Phase method optical frequency domain reflectometer denoising method based on Hampel filter

Abstract

The invention provides a phase method optical frequency domain reflectometer denoising method based on an HAMPEL filter. The denoising method comprises the steps that two sets of beat frequency signals of a phase method optical frequency domain reflectometer are collected to serve as reference signals and measurement signals; respectively carrying out interpolation resampling on the reference signal and the measurement signal to realize nonlinear compensation of the signals; respectively carrying out FFT (Fast Fourier Transform) on the two groups of beat frequency signals after nonlinear compensation, and extracting phase information in each frequency spectrum; obtaining a relative phase of the reference signal and the measurement signal according to the phase information in each frequency spectrum, and performing unwrapping and differential processing on the relative phase to obtain a differential relative phase; and a Hampel filter is adopted to carry out de-noising processing on the differential relative phase, and phase method optical frequency domain reflection data de-noising is completed. According to the invention, the technical problem that the noise reduction demand and the high spatial resolution demand cannot be balanced by the phi-OFDR phase data noise reduction processing in the prior art can be solved.

Description

Technical Field

[0001] The invention belongs to the technical field of optical fiber sensing, and in particular relates to a denoising method for a phase-method optical frequency domain reflectometer based on a Hampel filter. Background Art

[0002] Phase-based optical frequency domain reflectometry (Φ-OFDR), a popular area of ​​distributed fiber-optic sensing technology, is widely used in fields such as nondestructive monitoring of precision instruments, military defense, and large-scale structural health monitoring. By measuring the phase of Fresnel reflection or Rayleigh backscattering (RBS) in an optical fiber, Φ-OFDR enables quantitative measurement of time-dependent parameters such as distance, temperature, and strain. Compared to other distributed sensing technologies, such as phase-based optical time-domain reflectometry (Φ-OTDR), Φ-OFDR offers superior performance in terms of signal-to-noise ratio, spatial resolution, and sensing accuracy. Because Φ-OFDR achieves high spatial resolution, it enables more accurate measurements of physical quantities such as temperature and strain. However, when using a sensor with a fiber Bragg grating (FBG) train as the test fiber, fading noise caused by the FBG-free regions of the train can significantly degrade the signal-to-noise ratio, compromising signal quality.

[0003] Currently, there are several ways to suppress fading noise for Φ-OFDR phase data:

[0004] (1) Median filter and sliding average technology. This method enhances the signal by improving the signal-to-noise ratio. However, this method reduces random noise through smoothing, and also smoothes out subtle changes in the signal, resulting in a decrease in the high-frequency part and edge clarity of the signal, thereby reducing spatial resolution.

[0005] (2) Statistical filters and SG-I digital differentiators. This approach approximates the signal derivative by calculating the difference between adjacent sample points. While this process reduces noise, it can also lead to the loss of subtle changes in the signal, especially in areas of rapid signal change. This loss of detail directly affects the system's ability to detect subtle changes, thereby reducing spatial resolution.

[0006] (3) SG filter. The SG filter is a low-pass filter that reduces high-frequency signals while blurring the high-frequency details of the signal. When high-frequency signal components are filtered out, the resolution of the signal is affected. The larger the SG filter window, the more obvious the smoothing effect, but the greater the loss of resolution.

[0007] Methods such as median filters, sliding average techniques, and statistical filters play an important role in improving the signal-to-noise ratio and reducing noise, but they also have certain limitations. While these techniques reduce random noise, they also smooth out subtle signal variations, resulting in a decrease in high-frequency components and edge clarity, which in turn reduces spatial resolution. This is a common signal processing challenge in distributed fiber optic sensing applications. In practical applications, these factors must be balanced according to specific detection requirements to achieve optimal detection results. Summary of the Invention

[0008] The present invention aims to solve at least one of the technical problems existing in the prior art.

[0009] The present invention provides a phase method optical frequency domain reflectometry denoising method based on a HAMPEL filter, the denoising method comprising:

[0010] Collect two sets of beat frequency signals of phase-based optical frequency domain reflectometry as reference signals and measurement signals;

[0011] Interpolation and resampling are performed on the reference signal and the measurement signal respectively to achieve nonlinear compensation of the signal;

[0012] Perform FFT transformation on the two sets of beat frequency signals after nonlinear compensation to extract the phase information in each spectrum;

[0013] The relative phase of the reference signal and the measurement signal is obtained according to the phase information in each spectrum, and the relative phase is unwound and differentiated to obtain the differential relative phase;

[0014] Hampel filter is used to denoise the differential relative phase and complete the denoising of the phase method optical frequency domain reflection data.

[0015] Furthermore, interpolation and resampling of the reference signal and the measurement signal are performed separately, specifically including: capturing the phase information of the beat signal through an auxiliary interferometer, extracting the instantaneous phase using Hilbert transform, performing coordinate transformation and interpolation processing on the collected beat signal, and compensating for the nonlinear effect of the tunable laser.

[0016] Furthermore, after FFT transformation, the initial beat frequency phase of the reference signal is for Where v0 is the initial optical frequency of the tunable laser, τ r is the delay of the reference signal, and γ is the frequency sweep rate.

[0017] Furthermore, after FFT transformation, the initial beat frequency phase of the measurement signal is for Among them, τ mTo measure the delay of the signal.

[0018] Furthermore, the Hampel filter is used to perform denoising on the differential relative phase, specifically including:

[0019] (1) Input the Hampel filter to accept the window size N and the standard deviation threshold k;

[0020] (2) Calculate the median m and local standard deviation σ of the data subset S in the current window;

[0021] (3) Compare the deviation of each data point in the data subset S from the median m. If the deviation of a data point from the median m is greater than k times the local standard deviation σ, the data point is considered an outlier and is replaced by the median m of the data subset S.

[0022] (4) Slide to the next window with the data set and repeat (2)-(3) until the denoising process of all data is completed.

[0023] Furthermore, the local standard deviation σ is calculated according to Get, where x i is the i-th data point in subset S, i = 1, 2, ... N.

[0024] The technical solution of the present invention provides a phase-based optical frequency domain reflectometer denoising method based on a Hampel filter. The method calculates the phase difference between the measurement signal and the reference signal, and performs unwrapping and differential processing to obtain the differential relative phase. During the signal processing process, there is significant fading noise in the signal. In order to eliminate the noise, the present invention further sets the parameters of the Hampel filter, including the window size and the standard deviation threshold, to perform effective denoising. The denoised data not only retains the spatial resolution, but also improves the accuracy of the measurement. Compared with the prior art, the technical solution of the present invention can solve the technical problem in the prior art that the Φ-OFDR phase data denoising process cannot balance the noise reduction requirements and the high spatial resolution requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The accompanying drawings are included to provide a further understanding of the embodiments of the present invention, constitute a part of the specification, illustrate the embodiments of the present invention, and together with the description, explain the principles of the present invention. Obviously, the drawings described below are only some embodiments of the present invention, and those skilled in the art can derive other drawings based on these drawings without inventive effort.

[0026] Figure 1 A schematic flow chart of a Hampel filter-based phase-based optical frequency domain reflectometry denoising method according to a specific embodiment of the present invention is shown;

[0027] Figure 2 shows a differential relative phase diagram provided according to a specific embodiment of the present invention;

[0028] Figure 3 FIG. 4 shows a filtered differential relative phase diagram provided according to a specific embodiment of the present invention. DETAILED DESCRIPTION

[0029] It should be noted that, in the absence of conflict, the embodiments in this application and the features in the embodiments can be combined with each other. The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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. The following description of at least one exemplary embodiment is actually only illustrative and is in no way intended to limit the present invention and its application or use. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0030] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0031] Unless otherwise specifically stated, the relative arrangement of the parts and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present invention. At the same time, it should be understood that, for ease of description, the sizes of the various parts shown in the accompanying drawings are not drawn according to actual proportional relationships. The technology, methods and equipment known to those of ordinary skill in the relevant art may not be discussed in detail, but in appropriate cases, the technology, methods and equipment should be considered as a part of the specification. In all examples shown and discussed here, any specific value should be interpreted as being merely exemplary, rather than as a limitation. Therefore, other examples of the exemplary embodiments can have different values. It should be noted that similar numbers and letters represent similar items in the following drawings, and therefore, once an item is defined in an accompanying drawing, it does not need to be further discussed in subsequent drawings.

[0032] like Figure 1 As shown, according to a specific embodiment of the present invention, a phase method optical frequency domain reflectometry denoising method based on a Hampel filter is provided, and the method specifically includes:

[0033] Collect two sets of beat frequency signals of phase-based optical frequency domain reflectometry as reference signals and measurement signals;

[0034] Interpolation and resampling are performed on the reference signal and the measurement signal respectively to achieve nonlinear compensation of the signal;

[0035] Perform FFT transformation on the two sets of beat frequency signals after nonlinear compensation to extract the phase information in each spectrum;

[0036] The relative phase of the reference signal and the measurement signal is obtained according to the phase information in each spectrum, and the relative phase is unwound and differentiated to obtain the differential relative phase;

[0037] Hampel filter is used to denoise the differential relative phase and complete the denoising of the phase method optical frequency domain reflection data.

[0038] This configuration provides a denoising method for phase-based optical frequency domain reflectometry based on a Hampel filter. This method calculates the phase difference between the measured and reference signals, performs unwrapping and differential processing, and obtains the differential relative phase. However, significant fading noise is present in the signal during signal processing. To eliminate this noise, the present invention further adjusts the parameters of the Hampel filter, including the window size and standard deviation threshold, for effective denoising. The denoised data not only retains spatial resolution but also improves measurement accuracy.

[0039] OFDR system local oscillator signal E r (t) can be expressed as:

[0040] E r (t) = E o exp[j(2πv0t+πγt 2 )]

[0041] Among them, E o is the reference light amplitude, v0 is the initial optical frequency of the tunable laser, γ is the sweep rate, and t is time.

[0042] The detection signal light field E at position z in the optical fiber to be tested s (t) can be expressed as:

[0043] E s (t) = E o exp[j(2πv0(t-τ z )+πγ(t-τ Z ) 2 )]

[0044] Among them, τ Z Indicates the delay at position z in the optical fiber under test.

[0045] After neglecting the DC term and the constant term, the beat frequency signal of the local oscillator signal and the detection signal is:

[0046] I(t)=|E S (t)+E r (t)| 2

[0047] =[E S (t)+E r (t)][E S (t)+E r (t)] *

[0048] =E0 2 exp[j(2πv0τ z -πγτ Z 2 +2πf beat t)]

[0049] Among them, f beat is the beat frequency of the local oscillator signal and the detection signal.

[0050] After performing FFT transformation on the beat frequency signal, the initial beat frequency phase related to the FUT delay distance can be directly obtained. Its phase expression is:

[0051]

[0052] When temperature acts on an optical fiber, it changes the time delay and the refractive index of the light. The temperature change can be demodulated through phase changes. Based on this, the present application proposes a phase method optical frequency domain reflectometry denoising method based on a Hampel filter.

[0053] First, in the present invention, two sets of beat frequency signals of the phase method optical frequency domain reflectometer are collected as reference signals I r (t) and the measured signal I s (t). In order to correct the nonlinearity in the data, an auxiliary interferometer is used to interpolate and resample the two sets of data to achieve nonlinear compensation.

[0054] Specifically, in response to the nonlinear changes caused by the tunable laser source during the scanning process, the phase information of the beat signal is first captured by an auxiliary interferometer, and then the instantaneous phase is extracted using the Hilbert transform. The collected beat signal is then subjected to coordinate transformation and interpolation processing to effectively compensate for the nonlinear effects of the tunable laser.

[0055] Furthermore, the two groups of beat frequency signals after nonlinear compensation are respectively subjected to FFT transformation to extract the phase information in each spectrum.

[0056] Specifically, after FFT transformation, the initial beat frequency phases of the reference signal and the measurement signal can be obtained as and

[0057]

[0058] Among them, τ r is the delay of the reference signal, τ m To measure the delay of the signal.

[0059] Furthermore, the relative phase is obtained according to the phase information in each spectrum, and an unwrapping process is performed on the relative phase.

[0060] Specifically, the initial beat frequency phase difference between the reference signal and the measurement signal obtained in the previous step is calculated to obtain the relative phase The relative phase may have a jump of 2π, so the present invention performs unwinding processing to make it continuous and easy to analyze. In order to better analyze the relationship between temperature and relative phase, the present invention performs unwinding processing on the relative phase. Perform differential processing. Based on this, the relationship between the differential relative phase and temperature can be obtained as follows:

[0061]

[0062] in, For differential processing, is the relative phase after unwinding, ΔT is the temperature change, n is the refractive index of the optical fiber, c is the speed of light in vacuum, is the thermo-optical coefficient, α L is the optical fiber thermal expansion coefficient, and Δz is the system spatial resolution. Through experimental calibration, the linear coefficient of the differential relative phase and temperature change is 0.009035 rad / °C.

[0063] Furthermore, a Hampel filter is used to perform denoising on the differential relative phase to complete the denoising of the phase method optical frequency domain reflection data.

[0064] Due to the presence of fading noise in the signal, the present invention uses a Hampel filter to denoise the obtained differential relative phase, accepting a window size N and a standard deviation threshold k as input. Within each window, the filter first calculates the median m of the data subset S. For the data subset S, the local standard deviation σ is as follows:

[0065]

[0066] Among them, x i is the i-th data point in the subset S, and m is the median of the data subset S.

[0067] The Hampel filter identifies outliers by comparing the deviation of each data point from the median. If a data point deviates from the median m by more than k times the local standard deviation σ, that is, |x k -m|>k·σ, then the kth data point x k is considered an outlier.

[0068] Once an outlier is detected, the Hampel filter replaces that data point with the median m of the data subset. The Hampel filter iteratively applies this process in a sliding window that moves along the data set to mitigate the effects of fading noise on the overall data.

[0069] This method uses a Hampel filter to process phase data to suppress fading noise, effectively removing outliers without altering the overall data distribution. This approach significantly reduces the impact of outliers on statistical analysis, ensuring the accuracy and reliability of the results. Hampel utilizes the median and median absolute deviation to improve measurement accuracy while maximizing signal detail, enabling high spatial resolution and high-precision monitoring in distributed fiber optic sensing applications.

[0070] In order to have a further understanding of the present invention, XX of the present invention is described in detail below with reference to specific embodiments.

[0071] The beat frequency signals at temperatures of 20°C, 120°C, 130°C, 140°C, 150°C, and 220°C were collected. The signal at 20°C was used as the reference signal, and the signals at the other temperatures were used as the measurement signals.

[0072] The phase of the reference signal is subtracted from the phase of the measured signal at different temperatures, and then unwrapped and differentiated to obtain the differential relative phase at different temperatures, as shown in Figure 2. Figure 2 As shown in Figure 2, it can be clearly seen that there are outliers in the signal due to the influence of fading noise.

[0073] The denoising is further performed by Hampel filter, such as Figure 3 As shown, it can be seen that the outliers caused by fading noise are effectively suppressed.

[0074] In summary, the present invention provides a phase-based optical frequency domain reflectometry denoising method based on a Hampel filter. This method calculates the phase difference between the measurement signal and the reference signal, performs unwrapping and differential processing, obtains the differential relative phase, and further sets the parameters of the Hampel filter to perform effective denoising.

[0075] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A phase method optical frequency domain reflectometry denoising method based on HAMPEL filter, characterized in that: The denoising method comprises: Collect two sets of beat frequency signals of phase-based optical frequency domain reflectometry as reference signals and measurement signals; performing interpolation resampling on the reference signal and the measurement signal respectively to achieve nonlinear compensation of the signal; Perform FFT transformation on the two sets of beat frequency signals after nonlinear compensation to extract the phase information in each spectrum; Obtaining the relative phase of the reference signal and the measurement signal according to the phase information in each spectrum, and performing unwrapping and differential processing on the relative phase to obtain a differential relative phase; A Hampel filter is used to perform denoising on the differential relative phase to complete the denoising of the phase method optical frequency domain reflection data.

2. The HAMPEL filter-based phase-based optical frequency domain reflectometry denoising method according to claim 1, characterized in that: The interpolation and resampling of the reference signal and the measurement signal respectively specifically include: capturing the phase information of the beat signal through an auxiliary interferometer, extracting the instantaneous phase by using Hilbert transform, performing coordinate transformation and interpolation processing on the collected beat signal, and compensating for the nonlinear effect of the tunable laser.

3. The HAMPEL filter-based phase-based optical frequency domain reflectometry denoising method according to claim 1, characterized in that: After FFT transformation, the initial beat frequency phase of the reference signal is for Where v0 is the initial optical frequency of the tunable laser, τ r is the delay of the reference signal, and γ is the frequency sweep rate.

4. The method for denoising optical frequency domain reflectometry using a phase method based on a HAMPEL filter according to claim 1, wherein: After FFT transformation, the initial beat frequency phase of the measurement signal is for Among them, τ m To measure the delay of the signal.

5. The method for denoising optical frequency domain reflectometry using a phase method based on a HAMPEL filter according to claim 1, wherein: The denoising process of the differential relative phase using the Hampel filter specifically includes: (1) Input the Hampel filter to accept the window size N and the standard deviation threshold k; (2) Calculate the median m and local standard deviation σ of the data subset S in the current window; (3) Compare the deviation of each data point in the data subset S from the median m. If the deviation of a data point from the median m is greater than k times the local standard deviation σ, the data point is considered an outlier and is replaced by the median m of the data subset S. (4) Slide to the next window with the data set and repeat (2)-(3) until the denoising process of all data is completed.

6. The HAMPEL filter-based phase-based optical frequency domain reflectometry denoising method according to claim 5, characterized in that: The local standard deviation σ is calculated according to Get, where x i is the i-th data point in subset S, i = 1, 2, ... N.

Images