used to receive the electrical signal of the second photodetector; when performing optical full connection calculation, the multi-channel parallel data acquisition card is used to receive the electrical signal of the photodetector array and output it to the computer; The computer is also used to receive the electrical signals collected by the multi-channel parallel data acquisition card, convert the electrical signals into two-dimensional time-frequency signals, convert the two-dimensional time-frequency signals into one-dimensional tensors, and restore the one-dimensional signals after optical convolution calculation into two-dimensional time-frequency signals, output the judgment results of distributed optical fiber sensing events: background, knocking, walking, fence, and output the clock signal, weight and one-dimensional tensor to the microprocessor; The microprocessor is also used to convert the one-dimensional tensor into a time domain waveform signal and output it to the arbitrary waveform generator.

9. The distributed optical fiber sensing event recognition device based on optical computing according to claim 8, It is characterized in that The rate of the arbitrary waveform generator is 12.5 Gbps, the vertical resolution is 10 bits, the bandwidth is 5.3 GHz, and the corresponding time to send one bit is 80 ps.

10. The distributed optical fiber sensing event recognition device based on optical computing according to claim 8, It is characterized in that The dispersion coefficient of the single-mode optical fiber is 20 ps / (nm·km), the distance is 2 km, and the wavelength variation is 2 nm.

Images