Method for measuring inter-channel delay differences of optical fibers based on non-linear fitting

Abstract

The application relates to a method for measuring delay difference between optical fiber channels based on nonlinear fitting, belonging to the technical field of optical fiber communication and measurement. The method comprises the following steps: dividing a modulated optical signal into a reference path signal and at least one measuring path signal, inputting the signals into different cores of a multi-core optical fiber or different optical fibers for transmission, respectively, carrying out photoelectric conversion and resampling to obtain the sampled reference path signal and the at least one measuring path signal; then carrying out Fourier transformation and squaring on the sampled reference path signal and the at least one measuring path signal to obtain power spectrum of the corresponding signals; then carrying out waveform nonlinear fitting on the power spectrum to obtain a frequency domain correlation peak expression and carrying out inverse Fourier transformation to obtain a time domain correlation peak expression; and carrying out correlation and fitting on the sampled reference path signal and the at least one measuring path signal by using the time domain correlation peak expression to obtain the coordinates of the fitted correlation peak, and calculating the delay difference between the optical fiber channels. The application has high measurement precision, low bandwidth requirement and low calculation complexity.

Description

Technical Field

[0001] This invention belongs to the fields of optical fiber communication, optical fiber measurement, and optical fiber communication measurement technology, and specifically relates to a method and apparatus for measuring the time delay difference between optical fiber channels based on nonlinear fitting. Background Technology

[0002] With the rapid development of emerging technologies such as cloud computing, the Internet of Things, connected vehicles, and virtual reality, global communication data network traffic is growing exponentially. Statistics show that 99% of global communication data network traffic is related to data centers, presenting both significant challenges and opportunities for optical interconnects within these centers. To meet this demand, multi-core fiber integrates multiple cores within a single cladding, enabling parallel transmission and significantly improving fiber optic transmission capacity. Multi-core fiber (MCF), as a key technology for increasing fiber optic transmission capacity, has attracted widespread attention due to its ability to support spatial division multiplexing (SDM). However, signal transmission between different cores in a MCF system exhibits a skew delay, which affects system performance, particularly in MIMO and multidimensional modulation systems. Therefore, accurately measuring the inter-core skew delay of multi-core fiber is crucial for optimizing SDM system performance.

[0003] Currently, researchers have proposed various methods for measuring time delay differences, including cross-correlation synchronization, Time of Flight (TOF) based methods, microwave interferometry, and frequency-modulated continuous wave methods. While these methods improve measurement accuracy to some extent, they still suffer from high bandwidth requirements, complex equipment, and high computational complexity. For example, traditional cross-correlation synchronization techniques rely on measuring the cross-correlation peak values ​​of signals from different channels to determine the time delay difference, while the TOF method requires high-bandwidth photodetectors and high-speed oscilloscopes, and microwave interferometry requires complex frequency modulation and demodulation equipment. Furthermore, with the increase in signal transmission rates, the measurement accuracy of traditional methods is limited by the sampling rate and bandwidth, posing significant challenges in practical applications, especially in long-distance fiber optic transmission systems. Additionally, for traditional cross-correlation synchronization, some researchers have proposed upsampling methods to improve measurement accuracy, but these methods have high computational complexity, particularly in long-distance fiber optic transmission systems, resulting in significant consumption of measurement time and computational resources.

[0004] This invention addresses the shortcomings of existing long-distance fiber optic transmission systems, which still face challenges in achieving high-precision delay measurement due to high bandwidth requirements, complex equipment, and high computational complexity. It explores high-precision delay measurement solutions for these systems. Furthermore, this invention aims to research high-precision delay measurement schemes suitable not only for short-distance fiber optic transmission systems but also for medium- and long-distance systems. It seeks to overcome the shortcomings of existing technologies, such as low measurement accuracy and high complexity, which make them unsuitable for engineering measurements. Summary of the Invention

[0005] The purpose of this invention is to address the problems of high bandwidth requirements, complex equipment, and high computational complexity in existing methods for measuring the delay of long-distance optical fiber transmission systems, and to provide a method for measuring the delay difference between optical fiber channels based on nonlinear fitting.

[0006] To achieve the above objectives, the present invention adopts the following technical solution.

[0007] This invention provides a method for measuring the time delay difference between optical fiber channels based on nonlinear fitting, comprising the following steps:

[0008] The optical signal modulated by the modulator is divided into a reference path signal and at least one measurement path signal, which are then transmitted into different cores of a multi-core optical fiber or into different optical fibers and then subjected to photoelectric conversion and resampling.

[0009] The sampling is implemented using an oscilloscope.

[0010] Preferably, the sampled reference path signal and at least one measurement path signal obtained by photoelectric conversion and resampling are subjected to noise reduction, normalization and zero-padding preprocessing.

[0011] The power spectrum of the corresponding signal is obtained by performing Fourier transform on the sampled reference signal and at least one measurement signal and then squaring them.

[0012] Then, perform waveform nonlinear fitting on the power spectrum to obtain the frequency domain correlation peak expression, and then perform inverse Fourier transform to obtain the time domain correlation peak expression;

[0013] The method described above performs waveform nonlinear fitting through mathematical modeling, thereby achieving fitting of the measurement results. This makes the delay measurement accuracy no longer limited by the oscilloscope's sampling rate and signal rate, enabling high-precision delay measurement under low sampling rate conditions. This greatly reduces the demand for equipment bandwidth and significantly reduces experimental and engineering costs.

[0014] The time-domain correlation peak expression is used to perform correlation refitting on the sampled reference path signal and at least one measurement path signal to obtain the coordinates of the fitted correlation peak, and then the inter-optical channel delay difference is calculated.

[0015] The method achieves sub-picosecond measurement accuracy by performing cross-correlation calculations on the received signal and the reference signal and fitting the cross-correlation peak using nonlinear fitting. Specifically, it uses the time-domain correlation peak expression to perform correlation refitting on the sampled reference path signal and at least one measurement path signal, thus breaking through the sampling rate limitation of traditional cross-correlation methods.

[0016] The mathematical model expression for the frequency domain correlation peak is:

[0017]

[0018] The Indicates the corresponding frequency The power value, It is a coefficient. It is the slope of a linear function. It is the intercept of a linear function. It's frequency. It is the roll-off factor. It is the symbol rate.

[0019] The mathematical model expression for the time-domain correlation peak is:

[0020]

[0021] The express The corresponding amplitude, Indicates a delay. It is a coefficient. It is the slope of a linear function. It is the intercept of a linear function. It's frequency. It is the roll-off factor. It is the symbol rate.

[0022] The modulator is controlled by a control device to operate at the optimal bias point.

[0023] In specific implementation, it also includes: introducing the measurement path signal into an adjustable delay difference as a reference through an optical fiber signal delay device, and then repeatedly executing the method of delay difference between optical fiber channels to obtain the measurement result.

[0024] The introduction of an adjustable delay difference is specifically implemented by stretching the delay by k picoseconds using an optical fiber delay line, for a total of 20 to 100 times, where k is greater than or equal to 1 and less than or equal to 10.

[0025] The waveform nonlinear fitting is achieved using Gaussian, Lorentzian, raised cosine, linear, quadratic functions, or combinations thereof.

[0026] The process of using the time-domain correlation peak expression to perform correlation refitting on the sampled reference path signal and at least one measurement path signal involves selecting 5 to 10 points on both sides of the correlation peak for fitting.

[0027] The symbol rate of the optical signal ranges from 5 Gb / s to 100 Gb / s.

[0028] In summary, the proposed method achieves multi-channel cross-correlation measurement based on nonlinear fitting. By performing cross-correlation calculations on the received signal and the reference signal and using nonlinear fitting to fit a few points near the cross-correlation peak, the accuracy of delay measurement can be significantly improved while reducing computational complexity.

[0029] Compared with upsampling methods, it significantly reduces computational complexity, and improves the application limitations in high-precision, low-complexity scenarios, especially under the condition of poor adaptability to long-distance transmission.

[0030] The method is not only applicable to short-distance fiber optic transmission systems, but can also be extended to long-distance fiber optic transmission systems, and has broad application prospects.

[0031] Beneficial effects

[0032] Based on the above scheme, it can be seen that the method for measuring the inter-optical channel delay difference based on nonlinear fitting of the present invention has at least one of the following advantages over the prior art:

[0033] (1) High measurement accuracy: Traditional cross-correlation delay measurement methods are limited by signal transmission rate and oscilloscope sampling rate; This invention performs cross-correlation operation between the received signal and the reference signal, and uses nonlinear fitting to fit the cross-correlation peak, which can break through the sampling rate limitation of traditional cross-correlation methods and achieve sub-picosecond measurement accuracy.

[0034] (2) Low bandwidth requirement: Compared with traditional measurement methods, this method uses a mathematical model to fit the measurement results, and is no longer limited by the sampling rate and signal rate of the original data. It can achieve high-precision measurement under low sampling rate conditions, reduce the requirement for equipment bandwidth, and greatly reduce experimental and engineering costs.

[0035] (3) Low computational complexity: Compared with other high sampling rate upsampling methods, the resolution of delay measurement is increased by increasing the sampling rate of the signal. However, although the upsampling method improves the measurement accuracy to a certain extent, it also has some significant drawbacks, especially in multi-core fiber optic systems, which may limit its application effect. The upsampling method leads to a significant increase in computational complexity and poor adaptability to long-distance transmission, which limits its application in high-precision, low-complexity scenarios. This multi-channel cross-correlation measurement method based on nonlinear fitting technology can significantly improve the accuracy of delay measurement while reducing computational complexity by performing cross-correlation operation between the received signal and the reference signal and using nonlinear fitting technology to fit a few points near the cross-correlation peak. This method is not only applicable to short-distance fiber optic transmission systems, but can also be extended to long-distance fiber optic transmission systems, and has broad application prospects. Attached Figure Description

[0036] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the embodiments will be briefly described below.

[0037] Figure 1 This is a system diagram of the cross-correlation delay measurement scheme provided by the present invention;

[0038] Figure 2a The signal is generated by the signal generator (PPG);

[0039] Figure 2b The power spectrum of the signal emitted by the signal generator (PPG);

[0040] Figure 3a The signal received by the signal receiver (DSO)

[0041] Figure 3b The power spectrum of the received signal, i.e., the spectrum related to the signal and the fitting result;

[0042] Figure 4 This is a flowchart of the signal processing procedure.

[0043] Figure 5 The nonlinear fitting result, i.e. the time-domain fitting result of the cross-correlation of the two signals;

[0044] Figure 6 The nonlinear fitting measurement results of the motor delay line with a delay of 2 ps each time;

[0045] Figure 7 Comparison of computation time for nonlinear fitting and upsampling 100 times. Detailed Implementation

[0046] To better understand the technical solutions of the embodiments of the present invention, the embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0047] It should be understood that the described embodiments are merely some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0048] Example 1

[0049] This invention proposes a method for measuring the inter-channel delay difference in optical fibers based on nonlinear fitting, belonging to the field of optical fiber communication and measurement technology. Specifically, the method includes: splitting a modulated optical signal into a reference signal and at least one measurement signal, respectively inputting them into different cores of a multi-core optical fiber or into different optical fibers for transmission, and then performing photoelectric conversion and resampling to obtain the sampled reference signal and at least one measurement signal; next, performing Fourier transform on the sampled reference signal and at least one measurement signal, and then squaring them to obtain the power spectrum of the corresponding signal; then performing waveform nonlinear fitting on the power spectrum to obtain the frequency domain correlation peak expression, and then performing an inverse Fourier transform to obtain the time domain correlation peak expression; using the time domain correlation peak expression to perform correlation refitting on the sampled reference signal and at least one measurement signal to obtain the coordinates of the fitted correlation peak, and calculating the inter-channel delay difference in optical fibers. This invention offers high measurement accuracy, low bandwidth requirements, and low computational complexity.

[0050] like Figures 1 to 7 As shown, this invention discloses a method for measuring the inter-optical fiber channel delay difference based on nonlinear fitting, which includes the following steps:

[0051] The optical signal modulated by the modulator is divided into a reference path signal and at least one measurement path signal, which are then transmitted into different cores of a multi-core optical fiber or into different optical fibers and then subjected to photoelectric conversion and resampling.

[0052] The power spectrum of the corresponding signal is obtained by performing Fourier transform on the sampled reference signal and at least one measurement signal and then squaring them.

[0053] Then, perform waveform nonlinear fitting on the power spectrum to obtain the frequency domain correlation peak expression, and then perform inverse Fourier transform to obtain the time domain correlation peak expression;

[0054] The time-domain correlation peak expression is used to perform correlation refitting on the sampled reference path signal and at least one measurement path signal to obtain the coordinates of the fitted correlation peak, and then the inter-optical channel delay difference is calculated.

[0055] The device corresponding to this method includes: a signal generator, a modulation module; a modulator control device, an optical fiber coupler, a photodetector, an optical fiber signal delay device, a signal receiver, and a digital signal processing module.

[0056] The functions of each part are described below.

[0057] Signal generators are used to generate low-speed / high-speed digital signals.

[0058] The modulation module modulates the digital signal onto the laser using an intensity modulator.

[0059] The modulator control device provides feedback control signals, ensuring the modulator always operates at the optimal modulation point.

[0060] Fiber optic couplers are used to split modulated optical signals into multiple paths, which are then input into the fiber cores to be measured in a multi-core optical fiber.

[0061] The photodetector converts the received optical signal into an electrical signal and inputs it into an oscilloscope for sampling.

[0062] Optical fiber signal delay devices, such as adjustable optical fiber delay lines, piezoelectric ceramic optical fiber modulators, and other adjustable optical fiber delay devices, are used to introduce an adjustable delay difference as a measurement reference.

[0063] The signal receiver includes a data acquisition card, an integrated optical module, a sampling oscilloscope, and other signal receiving devices, used to receive and measure the signals of the channel to be measured, and perform subsequent digital signal processing.

[0064] The functions of the digital signal processing module include:

[0065] Signal preprocessing: Denoising, normalizing, and zero-padding the signal (to improve the resolution of the Fast Fourier Transform (FFT)).

[0066] Frequency domain analysis: The FFT of a signal can be calculated using MATLAB, Python, or other data processing tools, and the outputs are: spectrum (amplitude-frequency) and power spectrum.

[0067] Frequency domain modeling: Select Gaussian, Lorentzian, raised cosine, and other mixed mathematical models and use data processing tools to model the power spectrum.

[0068] Inverse Fourier Transform: The power spectrum, which is the spectrum of the cross-correlated signal, is fitted to obtain the frequency domain analytical expression, and the time domain correlation function is output through the inverse Fourier transform.

[0069] Time-domain model fitting: The obtained time-domain correlation function is used to fit the discrete points of the signal correlation peak to optimize the correlation peak.

[0070] Delay extraction: Extract the x-coordinate of the optimized relevant peak and calculate the delay.

[0071] A signal generator (PPG) is used to generate low-speed / high-speed digital signals. A signal transmitter generates digital signals. The signal can be a Prabs sequence, a Gold sequence, a Kasami sequence, or other signal sequences.

[0072] The modulation module modulates the digital signal onto the laser through intensity modulation. The modulation module can be a Mach-Zehnder modulator (MZM) or an intensity modulator. To ensure the accuracy of signal delay measurement, this invention innovatively incorporates a modulator control device in the experimental section, such as equipping the MZM with an automatic bias feedback controller. This automatic bias controller continuously adjusts and maintains the MZM at the optimal bias voltage by monitoring the power or average power of the output signal and using a method to find the power extreme value.

[0073] In this embodiment, the signal generator generates a PRBS11 digital signal with a transmission rate of 10Gb / s, such as... Figure 2a As shown, the signal is modulated onto a C-band laser using a Mach-Zehnder modulator (MZM). Figure 1 The transmission system shown consists of a 1551.72nm laser, an MZM, and a PPG. The modulator is controlled by a control unit to operate at the optimal bias point.

[0074] The modulated optical signal is split into a reference path signal and at least one measurement path signal by an optical fiber coupler. These signals are then transmitted into different cores of a multi-core optical fiber or into different optical fibers, and subsequently received and sampled separately. In this embodiment, a 7-core optical fiber with a length of 13km is used. Taking two cores as an example, core 2 is selected as the reference path and core 1 as the measurement path.

[0075] After fiber delay Transmission system After transmission, at the receiving end, the reference path signal directly converts the optical signal into an electrical signal via a photodetector (PD). The signal is sampled using an oscilloscope or directly via an integrated optical module. The measured signal is then delayed by an optical fiber signal delay device before being converted into an electrical signal by a photodiode (PD). It can be sampled using an oscilloscope or directly via an integrated optical module.

[0076] The reference signal (core 2) is directly input to an oscilloscope (DSO) with a sampling rate of 20 Gsa / s via a 40 GHz photodetector (PD). Figure 3aAs shown, the measurement signal (core 1) passes through an adjustable fiber delay line (ODL) and then is input to the other receiver of the oscilloscope via a PD input. The oscilloscope is controlled to sample both signals simultaneously. Because the oscilloscope's sampling rate is 20 GSa / s, the signal resolution is 50 ps, ​​which is far from meeting the synchronization requirements of high-speed signals.

[0077] The sampled reference path signal and at least one measurement path signal obtained by photoelectric conversion and resampling are subjected to denoising, normalization and zero-padding preprocessing (to improve FFT resolution) to prepare for subsequent mathematical modeling.

[0078] The transmission system consists of a 3dB fiber coupler, a multi-core fiber, and a photodetector (PD). Afterwards, the signal received by the oscilloscope was ,like Figure 3a As shown, its signal power spectrum is as follows Figure 3b As shown.

[0079] After processing, the positions of the relevant peaks can be obtained from the signal received by the oscilloscope.

[0080] .

[0081] The spectrum of the relevant peak can be expressed as:

[0082]

[0083] Right now: ,like Figure 3b As shown.

[0084] This invention performs cross-correlation calculations on the received reference signal and measurement signal to extract the correlation peak. In traditional cross-correlation delay measurement methods, the position (horizontal axis) of the correlation peak represents the delay measurement value of the two channels. However, traditional cross-correlation synchronization techniques are limited by sampling rate and bandwidth, resulting in insufficient measurement accuracy. The core of this invention lies in obtaining the optimal delay correlation peak through subsequent nonlinear fitting techniques. This improves the accuracy of delay difference measurement, thereby obtaining a higher resolution transmission delay time.

[0085] In both laboratory settings and practical engineering applications, it is difficult to directly obtain the time-domain waveform of the received signal. Obtain relevant peaks Specific waveform formulas, however, time-domain signals The frequency domain expression of the transmitted signal is obtained by performing a Fourier transform. This frequency domain expression is the signal spectrum. Squaring the spectra of the two signals yields the signal power spectrum. In other words, we can first calculate the signal's correlation frequency domain waveform, which is the signal power spectral density waveform. The received signal is processed by FFT to obtain the power spectrum waveform. Since the two signals are identical and the transmission systems they pass through are the same, their frequency domain waveforms are identical.

[0086] Mathematical modeling of the power spectrum waveform is performed. Traditional system transfer functions such as Gaussian, Lorentzian, raised cosine, linear, quadratic functions, and other piecewise hybrid mathematical models can be selected, and data processing tools can be used to model the power spectrum. The mathematical model is fitted to the power spectrum waveform (frequency domain waveform of the correlation peak), and the goodness of fit is used to verify the fitting effect and the correctness of the mathematical model selection.

[0087] right Figure 3b Waveform fitting of the power spectrum was performed. Observation of the graph showed that the waveform conformed to a piecewise function, and was closest to the piecewise forms of the traditional raised cosine transfer function and linear function. A mathematical model expression for the relevant peak spectrum was created:

[0088]

[0089] in: Indicates the corresponding frequency power value It is a coefficient. Slope of a linear function The intercept (constant) of a linear function. It's frequency. Roll-off factor, Symbol rate.

[0090] From the above formula and Figure 3b As shown, the power spectrum waveform is fitted by two parts, namely a linear function and a raised cosine function. The goodness of fit of the power spectrum waveform fitted by this formula is 0.99.

[0091] The time-domain representation after the inverse Fourier transform is the sum of two parts:

[0092]

[0093] A linear function contributes to the time domain as a squared sinc function, while a raised cosine function contributes as a modulated sinc function. Since the signal is a piecewise continuous and symmetrical signal, its square can be written as:

[0094]

[0095] in: express The corresponding amplitude, Indicates a delay. It is a coefficient. It is the slope of a linear function. It is the intercept (constant) of a linear function. It's frequency. It is the roll-off factor. It is the symbol rate.

[0096] The above formula is the waveform expression of the cross-correlation function of the received signal. Using this formula to fit the cross-correlation experimental data, a higher resolution delay time can be obtained, such as... Figure 4 The goodness of fit shown is 0.99.

[0097] The present invention also includes the following steps: changing the delay time variation to a relative delay variation benchmark, and performing multiple measurements to verify the measurement results, thereby better measuring the accuracy of the delay measurement of the present invention. Specifically, the delay is stretched by k picoseconds using an optical fiber delay line, for a total of 20 to 100 stretches, where k is greater than or equal to 1 and less than or equal to 10. Figure 1 In this study, the delay time change caused by stretching an adjustable fiber delay line (ODL) was used as the relative delay change (skew) benchmark. The channel transmission delay change was measured once for every 2ps stretch of the ODL, and a total of 50 stretches were performed to reach a delay change of 100ps to verify the measurement results.

[0098] It is known that the upsampling cross-correlation method improves the accuracy of delay measurements. However, when the sampling rate increases to a certain level, the measurement accuracy will no longer improve. A comparison was made between the measurement accuracy of the upsampling cross-correlation scheme and the fitting scheme proposed in this paper. It was found that the measurement accuracy of the fitting scheme is basically consistent with that of a 100x upsampling rate of 2000Gb / s. This demonstrates that the fitting scheme proposed in this invention achieves high measurement accuracy. Therefore, subsequent data processing and measurement accuracy verification were performed by comparing 50 sets of experimental data using both the 100x upsampling scheme (2000Gb / s) and the fitting scheme. Figure 5 As shown, this invention has improved the accuracy of cross-correlation delay measurement by 100 times, reaching 50 fs, thus achieving high-precision delay measurement.

[0099] For cross-correlation synchronization methods, the length of the correlated data directly affects the computational complexity of the correlation synchronization algorithm. With the same symbol rate and number of symbols, doubling the symbol sampling rate significantly increases the complexity of the correlation synchronization algorithm. In this experiment, the symbol rate was 10 Gb / s and the sampling rate was 20 Gb / s. Achieving a sampling rate of 2000 Gb / s increases the computational complexity by a factor of 10,000. In contrast, the fitting experiment, based on the cross-correlation of the original data, only uses 5-10 points on either side of the correlation peak for correlation peak fitting calculation, resulting in a significantly lower increased complexity than the upsampling method.

[0100] Under the same computational environment, the correlation peak fitting scheme and the 100x upsampling cross-correlation scheme were used to optimize 50 sets of experimental data. The time required was as follows: Figure 6 As shown, the average time required for the correlation peak fitting scheme is 0.1s, and the average time required for cross-correlation calculation at 2000Gb / s with a 100-fold upsampling is 1s, representing a 10-fold increase in calculation speed and significantly saving measurement time. For long-distance fiber optic measurement experiments, where longer prbs codes are needed to avoid over-period phenomena, the advantages of the fitting experiment scheme—its high efficiency, simplicity, and short calculation time—become even more pronounced. The cross-correlation synchronous fitting experiment scheme has higher research value for long-distance, real-time, and high-precision measurement systems. This experiment used the nonlinear least squares fitting function built into MATLAB. Further research can optimize the fitting algorithm to effectively improve experimental accuracy and reduce complexity.

[0101] The beneficial effects of this invention are as follows.

[0102] (1) High measurement accuracy: Traditional cross-correlation delay measurement methods are limited by signal transmission rate and oscilloscope sampling rate. This invention achieves sub-picosecond measurement accuracy by performing cross-correlation calculation on the received transmitted signal and reference signal, and using nonlinear fitting technology to fit the cross-correlation peak.

[0103] (2) Low bandwidth requirement: Compared with traditional measurement methods, this method uses a mathematical model to fit the measurement results, and is no longer limited by the sampling rate and signal rate of the original data. It can achieve high-precision measurement under low sampling rate conditions, reduce the requirement for equipment bandwidth, and greatly reduce experimental and engineering costs.

[0104] (3) Low computational complexity: Compared with other high sampling rate upsampling methods, the resolution of delay measurement is increased by increasing the sampling rate of the signal. However, although the upsampling method improves the measurement accuracy to a certain extent, it also has some significant drawbacks, especially in multi-core fiber optic systems, which may limit its application effect. The upsampling method leads to a significant increase in computational complexity and poor adaptability to long-distance transmission, which limits its application in high-precision, low-complexity scenarios. This multi-channel cross-correlation measurement method based on nonlinear fitting technology can significantly improve the accuracy of delay measurement while reducing computational complexity by performing cross-correlation operation between the received signal and the reference signal and using nonlinear fitting technology to fit a few points near the cross-correlation peak. This method is not only applicable to short-distance fiber optic transmission systems, but can also be extended to long-distance fiber optic transmission systems, and has broad application prospects.

[0105] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for measuring the time delay difference between optical fiber channels based on nonlinear fitting, characterized in that, Includes the following steps: The optical signal modulated by the modulator is divided into a reference path signal and at least one measurement path signal, which are then transmitted into different cores of a multi-core optical fiber or into different optical fibers and then subjected to photoelectric conversion and resampling. The power spectrum of the corresponding signal is obtained by performing Fourier transform on the sampled reference signal and at least one measurement signal and then squaring them. Then, perform waveform nonlinear fitting on the power spectrum to obtain the frequency domain correlation peak expression, and then perform inverse Fourier transform to obtain the time domain correlation peak expression; The time-domain correlation peak expression is used to perform correlation refitting on the sampled reference path signal and at least one measurement path signal to obtain the coordinates of the fitted correlation peak, and then the inter-optical channel delay difference is calculated.

2. The method for measuring inter-optical fiber channel delay difference based on nonlinear fitting according to claim 1, characterized in that, Also includes: The sampled reference path signal and at least one measurement path signal obtained by photoelectric conversion and resampling are subjected to noise reduction, normalization and zero-padding preprocessing.

3. The method for measuring inter-optical fiber channel delay difference based on nonlinear fitting according to claim 1, characterized in that, The modulator is controlled by a control device to operate at the optimal bias point.

4. The method for measuring inter-optical fiber channel delay difference based on nonlinear fitting according to claim 1, characterized in that, The mathematical model expression for the frequency domain correlation peak is: The Indicates the corresponding frequency The power value, It is a coefficient. It is the slope of a linear function. It is the intercept of a linear function. It's frequency. It is the roll-off factor. It is the symbol rate.

5. The method for measuring inter-optical fiber channel delay difference based on nonlinear fitting according to claim 4, characterized in that, The mathematical model expression for the time-domain correlation peak is: The express The corresponding amplitude, Indicates a delay. It is a coefficient. It is the slope of a linear function. It is the intercept of a linear function. It's frequency. It is the roll-off factor. It is the symbol rate.

6. The method for measuring inter-optical fiber channel delay difference based on nonlinear fitting according to claim 1, characterized in that, In specific implementation, it also includes: introducing the measurement path signal into an adjustable delay difference as a reference through an optical fiber signal delay device, and then repeatedly executing the method of delay difference between optical fiber channels to obtain the measurement result.

7. The method for measuring inter-optical fiber channel delay difference based on nonlinear fitting according to claim 6, characterized in that, The introduction of an adjustable delay difference is specifically implemented by stretching the delay by k picoseconds using an optical fiber delay line, for a total of 20 to 100 times, where k is greater than or equal to 1 and less than or equal to 10.

8. The method for measuring inter-optical fiber channel delay difference based on nonlinear fitting according to claim 1, characterized in that, The waveform nonlinear fitting is achieved using Gaussian, Lorentzian, raised cosine, linear, quadratic functions, or combinations thereof.

9. The method for measuring inter-optical fiber channel delay difference based on nonlinear fitting according to claim 1, characterized in that, The process of using the time-domain correlation peak expression to perform correlation refitting on the sampled reference path signal and at least one measurement path signal involves selecting 5 to 10 points on both sides of the correlation peak for fitting.

10. The method for measuring inter-optical fiber channel delay difference based on nonlinear fitting according to claim 1, characterized in that, The symbol rate of the optical signal ranges from 5 Gb / s to 100 Gb / s.

Images