Unscented kalman filter based phi-otdr data real-time processing system and method

Abstract

The application discloses a kind of Φ-OTDR data real-time processing system and method based on unscented Kalman filter.The system includes modulated signal generation module, data acquisition module, data combination reading module, phase calculation module, data transmission module, phase unpacking module and communication module.The method is: three adjacent pulsed light produces Rayleigh backscattering into unbalanced Michelson interferometer, and modulation voltage is applied to PZT piezoelectric ceramic, so that three-way interference RBS has additional phase;signal acquisition module acquires three-way light intensity signal and stores into data combination reading module, and reads the light intensity signal of corresponding detection sequence;Phase calculation module calculates the corresponding wrapped phase, and inputs data to unscented Kalman filter module for phase unpacking;Communication module sends the phase information unpacked to host computer.The application realizes the data real-time processing of Φ-OTDR system, and improves the accuracy and noise immunity of phase unpacking.

Description

Technical Field

[0001] This invention relates to the fields of optical sensing technology and FPGA signal processing technology, and in particular to a real-time data processing system and method based on unscented Kalman filtering and Φ-OTDR. Background Technology

[0002] Distributed Acoustic Sensor (DAS) is a sensing system based on Rayleigh scattering that detects sound waves by measuring the change in optical phase caused by changes in the axial strain of an optical fiber. In recent years, DAS performance has developed rapidly, and it has been widely used due to its ability to reconstruct frequency and phase information at any location on the optical fiber. Phase-sensitive optical time-domain reflectometer (Φ-OTDR) is an important area of ​​DAS systems, achieving distributed sensing by detecting backscattered Rayleigh light (RBS). Compared to traditional optical time-domain reflectometer systems, Φ-OTDR uses a highly coherent laser as the light source, resulting in higher sensitivity and responsivity. This technology has been widely applied in perimeter security inspection, structural health monitoring, and underwater sound source localization.

[0003] However, fiber optic sensing emphasizes real-time performance and accuracy. As sensing distance and sampling rate increase, the requirements for data processing speed and unpacking accuracy also increase accordingly. The correctness, real-time performance, and noise immunity of phase unpacking remain difficult problems to solve. Kalman filtering is widely used in signal processing, robot control, unmanned vehicles, and drones due to its strong filtering effect and strong noise immunity. However, the matrix multiplication involved can only be applied to linear systems. In the process of phase unpacking, complex trigonometric function calculations are involved, and standard Kalman filtering cannot be applied. Summary of the Invention

[0004] The purpose of this invention is to provide a real-time Φ-OTDR data processing system and method with high processing speed, high phase unpacking accuracy, and low hardware cost.

[0005] The technical solution to achieve the purpose of this invention is: a real-time data processing system based on unscented Kalman filtering, comprising a modulation signal generation module, a data acquisition module, a data combination and reading module, a phase calculation module, a data transmission module, a phase unpacking module, and a communication module;

[0006] The modulation signal generation module applies a signal to the PZT piezoelectric ceramic through the FPGA, so that the back Rayleigh scattering light RBS after the three-way interference has an additional phase of 2π / 3, -2π / 3, and 0, respectively.

[0007] The data acquisition module uses an analog-to-digital converter (ADC) to acquire Rayleigh backscattered light intensity information;

[0008] The data combination reading module stores light intensity data according to the number of channels and reads the corresponding detection sequence data;

[0009] The phase calculation module uses three light intensity data to calculate the corresponding envelope phase;

[0010] The data transmission module utilizes a high-speed FIFO to achieve data transmission between modules.

[0011] The phase unpacking module performs one-dimensional phase unpacking on the obtained packaged phase data;

[0012] The communication module transmits the unpacked data to the host computer.

[0013] Furthermore, the data combination reading module includes multiple storage units for storing data; the storage units employ dual-port RAM to support parallel read and write operations.

[0014] Furthermore, the data combination and reading module employs a control program written in Verilog to control the storage and retrieval of data.

[0015] Furthermore, the phase calculation module includes multiple logic units for performing data operations; the logic units employ pipeline technology.

[0016] A real-time data processing method for Φ-OTDR based on unscented Kalman filtering, employing the aforementioned real-time data processing system for Φ-OTDR, specifically includes the following steps:

[0017] Step 1: The modulation signal generation module converts the continuous light from the narrow linewidth laser into pulsed light through the acousto-optic modulator AOM. At the same time, the FPGA applies a modulation voltage to the PZT piezoelectric ceramic on the optical fiber to obtain an interference light intensity signal with phase modulation.

[0018] Step 2: The data acquisition module uses an analog-to-digital converter (ADC) to acquire the light intensity signal obtained in Step 1. The sampling rate corresponds to the sensing distance.

[0019] Step 3: The data combination reading module uses three dual-port RAMs to store the multi-channel light intensity data obtained in Step 2, corresponding to the three light intensity information channels. The control program controls the reading of the corresponding light intensity sequence. After reading, the light intensity data in the dual-port RAM is replaced with the light intensity data of the next moment of the channel.

[0020] Step 4: Transfer the light intensity data from Step 3 to the phase calculation module and use the phase-shifting arctangent algorithm to solve the wrapping phase;

[0021] Step 5: Transmit the packaged phase data obtained in Step 4 to the phase unpacking module through the data transmission module;

[0022] Step 6: The phase unpacking module performs one-dimensional phase unpacking using UKF.

[0023] Step 7: The communication module transmits the unpacked phase information to the host computer.

[0024] Furthermore, step 1 is specifically as follows:

[0025] Adding PZT piezoelectric ceramic to one arm of an unbalanced Michelson interferometer, and applying voltage to the PZT piezoelectric ceramic, causes the three Rayleigh backs after interference to have an additional modulated phase.

[0026] Furthermore, step 2 is specifically as follows:

[0027] Rayleigh backscattered light (RBS) from the two arms of the Michelson interferometer is interfered with by the coupler and then received by the data acquisition module, which consists of a photodetector (PD), an analog-to-digital converter (ADC), and an FPGA.

[0028] Furthermore, step 3 is specifically as follows:

[0029] Step 3.1: Store the data corresponding to all channels of the first light intensity in RAM1, store the data of the second light intensity in RAM2, and store the data of the third light intensity in RAM3.

[0030] Step 3.2: Use the control program to read the three-channel light intensity data;

[0031] Step 3.3: Replace the light intensity data in the dual-port RAM with the light intensity data of the next moment of the channel. Overwrite the data in RAM1 one by one according to the address. At this time, RAM2 and RAM3 still retain the original data. Then read the three light intensities, which is another detection sequence. Repeat the above operation to achieve continuous storage and reading.

[0032] Furthermore, step 4 is specifically as follows:

[0033] The three light intensity signals are added or subtracted and then used as the numerator and denominator to be input into the phase calculation module for arctangent operation to obtain the corresponding encapsulated phase. The first 1000 data points are taken and exported as a text document through Vivado, and then the image is drawn using Python.

[0034] Furthermore, step 6 is specifically as follows:

[0035] The initial parameters of the unscented Kalman filter are stored in the corresponding parameter storage RAM. The initial detection sequence calls the initial parameters. The control module stores the corresponding channel filter parameters in the corresponding address and calls them when the next set of detection sequences in the same channel arrives. The filter module adopts a pipeline design. The input signal is the sine and cosine information of the phase plus the initial filter parameters or the updated filter parameters. The output value of the unpacking module is also exported through Vivado and the curve is plotted using Python.

[0036] Compared with the prior art, the significant advantages of this invention are: (1) the overall data processing flow is completed by FPGA and the pipeline is designed, which improves the data processing speed; (2) one-dimensional phase unpacking is performed by unscented Kalman filtering, which improves the phase unpacking accuracy; (3) the design module reuse saves hardware resources and reduces hardware costs. Attached Figure Description

[0037] Figure 1 This is a schematic diagram of the structure of a Φ-OTDR real-time data processing system based on unscented Kalman filtering according to the present invention.

[0038] Figure 2 This is a flowchart illustrating a real-time data processing method for Φ-OTDR based on unscented Kalman filtering according to the present invention.

[0039] Figure 3 This is a schematic diagram of the encapsulation phase obtained by taking the first 1000 points after applying sinusoidal modulation to the sensing distance of 278m corresponding to channel 139 in an embodiment of the present invention.

[0040] Figure 4 This is an embodiment of the present invention. Figure 3 The result of further one-dimensional phase unwrapping of the wrapped phase results is shown in the figure. Detailed Implementation

[0041] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0042] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0043] Combination Figure 1 The present invention discloses a real-time data processing system for Φ-OTDR based on unscented Kalman filtering, comprising a modulation signal generation module, a data acquisition module, a data combination and reading module, a phase calculation module, a data transmission module, a phase unpacking module, and a communication module;

[0044] The modulation signal generation module applies a signal to the PZT piezoelectric ceramic through the FPGA, so that the back Rayleigh scattering light RBS after the three-way interference has an additional phase of 2π / 3, -2π / 3, and 0, respectively.

[0045] The data acquisition module uses an analog-to-digital converter (ADC) to acquire Rayleigh backscattered light intensity information;

[0046] The data combination reading module stores light intensity data according to the number of channels and reads the corresponding detection sequence data;

[0047] The phase calculation module uses three light intensity data to calculate the corresponding envelope phase;

[0048] The data transmission module utilizes a high-speed FIFO to achieve data transmission between modules.

[0049] The phase unpacking module performs one-dimensional phase unpacking on the obtained packaged phase data;

[0050] The communication module transmits the unpacked data to the host computer.

[0051] As a specific example, the data combination reading module includes multiple storage units for storing data; the storage units employ dual-port RAM to support parallel read and write operations.

[0052] As a specific example, the data combination reading module uses a control program written in Verilog to control the storage and reading of data.

[0053] As a specific example, the phase calculation module includes multiple logic units for performing data operations; the logic units employ pipeline technology to improve data processing speed.

[0054] Combination Figure 2 A real-time data processing method based on unscented Kalman filtering for Φ-OTDR includes the following steps:

[0055] Step 1: The modulation signal generation module converts the continuous light from the narrow linewidth laser into pulsed light through the acousto-optic modulator AOM. At the same time, the FPGA applies a modulation voltage to the PZT piezoelectric ceramic on the optical fiber to obtain an interference light intensity signal with phase modulation.

[0056] Step 2: The data acquisition module uses an analog-to-digital converter (ADC) to acquire the light intensity signal obtained in Step 1. The sampling rate corresponds to the sensing distance.

[0057] Step 3: The data combination reading module uses three dual-port RAMs to store the multi-channel light intensity data obtained in Step 2, corresponding to the three light intensity information channels. The control program controls the reading of the corresponding light intensity sequence. After reading, the light intensity data in the dual-port RAM is replaced with the light intensity data of the next moment of the channel.

[0058] Step 4: Transfer the light intensity data from Step 3 to the phase calculation module and use the phase-shifting arctangent algorithm to solve the wrapping phase;

[0059] Step 5: Transmit the packaged phase data obtained in Step 4 to the phase unpacking module through the data transmission module;

[0060] Step 6: The phase unpacking module performs one-dimensional phase unpacking using UKF.

[0061] Step 7: The communication module transmits the unpacked phase information to the host computer.

[0062] As a specific example, the application of a modulation voltage to the PZT piezoelectric ceramic on the optical fiber in step 1 to obtain an interferometric light intensity signal with phase modulation is as follows:

[0063] Adding PZT piezoelectric ceramic to one arm of an unbalanced Michelson interferometer, and applying voltage to the PZT piezoelectric ceramic, causes the three Rayleigh backs after interference to have an additional modulated phase.

[0064] As a specific example, the data acquisition module in step 2 uses an analog-to-digital converter (ADC) to acquire the light intensity signal obtained in step 1. The sampling rate corresponds to the sensing distance, as detailed below:

[0065] Rayleigh backscattered light (RBS) from the two arms of the Michelson interferometer is interfered with by the coupler and then received by the data acquisition module, which consists of a photodetector (PD), an analog-to-digital converter (ADC), and an FPGA.

[0066] As a specific example, the data combination reading module described in step 3 uses three dual-port RAMs to store the multi-channel light intensity data obtained in step 2, corresponding to three light intensity information channels. It also uses a control program to control the reading of the corresponding light intensity sequence. After reading, the light intensity data in the dual-port RAMs is replaced with the light intensity data of the next moment for that channel, as detailed below:

[0067] Step 3.1: Store the data corresponding to all channels of the first light intensity in RAM1, store the data of the second light intensity in RAM2, and store the data of the third light intensity in RAM3.

[0068] Step 3.2: Use the control program to read the three-channel light intensity data;

[0069] Step 3.3: Replace the light intensity data in the dual-port RAM with the light intensity data of the next moment of the channel. Overwrite the data in RAM1 one by one according to the address. At this time, RAM2 and RAM3 still retain the original data. Then read the three light intensities, which is another detection sequence. Repeat the above operation to achieve continuous storage and reading.

[0070] As a specific example, the method for solving the wrapping phase using the phase-shifting arctangent algorithm described in step 4 is as follows:

[0071] The three light intensity signals are added or subtracted and then used as the numerator and denominator to be input into the phase calculation module for arctangent operation to obtain the corresponding encapsulated phase. The first 1000 data points are taken and exported as a text document through Vivado, and then the image is drawn using Python.

[0072] As a specific example, the phase unpacking module described in step 6 performs one-dimensional phase unpacking of the wrapped phase using UKF, as detailed below:

[0073] The initial parameters of the unscented Kalman filter are stored in the corresponding parameter storage RAM. The initial detection sequence calls the initial parameters. The control module stores the corresponding channel filter parameters in the corresponding address and calls them when the next set of detection sequences in the same channel arrives. The filter module adopts a pipeline design. The input signal is the sine and cosine information of the phase plus the initial filter parameters or the updated filter parameters. The output value of the unpacking module is also exported through Vivado and the curve is plotted using Python.

[0074] Example

[0075] Combination Figure 1 This embodiment provides a real-time data processing system for Φ-OTDR based on unscented Kalman filtering, including a modulation signal generation module, a data acquisition module, a data combination and reading module, a phase calculation module, a data transmission module, a phase unpacking module, and a communication module.

[0076] The modulation signal generation module applies a signal to the PZT piezoelectric ceramic through the FPGA, so that the back Rayleigh scattering light RBS after the three-way interference has an additional phase of 2π / 3, -2π / 3, and 0, respectively.

[0077] The data acquisition module uses an analog-to-digital converter (ADC) to acquire Rayleigh backscattered light intensity information;

[0078] The data combination reading module stores light intensity data according to the number of channels and reads the corresponding detection sequence data;

[0079] The phase calculation module uses three light intensity data to calculate the corresponding envelope phase;

[0080] The data transmission module utilizes a high-speed FIFO to achieve data transmission between modules.

[0081] The phase unpacking module performs one-dimensional phase unpacking on the obtained packaged phase data;

[0082] The communication module transmits the unpacked data to the host computer.

[0083] As a specific example, the data combination reading module includes multiple storage units for storing data; the storage units employ dual-port RAM to support parallel read and write operations.

[0084] As a specific example, the data combination reading module uses a control program written in Verilog to control the storage and reading of data.

[0085] As a specific example, the phase calculation module includes multiple logic units for performing data operations; the logic units employ pipeline technology to improve data processing speed.

[0086] Combination Figure 2 This embodiment provides a real-time data processing method for Φ-OTDR based on unscented Kalman filtering, which includes the following steps:

[0087] Step 1: The modulation signal generation module converts the continuous light from the narrow linewidth laser into pulsed light through the acousto-optic modulator AOM. The system pulse light frequency is 20kHz. A sinusoidal voltage signal is applied to the piezoelectric ceramic at the corresponding sensing distance of 278m in channel 139 of the optical fiber. The three light intensities have modulation phases of 2π / 3, 0, and -2π / 3, respectively.

[0088] Step 2: The data acquisition module uses an analog-to-digital converter (ADC) to acquire the light intensity signal obtained in step 1, with a sampling rate of 50Mb / s, corresponding to acquiring one data point every 2m along the fiber length.

[0089] Step 3: The data combination reading module uses three dual-port RAMs to store the multi-channel light intensity data obtained in Step 2, corresponding to the three light intensity information channels. The control program controls the reading of the corresponding light intensity sequence. After reading, the light intensity data in the dual-port RAM is replaced with the light intensity data of the next moment of the channel.

[0090] Step 4: Transfer the light intensity data I1, I2, and I3 from Step 3 to the phase calculation modules RAM1, RAM2, and RAM3. Simultaneously, read the three data streams corresponding to the detection sequence, perform point-to-point processing on the data, calculate the encapsulated phase using the phase calculation modules, and extract the first 1000 data points. Export the extracted data as a text document using Vivado, and then use Python to draw an image, such as... Figure 3 As shown;

[0091] Depend on Figure 3 It can be seen that the result of the wrapping phase calculation process performed by the present invention conforms to the wrapping phase interval of [-π, π].

[0092] Step 5: Calculate the sine and cosine values ​​of the wrapped phase obtained in Step 5, and transmit them to the unscented Kalman filter one-dimensional phase unpacking module through a high-speed FIFO.

[0093] Step 6: The phase unpacking module performs one-dimensional phase unpacking using UKF.

[0094] The initial parameters of the unscented Kalman filter are stored in the corresponding parameter storage RAM. The initial detection sequence calls these initial parameters. The control module stores the corresponding channel's filter parameters in the corresponding address, which are then called when the next detection sequence arrives on the same channel. The filter module uses a pipeline design; the input signal is a sine and cosine signal encapsulating the phase, plus the initial or updated filter parameters. The output value of the unpacking module is also exported via Vivado, and curves are plotted using Python. The results are shown below. Figure 4 As shown.

[0095] Depend on Figure 4 It can be seen that the sinusoidal signal applied to the corresponding point of the optical fiber is well reproduced. The Φ-OTDR real-time data processing system and method based on unscented Kalman filtering provided by this invention can perform high-precision phase unpacking when processing multi-channel data simultaneously.

[0096] Step 7: The communication module transmits the unpacked phase information to the host computer.

[0097] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention specification, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A Φ-OTDR data real-time processing system based on unscented Kalman filter, characterized in that, The modulation signal generation module, the data acquisition module, the data combination reading module, the phase calculation module, the data transmission module, the phase unwrapping module and the communication module are included. The modulation signal generation module applies a signal to the PZT piezoelectric ceramic through the FPGA, so that the back Rayleigh scattering light RBS after three-way interference respectively has an extra phase of 2π / 3, -2π / 3 and 0. The data acquisition module acquires the Rayleigh backscattering light intensity information by using an analog-to-digital converter ADC. The data combination reading module stores the light intensity data according to the channel number and reads the corresponding detection sequence data. The phase calculation module calculates the corresponding wrapped phase by using three-way light intensity data. The data transmission module uses a high-speed FIFO to realize data transmission between modules. The phase unwrapping module performs one-dimensional phase unwrapping on the obtained wrapped phase data by UKF. The communication module transmits the unwrapped data to the upper computer.

2. The real-time Φ-OTDR data processing system based on unscented Kalman filter according to claim 1, characterized in that, The data combination reading module includes a plurality of storage units for storing data; the storage unit adopts a dual-port RAM to support parallel read and write operations.

3. The real-time Φ-OTDR data processing system based on unscented Kalman filter according to claim 1, characterized in that, The data combination reading module adopts a control program written based on verilog to control the storage and reading of data.

4. The real-time Φ-OTDR data processing system based on unscented Kalman filter according to claim 1, characterized in that, The phase calculation module includes a plurality of logic units for performing data operations; the logic unit adopts a pipeline technology.

5. A real-time processing method of Φ-OTDR data based on unscented Kalman filter, characterized in that, The method adopts the Φ-OTDR data real-time processing system of any one of claims 1-4, and specifically includes the following steps: Step 1: The modulation signal generation module converts the continuous light from the narrow linewidth laser into pulsed light through the acousto-optic modulator AOM, and at the same time, applies a modulation voltage to the PZT piezoelectric ceramic on the optical fiber through the FPGA to obtain the light intensity signal after interference with phase modulation; Step 2: The data acquisition module acquires the light intensity signal obtained in step 1 by using an analog-to-digital converter ADC, and the sampling rate corresponds to the sensing distance; Step 3: The data combination reading module stores the multi-channel light intensity data obtained in step 2 by using three dual-port RAMs, corresponding to three-way light intensity information, and controls the reading of the corresponding light intensity sequence by using a control program; after reading, the light intensity data in the dual-port RAM is replaced with the light intensity data at the next moment of the channel; Step 4: The light intensity data of step 3 is transmitted to the phase calculation module, and the wrapped phase is solved by using the phase shift-based arctangent algorithm; Step 5: The wrapped phase data obtained in step 4 is transmitted to the phase unwrapping module through the data transmission module; Step 6: The phase unwrapping module unwraps the wrapped phase by UKF in one dimension; Step 7: The communication module transmits the unwrapped phase information to the upper computer.

6. The real-time Φ-OTDR data processing method based on unscented Kalman filter according to claim 5, characterized in that, The step 1 is specifically as follows: A PZT piezoelectric ceramic is added to one arm of the unbalanced Michelson interferometer, and an extra modulation phase is added to the three-way Rayleigh backscattering light after interference by applying a voltage to the PZT piezoelectric ceramic.

7. The real-time Φ-OTDR data processing method based on unscented Kalman filter according to claim 5, characterized in that, The step 2 is specifically as follows: The Rayleigh backscattering light RBS of the two arms of the balanced Michelson interferometer is received by the data acquisition module after interference through the coupler; the data acquisition module is composed of a photodetector PD, an analog-to-digital converter ADC and an FPGA.

8. The real-time Φ-OTDR data processing method based on unscented Kalman filter according to claim 5, characterized in that, The step 3 is specifically as follows: Step 3.1, store the data of the first light intensity of all channels in the RAM1, the data of the second light intensity in the RAM2, and the data of the third light intensity in the RAM3; Step 3.2, control the reading of the three light intensity data by using the control program; Step 3.3, replace the light intensity data in the dual-port RAM with the light intensity data at the next moment of the channel, and cover the data in the RAM1 by address one by one, at this time, the RAM2 and the RAM3 still keep the original data, at this time, read the three light intensity, that is, another detection sequence, and cycle the above operation to realize continuous storage and reading.

9. The real-time Φ-OTDR data processing method based on unscented Kalman filter according to claim 5, characterized in that, The step 4 is specifically as follows: After the three light intensity signals are added and subtracted, the obtained signals are input into the phase calculation module as the numerator and the denominator to perform the arctangent operation, and the corresponding wrapped phase is obtained, and the obtained data are taken as the first 1000 to export a text document by using vivado, and an image is drawn by using python.

10. The real-time Φ-OTDR data processing method based on unscented Kalman filter according to claim 5, characterized in that, The step 6 is specifically as follows: Store the initial parameters of the unscented Kalman filter into the corresponding parameter storage RAM, call the initial parameters by using the initial detection sequence, store the corresponding channel filtering parameters into the corresponding address by using the control module, and call the filtering parameters when the next group of detection sequences arrive under the same channel; The filtering module adopts the pipeline design, the input signal is the sine and cosine information of the wrapped phase plus the initial filtering parameters or the updated filtering parameters, and the output value of the unpacking module is also exported by using vivado, and the curve is drawn by using python.

Images