Photon counting ocean current field velocity measurement sequence length adaptive selection method and system

Abstract

The application belongs to the field of underwater photon counting ocean current field velocity measurement, and particularly relates to a photon counting ocean current field velocity measurement sequence length self-adaptive selection method and system, and proposes a method for adaptively determining the optimal sequence length of photon arrival by using a sliding window to realize real-time inversion of high-precision time-varying flow field Doppler velocity. By calculating the high-order statistics of the time interval of adjacent photons, the time-varying disturbance factor and the instantaneous photon counting rate density are obtained. A target evaluation function is designed to adaptively solve the optimal reconstruction data length that meets the current local flow field characteristics, and the ocean current field Doppler frequency shift is calculated through 1-bit compressed sensing reconstruction to complete the velocity inversion. The application quantitatively balances the physical contradiction between the information sparsity bottom line and the frequency domain expansion, and greatly improves the inversion accuracy of time-varying ocean current velocity measurement.

Description

Technical Field

[0001] This application belongs to the field of underwater photon counting ocean current field velocity measurement, specifically relating to an underwater photon counting ocean current field Doppler velocity inversion method and system. Background Technology

[0002] Accurate ocean current velocity detection is crucial for autonomous navigation of underwater unmanned vehicles (AUVs), cold seep hydrothermal flux analysis, and deep-sea development. Photon-counting coherent Doppler velocimetry based on a single-photon avalanche diode (SPAD) utilizes the SPAD detector's nonlinear Poisson response and hardware dead-time constraints to convert extremely weak ocean coherent echo signals into a 1-bit binary discrete pulse timestamp sequence. Doppler beat frequencies are then calculated using 1-bit compressed sensing, enabling long-distance, high-precision ocean current velocity measurement.

[0003] However, in actual deep-sea environments, ocean current velocities typically exhibit strong time-varying characteristics. This presents a dilemma for 1-bit compressed sensing spectral reconstruction in terms of selecting the photon sequence length (time window): if the photon pulse sequence selected for a single reconstruction is too long (the observation time window is too wide), the Doppler frequency shift will accumulate in the time domain under the influence of the time-varying flow field, resulting in severe spectral broadening and a significant decrease in velocity measurement accuracy; if the photon pulse sequence selected for a single reconstruction is too short (the observation time window is too narrow), due to the intrinsic sparsity of photon counting detection, the number of effective photon events within the time window is severely insufficient, causing the observation matrix constraints required for compressed sensing reconstruction to be unsatisfied, and the algorithm to fail to converge.

[0004] Conventional methods using fixed photon sequence lengths are ill-suited to the complex marine environment characterized by variable turbidity and drastic time-varying flow velocities. Therefore, adaptively selecting the optimal data length based on the intensity characteristics of real-time returned photon signals, while suppressing the cumulative effects of time-varying flow fields and maintaining the signal-to-noise ratio of sparse photon count reconstruction, is a pressing technical challenge. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a method and system for adaptively selecting the length of a photon-counting ocean current field velocity measurement sequence. This method extracts the temporal statistical features of the arriving photon sequence to dynamically optimize the truncation length of the ocean current field photon arrival sequence, thereby retrieving the ocean current flow and balancing algorithm convergence with the accuracy of the ocean current field inversion. The specific technical solution is as follows: An adaptive selection method for the length of a photon-counted ocean current field velocity measurement sequence includes the following steps: Step 1: Complete the initial acquisition of the sliding window photon counting arrival sequence; Step 2: Extract the initial sliding time window The photon counting event time statistics characteristics achieved within the time window are used to calculate the photon count density in real time. With the time-varying perturbation factor of the flow field ; Step 3: Construct the joint optimal data constant objective evaluation function The photon counting density Flow field time-varying disturbance factor and the detector's inherent dead time As a constraint variable, the current optimal reconstructed data length is determined. ; Step 4: Using the current end time of the sliding window as a reference, dynamically extract a length along the historical timeline. Photon counting events arrive in a subset of the sequence ; Step 5: Deliver the captured photons to the arrival sequence subset The coherent Doppler frequency of the ocean current field is recovered from the 1-bit compressed sensing reconstruction algorithm, thus completing the inversion of the time-varying ocean current velocity.

[0006] Preferably, in step 2, the photon count density within the current time window The calculation equation is as follows: ; in, The total photon count events within the initial sliding time window. The fixed total duration of the initial sliding time window.

[0007] Preferably, in step 2, the time-varying perturbation factor of the flow field Extraction is achieved by calculating the time interval between all adjacent photons within the current window. The calculation equation is obtained from the time statistical characteristics and is as follows: ; In the formula, This indicates the calculation of variance. This indicates the calculation of mathematical expectation; the stated Used to quantitatively characterize the degree of photon stability caused by ocean current turbulence modulation.

[0008] Preferably, in step 3, the joint optimal data length objective evaluation function is used. The specific mathematical model is as follows: ; in, To ensure the minimum theoretical number of photons required for the 1-bit compressed sensing algorithm to converge in a noise-free environment, To fix the dead time for the hardware of a single-photon detector, As the frequency shift reference for the acousto-optic modulator, and The preset normalized weight adjustment coefficient satisfies , It is a dimensionless empirical constant.

[0009] Preferably, based on calculations At the current end time of the sliding window Based on this, the forward dynamic truncation length is... time series This sequence serves as a random sequence for completing 1-bit compressed sensing spectrum reconstruction; the dynamically truncated time series By constructing a binary discrete observation matrix that considers the dead-time effect, and using algorithms such as 1-bit iterative hard thresholding, high-precision sparse spectrum reconstruction is performed to search for the maximum resonance matching frequency point. .

[0010] Preferably, the radial velocity of the current is calculated using the Doppler velocity inversion equation. : ; in, The wavelength is the laser wavelength. After the calculation is completed, the sliding window advances forward to enter the adaptive solution loop of the next time period.

[0011] Preferably, in step 1, the detector receives a coherently mixed optical field of the frequency-shifted local oscillator light from the acousto-optic modulator and the echo signal light modulated by the ocean current field. Through the random response to the mixed optical field, it outputs a 1-bit binary pulse sequence, and records the sequence of photon arrival events that are effectively triggered by the photon counting and acquisition card. Set a basic sliding time window Its window length is The total number of photons in the current sliding window is .

[0012] A system for selecting the optimal sequence length for photon counting ocean current field velocity measurement includes a data acquisition module, a calculation module, and an output module; Data acquisition module: The detector receives the coherent mixed light field of the frequency-shifted local oscillator light from the acousto-optic modulator and the echo signal light modulated by the ocean current field. Through the random response to the mixed light field, it outputs a 1-bit binary pulse sequence; the photon counting acquisition card records the arrival sequence of validly triggered photon counting events. The computation module obtains the time-varying perturbation factor and instantaneous photon count rate density by calculating the higher-order statistics of the time interval between adjacent photons; it designs an objective evaluation function, adaptively solves for the optimal reconstructed data length that satisfies the current local flow field characteristics, and calculates the Doppler frequency shift of the ocean flow field through 1-bit compressed sensing reconstruction to complete the velocity inversion. Output module: Visualizes the results.

[0013] Compared with the prior art, the beneficial effects of this application are as follows: This invention breaks through the limitations of traditional fixed-length data truncation. In weak light / stable flow fields, it automatically increases the window to concentrate photons and ensure reconstruction convergence. In strong light / intense turbulence, it automatically shrinks the window to lock high instantaneous time-domain echo signals, suppressing spectral broadening caused by time-varying flow velocity. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the processing flow of this application. Detailed Implementation

[0015] The technical solution of this application will be described in detail below with reference to specific embodiments and accompanying drawings. It should be understood that the embodiments and specific features in the embodiments are detailed descriptions of the technical solution of this application, rather than limitations thereof. Specific technical features can be combined with each other.

[0016] Figure 1 As shown, a method for selecting the optimal sequence length in photon counting ocean current field velocity measurement is presented. This method dynamically optimizes the truncation length of the photon arrival sequence in the ocean current field. By extracting the temporal statistical features of the arrival photon sequence, the ocean current field velocity is inverted, balancing the convergence of the algorithm with the accuracy of the ocean current field inversion. The specific technical solution is as follows: Based on the time-varying characteristics of ocean current fields, a sliding window acquisition method for photon arrival event sequences is designed. A SPAD detector receives the coherently mixed optical field of the frequency-shifted local oscillator light from the acousto-optic modulator and the echo signal light modulated by the ocean current field. Through the random response to the mixed optical field, a 1-bit binary pulse sequence is output, and the sequence of effectively triggered photon arrival events is recorded using a Time-Correlated Single Photon Counting (TCSPC) card. Set a basic sliding time window. Its window length is The total number of photons in the current sliding window is .

[0017] Using the initial sliding event window The data within the time window is used to extract physical indicators characterizing the stationarity of the photon sequence. The number of photon counting events within the current time window is counted, and the real-time photon count density is calculated. This serves as a reference point to characterize the light intensity of the current detection environment. Simultaneously, the current window is calculated. Time interval between arrivals of adjacent photons Extracting the time interval sequence of adjacent photons. The temporal statistical characteristics are used to calculate the light intensity variation factor, which is then used to assess the degree of short-term mixing light field variation in the ocean current field. ; in, For time-varying perturbation factors, This indicates the calculation of variance. This indicates the calculation of mathematical expectation.

[0018] Optimal data length The selection of the time-varying flow field is essentially the result of a trade-off and compromise between two levels of physical mechanisms: signal sparsity constraints (requiring the event window to be as long as possible to gather enough photons to trigger algorithm convergence) and flow field time-varying constraints (requiring the time window to be as short as possible to suppress spectral broadening). Therefore, a time-varying flow field ... Time-varying disturbance factor and SPAD's inherent dead time Joint optimal data length objective evaluation function : ; in, The left-hand side of the formula is the sparsity constraint term. To ensure the minimum theoretical photon count required for the 1-bit compressed sensing algorithm to converge in a noise-free environment, the SPAD dead time is limited at high count rates. This can lead to counting saturation, and a correction term is introduced into the denominator. This ensures that even when the light intensity is extremely low and the counting rate is low... During descent, the system can drive Automatically stretched to ensure sufficient photon arrival events are counted. The right-hand term is the time-varying suppression term. This is the modulation fundamental frequency of the acousto-optic modulator. This is a dimensionless empirical constant. When the flow field fluctuates drastically, leading to a time-varying perturbation factor... When the denominator increases, the denominator increases sharply, driving... It automatically shortens, thus locking the field of view within a very short time window, avoiding severe accumulation of Doppler frequency shift in the time domain and causing spectral superposition. Normalized adjustment coefficient for weights It is used to balance the signal-to-noise ratio and inversion accuracy.

[0019] Based on the calculation At the current end time of the sliding window Based on this, the forward dynamic truncation length is... time series This sequence serves as a random sequence for completing 1-bit compressed sensing spectrum reconstruction.

[0020] Dynamically extracted time series By constructing a binary discrete observation matrix that considers the dead time effect, high-precision sparse spectrum reconstruction is performed using algorithms such as 1-bit iterative hard thresholding to search for the maximum resonance matching frequency point, and the current radial velocity of the ocean current is calculated using the Doppler velocity inversion equation. : ; in, The wavelength is the laser wavelength. After the calculation is completed, the sliding window advances forward to enter the adaptive solution loop of the next time period.

[0021] A system for selecting the optimal sequence length for photon counting ocean current field velocity measurement includes a data acquisition module, a calculation module, and an output module; Data acquisition module: The detector receives the coherent mixed light field of the frequency-shifted local oscillator light from the acousto-optic modulator and the echo signal light modulated by the ocean current field. Through the random response to the mixed light field, it outputs a 1-bit binary pulse sequence; the photon counting acquisition card records the arrival sequence of validly triggered photon counting events. The computation module obtains the time-varying perturbation factor and instantaneous photon count rate density by calculating the higher-order statistics of the time interval between adjacent photons; it designs an objective evaluation function to adaptively solve for the optimal reconstructed data length that satisfies the current local flow field characteristics; and it calculates the Doppler frequency shift of the ocean flow field through 1-bit compressed sensing reconstruction to complete the velocity inversion. Output module: Visualizes the results.

[0022] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for adaptively selecting the length of a photon-counted ocean current field velocity measurement sequence, characterized in that, Includes the following steps: Step 1: Complete the initial acquisition of the sliding window photon counting arrival sequence; Step 2: Extract the initial sliding time window The photon counting event time statistics characteristics achieved within the time window are used to calculate the photon count density in real time. With the time-varying perturbation factor of the flow field ; Time-varying perturbation factor of flow field Extraction is achieved by calculating the time interval between all adjacent photons within the current window. The calculation equation is obtained from the time statistical characteristics and is as follows: ； In the formula, This indicates the calculation of variance. This indicates the calculation of mathematical expectation; the stated Used to quantitatively characterize the degree of photon stability caused by ocean current turbulence modulation; Step 3: Construct the joint optimal data constant objective evaluation function The photon counting density Flow field time-varying disturbance factor and the detector's inherent dead time As a constraint variable, the current optimal reconstructed data length is determined. ; Joint optimal data length objective evaluation function The specific mathematical model is as follows: ； in, To ensure the minimum theoretical number of photons required for the 1-bit compressed sensing algorithm to converge in a noise-free environment, To fix the dead time for the hardware of a single-photon detector, As the frequency shift reference for the acousto-optic modulator, and The preset normalized weight adjustment coefficient satisfies , It is a dimensionless empirical constant; Based on the calculation At the current end time of the sliding window Based on this, the forward dynamic truncation length is... time series This sequence serves as a random sequence for completing 1-bit compressed sensing spectrum reconstruction; the dynamically truncated time series By constructing a binary discrete observation matrix that considers the dead-time effect, and using algorithms such as 1-bit iterative hard thresholding, high-precision sparse spectrum reconstruction is performed to search for the maximum resonance matching frequency point. ; Step 4: Using the current end time of the sliding window as a reference, dynamically extract a length along the historical timeline. Photon counting events arrive in a subset of the sequence ; Step 5: Deliver the captured photons to the arrival sequence subset The coherent Doppler frequency of the ocean current field is recovered from the 1-bit compressed sensing reconstruction algorithm, thus completing the inversion of the time-varying ocean current velocity; The radial velocity of the current ocean current was calculated using the Doppler velocity inversion equation. : ； in, The wavelength is the laser wavelength. After the calculation is completed, the sliding window advances forward to enter the adaptive solution loop of the next time period.

2. The adaptive selection method for the length of the photon-counting ocean current field velocity measurement sequence according to claim 1, characterized in that, In step 2, the photon count density within the current time window The calculation equation is as follows: ； in, The total photon count events within the initial sliding time window. The fixed total duration of the initial sliding time window.

3. The adaptive selection method for the length of the photon-counting ocean current field velocity measurement sequence according to claim 1, characterized in that: In step 1, the detector receives the coherently mixed optical field of the frequency-shifted local oscillator light from the acousto-optic modulator and the echo signal light modulated by the ocean current field. Through the random response to the mixed optical field, it outputs a 1-bit binary pulse sequence, and records the sequence of photon arrival events that are effectively triggered by the pulse counting and acquisition card. Set a basic sliding time window Its window length is The total number of photons in the current sliding window is .

4. A photon-counting ocean current field velocity sequence length adaptive selection system, adapted to the method described in any one of claims 1-3, characterized in that, It includes a data acquisition module, a calculation module, and an output module; Data acquisition module: The detector receives the coherent mixed light field of the frequency-shifted local oscillator light from the acousto-optic modulator and the echo signal light modulated by the ocean current field. Through the random response to the mixed light field, it outputs a 1-bit binary pulse sequence; the photon counting acquisition card records the arrival sequence of validly triggered photon counting events. The computation module obtains the time-varying perturbation factor and instantaneous photon count rate density by calculating the higher-order statistics of the time interval between adjacent photons; it designs an objective evaluation function, adaptively solves for the optimal reconstructed data length that satisfies the current local flow field characteristics, and calculates the Doppler frequency shift of the ocean flow field through 1-bit compressed sensing reconstruction to complete the velocity inversion. Output module: Visualizes the results.

Images