Multi-channel optical delay measurement method and device based on compressed sensing

Abstract

The present invention discloses a multi-channel optical time delay measurement method based on compressive sensing, which is used to simultaneously measure the optical time delays of S reflection points at different positions or S parallel channels in the optical link to be measured, where S is an integer greater than or equal to 3. The method includes the following steps: S1. Intensity modulate an incoherent optical carrier with a microwave swept-frequency signal composed of M frequency points randomly selected from a frequency range with N frequency points to obtain a detection optical signal that is sparse in frequency, where M << N; S2. Let the detection optical signal pass through the optical link to be measured and then perform photoelectric conversion to obtain a microwave signal with sparse frequency carrying time delay information; S3. Use the compressive sensing method to calculate the optical time delay. The present invention also discloses a multi-channel optical time delay measurement device based on compressive sensing. Compared with the prior art, the present invention has the advantages of fast measurement speed, high measurement accuracy, and simple structure.

Description

Technical Field

[0001] This invention relates to a method for measuring optical delay, and more particularly to a multi-channel optical delay measurement method and apparatus. Background Technology

[0002] Optical delay measurement is a fundamental requirement in fields such as distributed fiber optic sensing, distributed radar networks, optically controlled phased arrays, and wireless communication. In these fields, optical delay measurement often involves multi-channel scenarios. For example, in 5G / 6G wireless communication and array radar, it is typically necessary to measure the delay consistency of multiple channels (such as a subarray), requiring simultaneous measurement of multiple channels. Similarly, in distributed fiber optic sensing, there are often multiple reflection points within the fiber; delay measurement must be able to distinguish these reflection points and accurately obtain the delay (length, position) information of each reflection point.

[0003] Currently, commonly used multi-channel optical time delay measurement techniques mainly include time-domain measurement methods and frequency-domain measurement methods. Time-domain measurement methods, primarily using optical time-domain reflectometry (OTDR) as an example, are suitable for large-scale time delay measurements and are simple in structure and low in cost. Although OTDRs can measure distances of hundreds of kilometers, their measurement distance and resolution are mutually constrained by pulse width, generally limiting their resolution to the meter level, and they also have measurement blind zones.

[0004] Frequency domain measurement methods, such as coherent-optical frequency-domain reflectionometry (C-OFDR), offer high resolution over short distances (up to sub-picosecond levels). However, the measurement distance is limited by the coherence and frequency scanning nonlinearity of the swept laser. To overcome this problem, Hu et al. proposed an optical time delay measurement scheme based on an optical resonator, achieving a spatial resolution of 5 cm over 100 km of fiber. While such methods can achieve high accuracy over long distances, the measurement time is long (usually several minutes) and they are sensitive to environmental conditions, making them unsuitable for high-speed time delay measurements.

[0005] Incoherent-optical frequency-domain reflectometry (I-OFDR) typically involves scanning the frequency of a microwave signal modulated on the light using a vector network analyzer. The frequency response of the optical link is obtained in the electrical domain, and the time-domain response is derived through inverse Fourier transform, allowing for the calculation of the time delay at each reflection point. Thanks to the precise scanning of the microwave source and the more stable amplitude and phase detection and higher resolution spectral analysis capabilities of the vector network analyzer, the accuracy of the time delay can reach the sub-picosecond level. However, to improve resolution, scanning over a wide frequency range is required, which limits the measurement speed due to the limited frequency sweep rate.

[0006] In summary, to improve latency resolution and achieve multi-channel measurement capabilities, the methods described above often require increasing the detection bandwidth. This leads to a large sampling rate and data volume, increasing the complexity, cost, and data processing time of the measurement system, making it difficult to overcome the constraint between measurement accuracy and speed. In multi-channel scenarios, such as wireless communication and array radar, it is necessary to process data from multiple signal sources simultaneously. Traditional sampling methods impose enormous data processing pressure, making it difficult to meet real-time requirements. Summary of the Invention

[0007] The technical problem to be solved by the present invention is to overcome the shortcomings of the existing technology and provide a multi-channel optical time delay measurement method based on compressed sensing, which has the advantages of fast measurement speed, high measurement accuracy and simple structure.

[0008] The present invention specifically adopts the following technical solutions to solve the above-mentioned technical problems:

[0009] A multi-channel optical delay measurement method based on compressed sensing is used to simultaneously measure the optical delay of S different reflection points or S parallel channels in the optical link under test, where S is an integer greater than or equal to 3; the method includes the following steps:

[0010] S1. Use a microwave sweep signal consisting of M frequency points randomly selected from a frequency range with N frequency points to perform intensity modulation on an incoherent optical carrier to obtain a frequency-sparse probe optical signal, M << N.

[0011] S2. The probe light signal is converted into a photoelectric signal after passing through the optical link under test to obtain a microwave signal with sparse frequency carrying time delay information.

[0012] S3. Using the frequency response obtained by sampling the microwave signal with sparse frequencies carrying time delay information in the digital domain as the observation value I, and the discrete Fourier transform matrix as the sparse matrix ψ, a measurement matrix Φ is constructed based on the position information of each sweep frequency point in the microwave sweep signal with sparse frequencies in the full frequency range. According to the formula I = Φψθ, the time-domain response information of the sparse signal θ is reconstructed using the compressed sensing reconstruction method, thereby obtaining the optical time delay τ1~τ2 of the S different reflection points or S parallel channels. s .

[0013] Furthermore, higher precision optical delay τ′1~τ′ S Specifically, the optical delay τ′1~τ′ is obtained by the following method: windowing is performed on the time-domain pulses corresponding to each channel of θ, and the windowed signal is converted to the frequency domain. Then, the converted frequency-domain signal is zero-padded, and the zero-padded frequency-domain signal is converted back to the time domain to obtain a higher precision optical delay τ′1~τ′. S .

[0014] Preferably, the incoherent optical carrier is a broadband optical signal generated by an amplified spontaneous emission source.

[0015] Preferably, the method for constructing the measurement matrix Φ is as follows: an N*N identity matrix is ​​generated, where the position of each row of the identity matrix corresponds to the position information of N frequency points in the frequency range; the M frequency points in the microwave sweep signal are extracted from the M rows corresponding to the identity matrix to construct an M×N order measurement matrix Φ.

[0016] Preferably, the compressed sensing reconstruction method is the Orthogonal Matching Pursuit (OMP) algorithm.

[0017] Based on the same inventive concept, the following technical solutions can also be obtained:

[0018] A multi-channel optical delay measurement device based on compressed sensing is used to simultaneously measure the optical delay of S different reflection points or S parallel channels in the optical link under test, where S is an integer greater than or equal to 3. The device includes: an intensity modulation module, used to intensity modulate an incoherent optical carrier with a microwave sweep signal composed of M frequency points randomly selected from a frequency range of N frequency points to obtain a frequency-sparse probe optical signal, where M << N; and a photoelectric detection module, used to perform photoelectric conversion on the probe optical signal after it passes through the optical link under test to obtain a microwave signal with sparse frequencies carrying delay information.

[0019] The signal acquisition and processing module is used to take the frequency response obtained by sampling the microwave signal with sparse frequencies carrying time delay information in the digital domain as the observation value I, and the discrete Fourier transform matrix as the sparse matrix ψ. Based on the position information of each sweep frequency point in the sparse frequency sweep signal within the full frequency range, a measurement matrix Φ is constructed. According to the formula I = Φψθ, the time-domain response information of the sparse signal θ is reconstructed using compressed sensing reconstruction method, thereby obtaining the optical time delays τ1 to τ2 of the S different reflection points or S parallel channels. s .

[0020] Furthermore, higher precision optical delay τ′1~τ′ s Specifically, the optical delay τ′1~τ′ is obtained by the following method: windowing is performed on the time-domain pulses corresponding to each channel of θ, and the windowed signal is converted to the frequency domain. Then, the converted frequency-domain signal is zero-padded, and the zero-padded frequency-domain signal is converted back to the time domain to obtain a higher precision optical delay τ′1~τ′. s .

[0021] Preferably, the incoherent optical carrier is a broadband optical signal generated by an amplified spontaneous emission source.

[0022] Preferably, the method for constructing the measurement matrix Φ is as follows: an N*N identity matrix is ​​generated, where the position of each row of the identity matrix corresponds to the position information of N frequency points in the frequency range; the M frequency points in the microwave sweep signal are extracted from the M rows corresponding to the identity matrix to construct an M×N order measurement matrix Φ.

[0023] Preferably, the compressed sensing reconstruction method is an orthogonal matching pursuit algorithm.

[0024] Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

[0025] This invention analyzes the sparsity of multi-channel delay in the time domain and proposes and constructs a frequency-sparse delay measurement method. This allows for the use of compressed sensing technology to reconstruct the original signal with less sampled data, achieving a fast, high-precision, and low-cost optical delay measurement system. Thanks to the sparse frequency demodulation method, this approach significantly reduces the sampling rate and data rate of the measurement system, lowering system complexity and cost. Furthermore, it greatly improves measurement speed while maintaining accuracy, providing a highly efficient testing method for delay calibration in array radar and 5G / 6G array communication systems. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of a specific embodiment of the optical delay measurement device of the present invention. Detailed Implementation

[0027] To address the shortcomings of existing technologies, this invention addresses the issue by employing a non-coherent light source as the optical carrier, which is intensity-modulated by a sparse frequency microwave sweep signal. This intensity-modulated signal is then used as the probe light signal. Since the optical link under test has multiple emission points or parallel channels, reflection points or channels at different locations will have different amplitudes and delays. A compressed sensing algorithm is then used to process the limited information in the frequency domain to calculate the delay, thus solving the problem of long measurement times required for large-scale frequency sweeps in traditional optical delay measurement systems. Furthermore, compressed sensing, through sparse signal reconstruction technology, can effectively suppress noise and interference, improving the robustness and accuracy of measurement results in optical delay measurement.

[0028] Compressed sensing, also known as compressed sampling or sparse sampling, is a novel sampling theory. By exploiting the sparsity of signals, it allows for the acquisition of discrete signals using random sampling at rates far lower than the Nyquist sampling rate. These discrete samples are then reconstructed using nonlinear reconstruction algorithms. Compressed sensing enables the acquisition of discrete signals through random sampling at rates significantly lower than traditional sampling methods, thus drastically reducing the overall sampling rate. Furthermore, by acquiring key signal information only through random sampling, it avoids the collection of a large amount of redundant information found in traditional sampling methods, significantly reducing sampling time and ensuring rapid measurement. In multi-channel measurements, this translates to savings in storage space, reduced bandwidth requirements for data transmission, and reduced complexity in subsequent data processing. Moreover, by reducing the amount and complexity of sampled data, compressed sensing technology lowers the demands on hardware performance and resources, thereby reducing hardware costs and power consumption. This is of great significance for the large-scale deployment and application of multi-channel measurement systems.

[0029] The specific technical solution proposed in this invention is as follows:

[0030] A multi-channel optical delay measurement method based on compressed sensing is used to simultaneously measure the optical delay of S different reflection points or S parallel channels in the optical link under test, where S is an integer greater than or equal to 3; the method includes the following steps:

[0031] S1. Use a microwave sweep signal consisting of M frequency points randomly selected from a frequency range with N frequency points to perform intensity modulation on an incoherent optical carrier to obtain a frequency-sparse probe optical signal, M << N.

[0032] S2. The probe light signal is converted into a photoelectric signal after passing through the optical link under test to obtain a microwave signal with sparse frequency carrying time delay information.

[0033] S3. Using the frequency response obtained by sampling the microwave signal with sparse frequencies carrying time delay information in the digital domain as the observation value I, and the discrete Fourier transform matrix as the sparse matrix ψ, a measurement matrix Φ is constructed based on the position information of each sweep frequency point in the microwave sweep signal with sparse frequencies in the full frequency range. According to the formula I = Φψθ, the time-domain response information of the sparse signal θ is reconstructed using the compressed sensing reconstruction method, thereby obtaining the optical time delay τ1~τ2 of the S different reflection points or S parallel channels. S .

[0034] A multi-channel optical delay measurement device based on compressed sensing is used to simultaneously measure the optical delay of S different reflection points or S parallel channels in the optical link under test, where S is an integer greater than or equal to 3. The device includes: an intensity modulation module, used to intensity modulate an incoherent optical carrier with a microwave sweep signal composed of M frequency points randomly selected from a frequency range of N frequency points to obtain a frequency-sparse probe optical signal, where M << N; and a photoelectric detection module, used to perform photoelectric conversion on the probe optical signal after it passes through the optical link under test to obtain a microwave signal with sparse frequencies carrying delay information.

[0035] The signal acquisition and processing module is used to take the frequency response obtained by sampling the microwave signal with sparse frequencies carrying time delay information in the digital domain as the observation value I, and the discrete Fourier transform matrix as the sparse matrix ψ. Based on the position information of each sweep frequency point in the sparse frequency sweep signal within the full frequency range, a measurement matrix Φ is constructed. According to the formula I = Φψθ, the time-domain response information of the sparse signal θ is reconstructed using compressed sensing reconstruction method, thereby obtaining the optical time delays τ1 to τ2 of the S different reflection points or S parallel channels. S .

[0036] To facilitate public understanding, the technical solution of the present invention will be described in detail below using the multi-channel optical delay measurement of an optical link under test with multiple reflection points at different locations as an example, in conjunction with the accompanying drawings:

[0037] The optical delay measurement device in this implementation has the following structure: Figure 1As shown, the incoherent optical carrier generated by the amplified spontaneous emission broadband light source is filtered and amplified before being input to the optical carrier input port of the electro-optic modulator. The frequency-sparse microwave sweep signal generated by the microwave source modulates the incoherent optical carrier in the electro-optic modulator to obtain a frequency-sparse probe optical signal. The frequency-sparse microwave sweep signal is composed of M frequency points randomly selected from a frequency range with N frequency points, where M << N. The probe optical signal enters the optical link under test through an optical circulator for reflection measurement. Since there are multiple reflection points in the optical link, the reflection points at different locations will have different amplitudes and time delays. The reflected probe optical signal carries the amplitude and time delay information of these reflection points. After filtering and amplification, it is converted into a microwave signal with sparse frequency carrying time delay information by the photodetector. The signal acquisition and processing module performs digital sampling and quantization on the microwave signal and constructs a measurement matrix and a sparse orthogonal basis matrix by analyzing the frequency domain expression in the microwave signal. Then, the sparse signal, i.e., the optical time delay of each channel, is reconstructed using a compressed sensing signal reconstruction algorithm.

[0038] Let ω m ω represents the frequency of the microwave sweep signal, with a total of M sweep points. c If the center frequency of the incoherent optical carrier is given, then the expression for the intensity modulation signal returned in the fiber optic link under test is:

[0039]

[0040] Where S is the number of reflection points, A S Let τ be the amplitude of the s-th reflected signal. s Let E be the time delay of the s-th reflected signal, and E be the modulation coefficient.

[0041] After being amplified and filtered, the reflected light enters the photodetector, is converted into a photocurrent, and is received by the signal acquisition and processing module. The recovered microwave signal can be represented in the frequency domain as follows:

[0042]

[0043] In the formula, η is the responsivity of the photoelectric detection module, and G is the amplification factor of the optical amplification module.

[0044] In traditional schemes, the time delay value corresponding to the peak value of the time-domain spectrum represents the time delay of each channel, and other time delay values ​​can be approximated as zero. In this case, the time-domain signal has sparse characteristics, so the time delay can be calculated using compressed sensing algorithms.

[0045] The multi-channel optical delay measurement technology based on compressed sensing is specifically expressed as follows:

[0046] I=ΦD+ε (3)

[0047] In the formula, the sampling matrix is ​​I=[I(ω1),I(ω2),..,I(ω M )] T Φ is an M×N order measurement matrix, ε is an additive noise vector, and the original signal matrix is ​​a microwave linear sweep frequency signal, expressed as:

[0048]

[0049] D=[D(ω st +Δω),D(ω st +2Δω),..,D(ω) st +NΔω)] T

[0050] Where ω st This is the starting frequency.

[0051] The measurement matrix Φ is constructed as follows: An identity matrix B∈eye(N, N) is generated, where each row of the identity matrix corresponds to the position information of a frequency point. This is achieved by sampling the frequency points ω1, ω2...ω... M At full frequency ω st +Δω,ω st +2Δω...ω st The position information of +NΔω is used to construct a matrix, that is, the corresponding M rows are extracted to construct an M×N order measurement matrix Φ.

[0052] Equation (3) can be further decomposed into:

[0053] I=Φψθ+ε (5)

[0054] ψ is a sparse matrix, which in this scheme is represented by the discrete Fourier transform matrix, as follows:

[0055]

[0056] Among them W N =exp(-j2π / N).

[0057] From equation (5), we know that D = ψθ, where θ is the transformation vector of D. When the transformation vector θ has at most K non-zero elements, the signal D is K-sparse in the ψ domain, where K is the sparsity.

[0058] The main purpose of compressed sensing is to accurately reconstruct the original signal D using the measured value I. Since M << N, there are infinitely many D ∈ R when solving for I. N To ensure the uniqueness of the solution, the original signal of compressed sensing must be sparse or compressible, which transforms the above problem into solving the minimum norm l0 problem.

[0059]

[0060] Where ||θ||0 represents the l0 norm of θ, that is, the number of non-zero elements of θ.

[0061] However, minimizing the l0 norm is an NP-hard problem, and it is often non-convex, requiring the exploration of all possibilities, which is computationally difficult. Therefore, equation (7) can be transformed into the problem of finding the minimum l1 norm:

[0062]

[0063] There are currently several methods for solving equations (7) and (8). This paper selects the Orthogonal Matching Pursuit (OMP) algorithm to solve the compressed sensing signal reconstruction. Its general operation is as follows:

[0064] Input signals: Observation I, sensing matrix A = Φψ, sparsity K;

[0065] (1) Initialize residual r0 = I, index set The number of iterations t=1, the matrix

[0066] (2) Find the index value λ of the atom that best matches the residual vector, such that

[0067] Where a j Let j be the j-th column of matrix A;

[0068] (3) Update index set Λ t =Λ t-1 ∪λ t Update matrix A t =A t-1 ∪a t ;

[0069] (4)Solve I=A t θ t The least squares solution is such that

[0070] (5) Update residuals

[0071] (6) If t≥K, the iteration ends and the loop is exited; otherwise, t=t+1, and the process returns to step (2).

[0072] (7) Finally, complete the iterative output signal.

[0073] The solution obtained by the OMP algorithm is θ = [τ1, τ2, ..., τ]. s ] TThis refers to the time delay and location information of each reflection point.

[0074] To further improve the time-domain resolution, the present invention can further process the obtained θ as follows: windowing is performed on the sparse time-domain signal θ to suppress spectral leakage, and the windowed signal is subjected to FFT (Fast Fourier Transform) to convert the time-domain signal into a frequency-domain signal, and zero-padding is performed in the frequency domain; the signal after zero-padding in the frequency domain is then subjected to inverse Fourier Transform (IFFT) to convert the frequency-domain signal back into a time-domain signal.

[0075] Because zero-padding in the frequency domain increases the number of FFT points, the number of time-domain signal points after the IFFT also increases accordingly, thereby improving the time-domain resolution. The process of windowing the time-domain FFT to the frequency domain and then performing an IFFT back to the time domain after zero-padding in the frequency domain is an effective method to improve signal measurement accuracy. Windowing can reduce spectral leakage and improve spectral characteristics, while zero-padding in the frequency domain can improve frequency-domain resolution and alleviate the picket fence effect. The increased number of time-domain signal points after the IFFT improves resolution, thus enhancing measurement accuracy.

Claims

1. A multi-channel optical delay measurement method based on compressed sensing, used to simultaneously measure the optical delay of S different reflection points or S parallel channels in the optical link under test, where S is an integer greater than or equal to 3; characterized in that, The method includes the following steps: S1. Intensity modulation of an incoherent optical carrier is performed using a microwave sweep signal consisting of M frequency points randomly selected from a frequency range of N frequency points, resulting in a frequency-sparse probe optical signal, M < <N; S2. The probe light signal is converted into a photoelectric signal after passing through the optical link under test to obtain a microwave signal with sparse frequency carrying time delay information. S3. Using the frequency response obtained by sampling the microwave signal with sparse frequencies carrying time delay information in the digital domain as the observation value I, and the discrete Fourier transform matrix as the sparse matrix ψ, a measurement matrix Φ is constructed based on the position information of each sweep frequency point in the microwave sweep signal with sparse frequencies in the full frequency range. According to the formula I = Φψθ, the time-domain response information of the sparse signal θ is reconstructed using the compressed sensing reconstruction method, thereby obtaining the optical time delays τ1 to τ2 of the S different reflection points or S parallel channels. S .

2. The multi-channel optical delay measurement method based on compressed sensing as described in claim 1, characterized in that, Higher precision optical delay τ1′~τ S Specifically, the optical delay τ1′~τ is obtained by: windowing the time-domain pulses corresponding to each channel of θ, converting the windowed signal to the frequency domain, zero-padding the converted frequency-domain signal, and then converting the zero-padding frequency-domain signal back to the time domain to obtain a higher precision optical delay τ1′~τ. S ′.

3. The multi-channel optical delay measurement method based on compressed sensing as described in claim 1 or 2, characterized in that, The incoherent optical carrier is a broadband optical signal generated by an amplified spontaneous emission source.

4. The multi-channel optical delay measurement method based on compressed sensing as described in claim 1 or 2, characterized in that, The specific method for constructing the measurement matrix Φ is as follows: Generate an N*N identity matrix, where the position of each row of the identity matrix corresponds to the position information of N frequency points in the frequency range; extract the M frequency points in the microwave sweep signal from the M rows corresponding to the identity matrix to construct an M×N order measurement matrix Φ.

5. The multi-channel optical delay measurement method based on compressed sensing as described in claim 1 or 2, characterized in that, The compressed sensing reconstruction method is an orthogonal matching pursuit algorithm.

6. A multi-channel optical delay measurement device based on compressed sensing, used to simultaneously measure the optical delay of S different reflection points or S parallel channels in the optical link under test, where S is an integer greater than or equal to 3; characterized in that, The device includes: The intensity modulation module is used to intensity modulate an incoherent optical carrier using a microwave sweep signal consisting of M randomly selected frequency points from a frequency range of N frequency points, to obtain a frequency-sparse probe optical signal, where M < <N; The photoelectric detection module is used to cause the detection light signal to undergo photoelectric conversion after passing through the optical link under test, so as to obtain a microwave signal with sparse frequency carrying time delay information; The signal acquisition and processing module is used to take the frequency response obtained by sampling the microwave signal with sparse frequencies carrying time delay information in the digital domain as the observation value I, and the discrete Fourier transform matrix as the sparse matrix ψ. Based on the position information of each sweep frequency point in the microwave sweep signal with sparse frequencies in the full frequency range, a measurement matrix Φ is constructed. According to the formula I = Φψθ, the time-domain response information of the sparse signal θ is reconstructed using compressed sensing reconstruction method, thereby obtaining the optical time delays τ1 to τ2 of the S different reflection points or S parallel channels. S .

7. The multi-channel optical delay measurement device based on compressed sensing as described in claim 6, characterized in that, Higher precision optical delay τ1′~τ S Specifically, the optical delay τ1′~τ is obtained by: windowing the time-domain pulses corresponding to each channel of θ, converting the windowed signal to the frequency domain, zero-padding the converted frequency-domain signal, and then converting the zero-padding frequency-domain signal back to the time domain to obtain a higher precision optical delay τ1′~τ. S ′.

8. The multi-channel optical delay measurement device based on compressed sensing as described in claim 6 or 7, characterized in that, The incoherent optical carrier is a broadband optical signal generated by an amplified spontaneous emission source.

9. The multi-channel optical delay measurement device based on compressed sensing as described in claim 6 or 7, characterized in that, The specific method for constructing the measurement matrix Φ is as follows: Generate an N*N identity matrix, where the position of each row of the identity matrix corresponds to the position information of N frequency points in the frequency range; extract the M frequency points in the microwave sweep signal from the M rows corresponding to the identity matrix to construct an M×N order measurement matrix Φ.

10. The multi-channel optical delay measurement device based on compressed sensing as described in claim 6 or 7, characterized in that, The compressed sensing reconstruction method is an orthogonal matching pursuit algorithm.

Images