Pacific cod natural population migration path monitoring method and system

Abstract

The invention provides a Pacific Ocean cod natural population migration path monitoring method and system. The method comprises the following steps: collecting a coding sound signal emitted by a cod population carrying an acoustic tag at a target node through a double-acoustic sensor vertical receiving array; determining an autocorrelation function of each channel signal in the coded sound signal at the target node and a cross-correlation function between the channel signals, and separating a plurality of signal components which are aliasing due to a seawater medium frequency dispersion effect based on a bending frequency dispersion track in a time-frequency domain of each autocorrelation function and each cross-correlation function; screening out a modal order set of dominant signal components, and converting all signal components into distance information from a sound source to a vertical receiving array of a target node double acoustic sensor through the modal order set; and reconstructing a three-dimensional migration path of the Pacific Ocean cod population in the target sea area according to the distance information of all the nodes. By adopting the scheme of the invention, the order leading the cod individual sound signal propagation mode can be identified from the sea area aliasing sound signals.

Description

Methods and Systems for Monitoring the Migration Routes of Natural Pacific Cod Populations Technical Field

[0001] This application relates to the field of fish monitoring technology, and more specifically, to a method and system for monitoring the natural migration routes of Pacific cod populations. Background Technology

[0002] Monitoring the migration routes of natural marine populations is a crucial foundation for fisheries resource management and marine ecological protection. Traditional monitoring methods, such as manual observation, satellite remote sensing, and trawling surveys, have limitations, including limited monitoring range, low spatiotemporal resolution, and significant disturbance to fish behavior. These methods struggle to achieve accurate, continuous, and non-invasive tracking of large-scale, long-term, and three-dimensional migration behaviors. In recent years, passive monitoring methods based on underwater acoustic technology have gained increasing attention. These methods identify and locate individuals by receiving coded signals emitted by fish carrying acoustic tags, providing a new technological approach for studying the migration of natural populations.

[0003] In complex shallow-sea waveguides, acoustic signals propagate through multiple paths, and normal modes of different orders interfere and alias at the receiving point, forming a complex composite signal. Although existing signal processing techniques (such as time-frequency analysis) can reveal the dispersion structure of the signal to some extent, they lack the ability to identify the dominant mode order combination in the current sound field. Misjudgment or ambiguity of the mode order will directly lead to multiple or even incorrect solutions in the subsequent sound source distance inversion, making the positioning results unreliable. Existing solutions either rely on strong assumptions about prior knowledge of the environment and sound sources or are limited to information from a single receiving point and lack spatial resolution. When dealing with unknown and dynamically changing natural population monitoring scenarios, their robustness and universality are significantly insufficient. This seriously restricts the ability to achieve accurate and reliable three-dimensional migration path reconstruction using passive acoustic technology. Therefore, how to identify the order of the dominant cod individual acoustic signal propagation mode from the aliased acoustic signals in the sea has become a difficult problem for the industry. Summary of the Invention

[0004] This application provides a method and system for monitoring the natural migration routes of Pacific cod populations, which can identify the order of the dominant acoustic signal propagation mode of individual cod from the superimposed acoustic signals in the sea area.

[0005] In a first aspect, this application provides a method for monitoring the natural migration path of Pacific cod populations. The method involves pre-deploying a dual-acoustic sensor vertical receiving array with multiple nodes in a target sea area. The method includes: selecting one node in the target sea area as the target node; collecting coded acoustic signals emitted by a cod population carrying acoustic tags at the target node using the dual-acoustic sensor vertical receiving array; determining the autocorrelation function of each channel signal and the cross-correlation function between each channel signal in the coded acoustic signal at the target node; separating multiple signal components aliased due to the dispersion effect of the seawater medium based on the curved dispersion trajectory in the time-frequency domain of each autocorrelation function and cross-correlation function; selecting the mode order set of the dominant signal components; converting all signal components into distance information from the sound source to the dual-acoustic sensor vertical receiving array at the target node using the mode order set; further determining the distance information from the sound source to the dual-acoustic sensor vertical receiving arrays at the remaining nodes; and reconstructing the three-dimensional migration path of the Pacific cod population in the target sea area based on all the distance information and the time difference of the signals arriving at different nodes.

[0006] In some embodiments, before acquiring the coded acoustic signals emitted by a cod population carrying acoustic tags via a dual hydrophone vertical receiving array, the method further includes: pre-deploying multiple individual cod carrying acoustic tags to the target sea area.

[0007] In some embodiments, determining the autocorrelation function of each channel signal and the cross-correlation function between each channel signal in the coded acoustic signal at the target node specifically includes: acquiring a set of dual-channel signal data in the coded acoustic signal at the target node; preprocessing the signal data of each channel to obtain a stable data segment for each channel; determining the autocorrelation function of the stable data segment corresponding to each channel; and determining the cross-correlation function between the stable data segments corresponding to two different channels.

[0008] In some embodiments, separating multiple signal components aliased due to the dispersion effect of seawater medium based on the curved dispersion trajectory in the time-frequency domain of each autocorrelation function and cross-correlation function specifically includes: a preset reference distance for sound wave propagation in the target sea area; converting the curved dispersion trajectory in each autocorrelation function and cross-correlation function into a straight line parallel to the frequency axis based on the preset reference distance, thereby obtaining the frequency domain signal of each autocorrelation function and cross-correlation function; extracting characteristic frequency peaks from each frequency domain signal, thereby obtaining multiple signal components aliased due to the dispersion effect of seawater medium.

[0009] In some embodiments, selecting the modal order set of the dominant signal component specifically includes: extracting the main characteristic frequency peak with the largest amplitude from all autocorrelation functions and cross-correlation functions; obtaining a theoretical characteristic frequency set; matching all main characteristic frequency peaks with the theoretical characteristic frequency set to obtain a candidate modal order set that meets the frequency characteristics; and selecting the candidate modal order set by the constraint relationship between the characteristic frequency peaks of all autocorrelation functions and cross-correlation functions to obtain the modal order set of the dominant signal component.

[0010] In some embodiments, converting all signal components into distance information from the sound source to the target node's dual acoustic sensor vertical receiving array by means of the modal order set specifically includes: for each group of modal orders in the modal order set, obtaining the observed characteristic frequency value of the signal component corresponding to each group of modal orders; determining the theoretical characteristic frequency value corresponding to each group of modal orders; and determining the distance information from the sound source to the target node's dual acoustic sensor vertical receiving array based on the theoretical dispersion relation corresponding to each group of modal orders, the observed characteristic frequency value, the theoretical characteristic frequency value, and a preset reference distance.

[0011] In some embodiments, reconstructing the three-dimensional migration path of the Pacific cod population in the target sea area based on the time differences of all distance information and signals arriving at different nodes specifically includes: acquiring all distance information and the time differences of all signals arriving at different nodes; inputting all distance information and the time differences of all signals arriving at different nodes into a positioning solution model to obtain the spatial position of individual cod in the target sea area; repeatedly monitoring to obtain the spatial position of individual cod in the target sea area at different times, and fitting all spatial positions into the movement trajectory of individual cod; continuing to determine the movement trajectory of other individual cod, and reconstructing all movement trajectories into the three-dimensional migration path of the cod population in the target sea area.

[0012] Secondly, this application provides a monitoring system for the natural migration path of Pacific cod populations, comprising: a data acquisition module, used to select a node in a target sea area as a target node, and to acquire coded acoustic signals emitted by a cod population carrying acoustic tags at the target node through a dual acoustic sensor vertical receiving array; a processing module, used to determine the autocorrelation function and cross-correlation function of each channel signal in the coded acoustic signal at the target node, and to separate multiple signal components aliased due to the dispersion effect of seawater medium based on the curved dispersion trajectory in the time-frequency domain of each autocorrelation function and cross-correlation function; the processing module is further used to filter out the mode order set of the dominant signal component, and to convert all signal components into distance information from the sound source to the dual acoustic sensor vertical receiving array at the target node through the mode order set, and to further determine the distance information from the sound source to the dual acoustic sensor vertical receiving arrays at the remaining nodes; and an execution module, used to reconstruct the three-dimensional migration path of the Pacific cod population in the target sea area based on all the distance information and the time difference of the signals arriving at different nodes.

[0013] Thirdly, this application provides a computer device, the computer device including a memory and a processor, the memory storing code, and the processor being configured to acquire the code and execute the above-described method for monitoring the natural migration paths of Pacific cod populations.

[0014] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for monitoring the natural migration routes of Pacific cod populations.

[0015] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects: In the method and system for monitoring the natural migration path of Pacific cod populations provided in this application, a node in the target sea area is first selected as the target node. At the target node, the coded acoustic signals emitted by the cod population carrying acoustic tags are collected through a dual acoustic sensor vertical receiving array. The autocorrelation function of each channel signal and the cross-correlation function between each channel signal in the coded acoustic signal at the target node are determined. Based on the curved dispersion trajectory in the time-frequency domain of each autocorrelation function and cross-correlation function, multiple signal components superimposed due to the dispersion effect of the seawater medium are separated. The mode order set of the dominant signal component is selected. All signal components are converted into distance information from the sound source to the dual acoustic sensor vertical receiving array at the target node through the mode order set. The distance information from the sound source to the dual acoustic sensor vertical receiving arrays at the remaining nodes is further determined. The three-dimensional migration path of the Pacific cod population in the target sea area is reconstructed based on all the distance information and the time difference of the signals arriving at different nodes.

[0016] Therefore, this application constructs an information sensing network with spatial distribution and vertical structure by deploying a dual-hydrophone vertical receiving array with multiple nodes in the target sea area. This enhances the spatial dimension and reliability of the sound field information at the data acquisition level, providing a physical basis for distinguishing aliased modes. Secondly, by calculating the autocorrelation function of each channel signal at the target node and the cross-correlation function between channels, and separating multiple signal components aliased due to seawater dispersion effects based on their curved dispersion trajectories in the time-frequency domain, this scheme achieves decoupling and feature extraction of complex composite signals, overcoming the shortcomings of traditional methods in insufficient feature information extraction under aliasing conditions. Furthermore, this application selects the dominant signal component mode based on the constraint relationship of characteristic frequency peaks between all autocorrelation functions and cross-correlation functions. This method utilizes the inherent correlation of modal orders in the vertical space and cross-correlation domain of the sound field to construct a discrimination mechanism independent of prior assumptions about the sound source or environment. This enables robust and unique identification of dominant modal order combinations without requiring strong external assumptions, fundamentally eliminating the multi-valued ranging problem caused by modal ambiguity. Based on this, signal components are converted into distance information through theoretical dispersion relations, and combined with the orientation and time difference of arrival between multiple nodes, a high-confidence three-dimensional migration path is finally reconstructed. This forms a complete technical chain from array deployment, signal demixing, modal identification to positioning reconstruction, systematically improving the ability and reliability of continuous, accurate, and non-invasive three-dimensional monitoring of fish migration paths in complex shallow sea environments. In summary, the scheme proposed in this application can identify the order of the dominant cod individual acoustic signal propagation mode from aliased acoustic signals in the sea area. Attached Figure Description

[0017] Figure 1 is an exemplary flowchart of a method for monitoring the natural migration path of Pacific cod populations according to some embodiments of this application; Figure 2 is an exemplary flowchart of separating signal components according to some embodiments of this application; Figure 3 is an exemplary flowchart of determining a three-dimensional migration path according to some embodiments of this application; Figure 4 is a schematic diagram of a system for monitoring the natural migration path of Pacific cod populations according to some embodiments of this application; Figure 5 is a schematic diagram of a computer device for implementing the method for monitoring the natural migration path of Pacific cod populations according to some embodiments of this application. Detailed Implementation

[0018] To better understand the technical solution of this application, the technical solution of this application will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0019] Referring to Figure 1, which is an exemplary flowchart of a method for monitoring the natural migration path of Pacific cod populations according to some embodiments of this application, the method mainly includes the following steps: In step 101, a node in the target sea area is selected as the target node, and the coded acoustic signals emitted by the cod population carrying acoustic tags are collected at the target node through a dual acoustic sensor vertical receiving array.

[0020] In some embodiments, before selecting a node in the target sea area as the target node, a dual acoustic sensor vertical receiving array with multiple nodes is pre-deployed in the target sea area. As a preferred embodiment, the dual acoustic sensor vertical receiving array with multiple nodes pre-deployed in the target sea area in this application can be implemented by the following steps: pre-setting multiple nodes in the target sea area; and deploying two acoustic sensors at each node at a preset vertical interval to form a dual acoustic sensor vertical receiving array.

[0021] In specific implementation, the pre-setting of multiple nodes in the target sea area can be achieved in the following way: First, based on historical fishery data, known migration corridors, and marine hydrological models, the geographical coordinates of the nodes are initially planned along the possible migration routes of cod, such as the edge of the continental shelf, specific isobaths, or the periphery of known spawning / feeding grounds. Then, an acoustic propagation model is used to simulate the coverage range and received signal-to-noise ratio of acoustic signals at these candidate points, and the node positions are optimized to ensure that the network can effectively cover the target waters and form positioning baselines on key channels. Finally, the latitude and longitude coordinates of each node are determined. In other embodiments, a uniform sampling method can also be used to take one coordinate every 5 nautical miles in the target sea area as a node.

[0022] In practical implementation, deploying two acoustic sensors at a preset vertical interval at each node to form a dual-acoustic sensor vertical receiving array can be achieved as follows: Based on the typical sound velocity profile and wavelength of the target sea area, calculate and set a vertical spacing capable of effectively distinguishing different normal modes. During deployment, use conventional marine fixing devices such as anchors, mooring buoys, or bottom supports to precisely fix the two acoustic sensors at a preset depth, ensuring they are vertically collinear and maintain stable attitude. The acoustic sensors are connected to the data logger or surface communication buoy at this node via underwater cables, completing the physical deployment and electrical connection of the array. The calculation and setting of a vertical spacing capable of effectively distinguishing different normal modes based on the typical sound velocity profile and wavelength of the target sea area refers to using an existing normal mode model, such as range-dependent acoustics, based on the typical sound velocity profile of the target sea area. The Model (RAM) is used to solve for the vertical wavenumber of each normal mode within the system's operating frequency band. The core is to ensure that the vertical spacing between the two acoustic sensors can sensitively detect the phase difference between adjacent modes. One possible calculation principle is that the vertical spacing should be greater than half of the ambiguity interval corresponding to the difference in vertical wavenumber between adjacent modes at the lowest operating frequency, i.e., vertical spacing > π / difference in vertical wavenumber between adjacent modes, to avoid the inability to distinguish the minimum value of acoustic field interference caused by adjacent modes. At the same time, to avoid spatial aliasing, the vertical spacing usually needs to be less than half the wavelength of the sound wave in the vertical direction at the highest operating frequency. Finally, a specific value is arbitrarily selected from the above range (i.e., half the wavelength of the sound wave in the vertical direction at the highest operating frequency > vertical spacing > π / difference in vertical wavenumber between adjacent modes) as the preset vertical spacing.

[0023] It should be noted that the dual acoustic sensor vertical receiving array in this application is an acoustic signal receiving device composed of two receiving sensors arranged vertically in seawater.

[0024] Furthermore, it should be noted that the acoustic sensor in this application is a hydrophone.

[0025] In some embodiments, multiple cod individuals carrying acoustic tags are pre-deployed to the target sea area before the coded acoustic signals emitted by the cod population carrying acoustic tags are collected by a dual acoustic sensor vertical receiving array. As a preferred embodiment, the pre-deployment of multiple cod individuals carrying acoustic tags to the target sea area in this application can be achieved in the following manner: First, from the target cod population, cod individuals without scratches on their body surface, without parasite infection, and without deformities are selected as the tagging objects. Then, the acoustic tag is aseptically implanted into the body cavity and sutured using a surgical procedure known in the industry. After the surgery is completed and the cod individuals have fully recovered, they are released into the sea in a habitat of a known target cod population in the target sea area. The above steps are repeated multiple times to complete the deployment of multiple cod individuals carrying acoustic tags. The encoded sequence of the sound waves emitted by the acoustic tag of each cod individual is different and uniquely corresponding.

[0026] In practice, the acquisition of coded acoustic signals emitted by cod populations carrying acoustic tags at the target node using a dual acoustic sensor vertical receiving array can be achieved in the following way: the two acoustic sensor channels in the dual acoustic sensor vertical receiving array work continuously in parallel, synchronously listening to and recording underwater sounds. When the signal processing unit detects in real time that an acoustic pulse conforming to a pre-set encoding rule appears in either channel, the system determines that a valid tag signal has appeared. Subsequently, the system synchronously triggers and records the complete time-domain waveforms of the two channels within the time period containing the signal, and assigns the same precise timestamp to these two data segments to form a set of dual-channel signal data. This set of dual-channel signal data is then used as the coded acoustic signal emitted by the cod population carrying acoustic tags.

[0027] It should be noted that the coded acoustic signal in this application refers to an acoustic signal actively emitted by an acoustic tag installed on the cod and possessing a uniquely identifiable characteristic.

[0028] In step 102, the autocorrelation function of each channel signal and the cross-correlation function between each channel signal in the encoded acoustic signal at the target node are determined, and multiple signal components that are superimposed due to the dispersion effect of seawater medium are separated based on the bending dispersion trajectory in the time-frequency domain of each autocorrelation function and cross-correlation function.

[0029] In some embodiments, determining the autocorrelation function of each channel signal and the cross-correlation function between each channel signal in the coded acoustic signal at the target node can be achieved by the following steps: acquiring a set of dual-channel signal data in the coded acoustic signal at the target node; preprocessing the signal data of each channel to obtain a stable data segment for each channel; determining the autocorrelation function of the stable data segment corresponding to each channel; and determining the cross-correlation function between the stable data segments corresponding to two different channels.

[0030] In specific implementation, the signal data of each channel is preprocessed separately to obtain a stable data segment for each channel. This can be achieved by preprocessing the signal data of each channel separately, including but not limited to filtering and gain compensation. The preprocessing uses existing signal preprocessing algorithms in the field of signal processing. Subsequently, a signal segment is extracted using an energy threshold detection method. Specifically, the short-time average energy of the preprocessed signal within a sliding time window is calculated, and a detection threshold determined by the background noise level is set. For example, it can be set as the sum of the average energy of the background noise and three times the standard deviation of the background noise. When the short-time average energy first... When the detection threshold is exceeded, it is determined to be the starting point of the coded acoustic pulse. After the starting point, the short-time average energy is continuously monitored. When it falls back and is lower than the detection threshold, it is determined to be the ending point of the same coded acoustic pulse. Finally, the complete signal segment between each starting point and ending point is extracted as an effective coded pulse segment, and the coded pulse segment extracted from each channel is used as the stable data segment of each channel. The average energy and standard deviation of the background noise can be obtained by conducting experiments in the target sea area in advance, collecting acoustic signals before releasing cod individuals carrying acoustic tags, and using the average energy and standard deviation of the acoustic signals as the average energy and standard deviation of the background noise.

[0031] It should be noted that, in this application, the stable data segment refers to the preprocessed signal data in the selected channel.

[0032] In some embodiments, referring to FIG2, which is an exemplary flowchart of separating signal components according to some embodiments of the present application, the separation of multiple signal components aliased due to the dispersion effect of seawater medium based on the curved dispersion trajectory in the time-frequency domain of each autocorrelation function and cross-correlation function can be achieved by the following steps: In step 1021, a reference distance for sound wave propagation in the target sea area is preset; In step 1022, the curved dispersion trajectory in each autocorrelation function and cross-correlation function is converted into a straight line parallel to the frequency axis based on the preset reference distance, thereby obtaining the frequency domain signal of each autocorrelation function and cross-correlation function; In step 1023, characteristic frequency peaks are extracted from each frequency domain signal, thereby obtaining multiple signal components aliased due to the dispersion effect of seawater medium.

[0033] It should be noted that the reference distance in this application is a preset value that is randomly selected within the effective propagation distance range of the sound wave in the environment. Specifically, the effective propagation distance range of the sound wave in the environment can be estimated using a normal mode model, such as the RAM model, based on the typical water depth, sound velocity profile, and system operating frequency band of the target sea area. A suitable value, such as the median value of the effective propagation distance range, is then selected within this effective propagation distance range. The typical water depth, sound velocity profile, and system operating frequency band of the target sea area can all be pre-calibrated through experiments. In other embodiments, the reference distance can also be preset using other methods, which are not limited here.

[0034] In practical implementation, based on a preset reference distance, the curved dispersion trajectories in each autocorrelation function and cross-correlation function are converted into straight lines parallel to the frequency axis, thus obtaining the frequency domain signals of each autocorrelation function and cross-correlation function. This can be achieved in the following way: First, divide the reference distance by the average sound speed of the target sea area, and use the quotient as the reference time. The average sound speed of the target sea area can be pre-calibrated experimentally, which will not be elaborated here. Then, construct a nonlinear time-domain resampling function based on this average sound speed. The mathematical expression of this nonlinear time-domain resampling function is: h(t) = sqrt(t^2 + T^2), where t is the time variable, T is the reference time, sqrt is the square root function, and h(t) is the reference time. To resample the time point t, the nonlinear time-domain resampling function is used to resample each autocorrelation function and cross-correlation function, thereby transforming the curved dispersion trajectory in each autocorrelation function and cross-correlation function into a straight line parallel to the frequency axis. Specifically, for each autocorrelation function and cross-correlation function, with time t as the variable, the time axis of each autocorrelation function and cross-correlation function is transformed by the nonlinear time-domain resampling function and multiplied by a compensation factor, which is the square root of the derivative of the nonlinear time-domain resampling function, i.e., sart(h(t)). Finally, the frequency domain signal obtained by Fourier transforming the resampled signal is used as the frequency domain signal of each autocorrelation function and cross-correlation function.

[0035] It should be noted that the frequency domain signal in this application is a frequency domain signal used to present the individual signal components successfully separated from the original aliased signal.

[0036] In specific implementation, extracting characteristic frequency peaks from each frequency domain signal to obtain multiple signal components aliased due to the dispersion effect of seawater can be achieved in the following way: extracting local maxima points from each frequency domain signal, and extracting signal segments to the left and right of the local maxima points that are greater than the amplitude threshold as signal components according to a preset amplitude threshold, thereby obtaining multiple signal components aliased due to the dispersion effect of seawater. The amplitude threshold is preset according to the background noise level. For example, an experiment is conducted in the target sea area in advance, and the acoustic signal is collected before releasing cod individuals carrying acoustic tags, and the average amplitude value of the acoustic signal is used as the amplitude threshold.

[0037] It should be noted that, in this application, the signal component is the acoustic component that characterizes the propagation of sound waves along a specified propagation path in an ocean waveguide.

[0038] Furthermore, it should be noted that the seawater medium dispersion effect in this application refers to the phenomenon that the propagation speed of sound waves in the ocean varies with frequency. This causes different frequency components of the broadband signal emitted by the sound source to arrive at the receiving point at different times after propagation, which manifests as a nonlinear curved trajectory in the time-frequency domain. Moreover, the trajectories of different normal modes (i.e., different propagation paths) overlap with each other. Step 102 introduces nonlinear time-domain resampling based on a preset reference distance to reverse the physical process, thereby offsetting the frequency-related group delay caused by dispersion. This straightens the curved trajectories corresponding to each normal mode into straight lines parallel to the frequency axis, transforming the complex time-frequency aliasing problem into an intuitive frequency domain peak detection problem. Moreover, it only requires limited spatial information provided by the hydrophone to achieve efficient separation of aliased signals, greatly reducing the difficulty of mode separation and identification.

[0039] In step 103, the modal order set of the dominant signal component is selected, and all signal components are converted into distance information from the sound source to the vertical receiving array of the dual acoustic sensors at the target node through the modal order set. The distance information from the sound source to the remaining nodes is then determined.

[0040] In some embodiments, the selection of the modal order set of the dominant signal component can be achieved by the following steps: extracting the main characteristic frequency peak with the largest amplitude from all autocorrelation functions and cross-correlation functions respectively; obtaining the theoretical characteristic frequency set; matching all main characteristic frequency peaks with the theoretical characteristic frequency set to obtain a candidate modal order set that meets the frequency characteristics; and selecting the candidate modal order set by the constraint relationship between the characteristic frequency peaks of all autocorrelation functions and cross-correlation functions to obtain the modal order set of the dominant signal component.

[0041] In practice, the main characteristic frequency peak with the largest amplitude can be extracted from all autocorrelation functions and cross-correlation functions in the following way: First, obtain the frequency domain signal of each autocorrelation function and cross-correlation function. Then, take the frequency peak with the largest amplitude in each frequency domain signal as the main characteristic frequency peak of each autocorrelation function and cross-correlation function.

[0042] It should be noted that, in this application, the main characteristic frequency peak refers to the characteristic frequency component used to describe the dominant acoustic field interference structure.

[0043] In practice, obtaining the theoretical characteristic frequency set can be achieved as follows: First, obtain the cutoff frequencies of all possible normal modes that can be excited within the target sea area, and calculate the theoretical characteristic frequency value between every two cutoff frequencies according to the formula μ=sqrt(∣vi^2-vj^2∣), where μ is the theoretical characteristic frequency value, vi is the cutoff frequency of the i-th mode, vj is the cutoff frequency of the j-th mode, and sqrt is the square root function. Finally, the set of all calculated theoretical characteristic frequencies is taken as the theoretical characteristic frequency set. This can be obtained directly by querying global or regional publicly available marine acoustics and geophysics databases, such as the National Environmental Information Center database of the National Oceanic and Atmospheric Administration (NOAA). In a preferred embodiment, the cutoff frequencies of the possible excited normal modes can also be determined by using environmental parameters such as the water depth of the target sea area, the acoustic characteristics of the seabed, and the pre-calibrated sound velocity profile, combined with the operating frequency band of the signal processing system, and by establishing and solving the eigenvalue equations of the normal modes under the waveguide environment using a sound field model based on normal mode theory. This allows the calculation of the cutoff frequencies of all possible excited normal modes under the environment. The cutoff frequency v of each mode is determined by the eigenvalue (i.e., vertical wavenumber) of that mode and can be obtained by standard numerical algorithms (such as the finite difference method or the transfer matrix method). In other embodiments, the cutoff frequencies of all possible excited normal modes in the target sea area can also be obtained by other methods, which are not limited here.

[0044] It should be noted that the theoretical characteristic frequency set in this application is a set of all possible acoustic field interference characteristic frequencies that are pre-calculated based on the acoustic environment characteristics of the target sea area.

[0045] In practice, matching all principal characteristic frequency peaks with the theoretical characteristic frequency set to obtain a candidate modal order set that meets the frequency characteristics can be achieved in the following way: First, set a tolerance threshold for frequency matching, for example, a relative error of no more than 2%. Then, compare the frequency value of each principal characteristic frequency peak with each theoretical value in the theoretical characteristic frequency set. If the relative error between the two meets the tolerance threshold for frequency matching, then the two-mode combination corresponding to the theoretical characteristic frequency is taken as a possible modal order combination of the principal characteristic frequency peak. Next, obtain the sets of possible modal order combinations corresponding to each of the four principal characteristic frequency peaks. Finally, combine the modal order combinations in these four sets to obtain all possible quadruplets. Each quadruplet consists of four possible modal order combinations, corresponding to the four principal characteristic frequency peaks. The set of all these quadruplets is the candidate modal order set.

[0046] It should be noted that the candidate modal order set in this application is a set of modal order combinations corresponding to the four main characteristic frequency peaks, which are initially screened by frequency matching. Each element is a quadruple, which corresponds to the possible modal order combinations of the four main characteristic frequency peaks.

[0047] In specific implementation, the candidate modal order set is screened based on the constraint relationship between the characteristic frequency peaks of all autocorrelation functions and cross-correlation functions, and the modal order set of the dominant signal component can be implemented in the following manner: for each quadruple in the candidate modal order set, screening is performed according to the constraint relationship between the modal orders of the autocorrelation function and the main characteristic frequency peak of the cross-correlation function. Using hydrophone 1 and hydrophone 2 to represent the signal channels at two receiving depths, the constraint relationship specifically includes: the possible modal order combinations corresponding to the main characteristic frequency peak of the cross-correlation function (denoted as the cross-correlation 1-2 main peak) calculated based on the signal of hydrophone 1 (as the reference signal) and the signal of hydrophone 2 (as the input signal) are denoted as (m, n), where m represents the possible modal order from the reference signal (hydrophone 1), and n represents the possible modal order from the input signal (hydrophone 2). For this main characteristic frequency peak, its modal order must satisfy m > n. The possible modal order combinations corresponding to the main characteristic frequency peak of the cross-correlation function (denoted as the cross-correlation 2-1 main peak) calculated based on the signal of hydrophone 2 (as the reference signal) and the signal of hydrophone 1 (as the input signal) are denoted as (q, p), where q represents the possible modal order from the reference signal (hydrophone 1), and p represents the possible modal order from the input signal (hydrophone 2). For this main characteristic frequency peak, its modal order must satisfy q < p. The two modal orders in the possible modal order combination corresponding to the main characteristic frequency peak of the autocorrelation function of hydrophone 1 must be equal to at least one of the modal orders in the possible modal order combinations corresponding to the cross-correlation 1-2 main peak and the cross-correlation 2-1 main peak respectively. The two modal orders in the possible modal order combination corresponding to the main characteristic frequency peak of the autocorrelation function of hydrophone 2 must also be equal to at least one of the modal orders in the possible modal order combinations corresponding to the cross-correlation 1-2 main peak and the cross-correlation 2-1 main peak respectively. The set composed of the four groups of modal order arrays that satisfy all the above constraint conditions is used as the modal order set of the dominant signal component. In particular, if the remaining quadruples are not unique, the energy mask of the acoustic field coherence component calculated in advance according to the acoustic field model and the receiving depth can be used for screening to eliminate those quadruples that contain modal combinations with extremely low energy at the current receiving depth, and finally a unique set of quadruples is obtained. In other embodiments, other methods can also be used for re-screening when the remaining quadruples are not unique, which is not limited here.

[0048] It should be noted that in this application, the modal order set is a parameter set used to describe different propagation paths of sound waves in an ocean waveguide.

[0049] In some embodiments, converting all signal components into distance information from the sound source to the target node's dual acoustic sensor vertical receiving array by means of the modal order set can be achieved by the following steps: for each group of modal orders in the modal order set, obtain the observed characteristic frequency value of the signal component corresponding to each group of modal orders; determine the theoretical characteristic frequency value corresponding to each group of modal orders; and determine the distance information from the sound source to the target node's dual acoustic sensor vertical receiving array based on the theoretical dispersion relation corresponding to each group of modal orders, the observed characteristic frequency value, the theoretical characteristic frequency value, and a preset reference distance.

[0050] In practice, the observed characteristic frequency value of the signal component corresponding to each modal order can be obtained in the following way: the frequency value corresponding to the peak value of the main characteristic frequency peak of the signal component corresponding to each modal order is taken as the observed characteristic frequency value of each modal order.

[0051] It should be noted that the observed characteristic frequency values ​​in this application are the characteristic frequency observation values ​​used to characterize the acoustic field interference structure generated by the selected propagation mode.

[0052] In practice, the theoretical characteristic frequency value corresponding to each group of modal orders can be determined in the following way: For each group of modal orders, the theoretical characteristic frequency value of each group of modal orders is calculated according to the formula F=sqrt(v1^2-v2^2), where F is the theoretical characteristic frequency value, v1 and v2 are a group of modal orders, and sqrt is the square root function.

[0053] It should be noted that the theoretical characteristic frequency value in this application is a theoretical value of the characteristic frequency used to characterize the acoustic field interference structure generated by the selected propagation mode.

[0054] In specific implementation, the distance information from the sound source to the target node's dual acoustic sensor vertical receiving array, based on the theoretical dispersion relationship corresponding to each modal order, the observed characteristic frequency value, the theoretical characteristic frequency value, and the preset reference distance, can be determined in the following way: For each modal order, the square of the ratio of the theoretical characteristic frequency value of each modal order to the observed characteristic frequency value corresponding to each modal order is taken as the theoretical dispersion coefficient of each modal order. This theoretical dispersion coefficient is the coefficient describing the theoretical dispersion relationship of each modal order. Subsequently, the theoretical dispersion coefficient of each modal order is multiplied by the preset reference distance, and the resulting product is taken as the distance estimate of each modal order. Finally, the average of all distance estimates is taken as the distance information from the sound source to the target node.

[0055] In step 104, the three-dimensional migration path of the Pacific cod population in the target sea area is reconstructed based on all distance information and the time difference of signal arrival at different nodes.

[0056] In some embodiments, referring to FIG3, which is an exemplary flowchart of determining a three-dimensional migration path according to some embodiments of the present application, the reconstruction of the three-dimensional migration path of the Pacific cod population in the target sea area based on all distance information and the time difference of signal arrival at different nodes can be achieved by the following steps: In step 1041, all distance information and the time difference of signal arrival at different nodes are obtained; in step 1042, all distance information and the time difference of signal arrival at different nodes are input into the positioning solution model to obtain the spatial position of individual cod in the target sea area; in step 1043, the spatial position of individual cod in the target sea area at different times is repeatedly monitored and all spatial positions are fitted into the movement trajectory of individual cod; in step 1044, the movement trajectories of other individual cod are further determined and all movement trajectories are reconstructed into the three-dimensional migration path of the cod population in the target sea area.

[0057] In practice, obtaining all distance information and the time difference of the signal arriving at different nodes can be achieved in the following way: First, obtain the timestamp of the acoustic signal of a specific coded sequence received at each node, where each specific coded sequence corresponds to a unique cod individual. Then, calculate the time difference of the acoustic signal of the same coded sequence received at different nodes.

[0058] In specific implementation, all distance information and the time difference of signal arrival at different nodes are input into the positioning calculation model to obtain the spatial position of the cod in the target sea area. This can be achieved by using a positioning algorithm based on time difference of arrival as the core calculation model. This model takes the precise time difference of signal arrival at different nodes and the known three-dimensional coordinates of each node as input to directly calculate the three-dimensional spatial position of the cod at a certain moment. For example, in this application, the four-station time difference of arrival (TDOA) model is selected. In other embodiments, other existing calculation methods can also be used to calculate the spatial coordinates of the cod. This is not limited here.

[0059] In specific implementation, the spatial positions of individual cod in the target sea area at different times are repeatedly monitored, and all spatial positions are fitted into the movement trajectory of the individual cod. This can be achieved by repeating steps 101 to 103 of this application over a period of time, continuously listening to, locating, and solving the coded signals emitted by the target cod, thereby obtaining a series of discrete three-dimensional spatial position points in time sequence. Subsequently, the obtained series of discrete three-dimensional spatial position points are fitted into the movement trajectory of the individual cod. Kalman filtering, a technique in the prior art, can be used for fitting. In other embodiments, other existing techniques can also be used to fit the movement trajectory of the individual cod, which is not limited here.

[0060] In specific implementation, the movement trajectories of other cod individuals are further determined, and all movement trajectories are reconstructed into a three-dimensional migration path of the cod population in the target sea area. This can be achieved in the following way: the movement trajectory of each cod individual is obtained, and on a unified three-dimensional geographic information system, the existing three-dimensional kernel density estimation (KDE) technology is used. For example, the three-dimensional kernel density estimation mentioned in the literature "Efficient 3D Movement-Based Kernel Density Estimator and Application to Wildlife Ecology" can be selected. The movement trajectories of all cod individuals are statistically analyzed to obtain high-density aggregation areas and high-density connecting zones. The high-density aggregation areas are regarded as the activity hotspots of the cod population in the target sea area, and the high-density connecting zones are regarded as the migration corridors of the cod population in the target sea area. Finally, a three-dimensional migration path map of cod individuals in the target sea area is drawn. In other embodiments, other existing technologies can also be used to draw the three-dimensional migration path map, which is not limited here.

[0061] In another aspect, in some embodiments, this application provides a Pacific cod natural population migration path monitoring system. Referring to Figure 4, which is a structural schematic diagram of a Pacific cod natural population migration path monitoring system according to some embodiments of this application, the Pacific cod natural population migration path monitoring system 400 includes: a data acquisition module 401, a processing module 402, and an execution module 403, which are described below: The data acquisition module 401 is mainly used to select a node in the target sea area as the target node, and to acquire the coded acoustic signals emitted by the cod population carrying acoustic tags at the target node through a dual acoustic sensor vertical receiving array; The processing module 402 is mainly used to determine the encoding at the target node. The autocorrelation function of each channel signal and the cross-correlation function between each channel signal are used to separate multiple signal components that are mixed due to the dispersion effect of seawater medium based on the curved dispersion trajectory in the time-frequency domain of each autocorrelation function and cross-correlation function. It should be noted that the processing module 402 in this application is also used to filter out the mode order set of the dominant signal component, and convert all signal components into distance information from the sound source to the vertical receiving array of the dual acoustic sensor at the target node through the mode order set, and continue to determine the distance information from the sound source to the vertical receiving array of the dual acoustic sensor at the remaining nodes; the execution module 403 in this application is mainly used to reconstruct the three-dimensional migration path of the Pacific cod population in the target sea area based on all the distance information and the time difference of the signal arriving at different nodes.

[0062] In addition, this application also provides a computer device, the computer device including a memory and a processor, the memory storing code, the processor being configured to acquire the code and execute the above-described method for monitoring the natural migration path of Pacific cod populations.

[0063] In some embodiments, referring to FIG5, this figure is a schematic diagram of the structure of a computer device for implementing a method for monitoring the natural migration routes of Pacific cod populations according to some embodiments of this application. The method for monitoring the natural migration routes of Pacific cod populations in the above embodiments can be implemented by the computer device 500 shown in FIG5, which includes at least one processor 501, a communication bus 502, a memory 503, and at least one communication interface 504.

[0064] Processor 501 can be a general-purpose central processing unit (CPU) or an application-specific integrated circuit (ASIC).

[0065] The communication bus 502 can be used to transmit information between the aforementioned components.

[0066] Memory 503 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CDROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital versatile optical discs, Blu-ray discs, etc.), magnetic disks or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. Memory 503 may exist independently and be connected to processor 501 via communication bus 502. Memory 503 may also be integrated with processor 501.

[0067] The memory 503 stores program code for executing the scheme of this application, and its execution is controlled by the processor 501. The processor 501 executes the program code stored in the memory 503. The program code may include one or more software modules. In the above embodiment, the method for monitoring the natural migration path of Pacific cod populations can be implemented by the processor 501 and one or more software modules in the program code in the memory 503.

[0068] Communication interface 504 uses any transceiver-like device to communicate with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0069] In a specific implementation, as one example, a computer device may include multiple processors, each of which may be a single-core (single CPU) processor or a multi-core (multi CPU) processor. Here, a processor may refer to one or more devices, circuits, and / or processing cores used to process data (e.g., computer program instructions).

[0070] The aforementioned computer device can be a general-purpose computer device or a special-purpose computer device. In specific implementations, the computer device can be a desktop computer, a portable computer, a network server, a handheld digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. This application does not limit the type of computer device.

[0071] In addition, this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for monitoring the natural migration routes of Pacific cod populations.

[0072] In summary, the Pacific cod natural population migration path monitoring method and system disclosed in this application firstly selects a node in the target sea area as the target node, and collects the coded acoustic signals emitted by the cod population carrying acoustic tags at the target node through a dual acoustic sensor vertical receiving array; determines the autocorrelation function of each channel signal and the cross-correlation function between each channel signal in the coded acoustic signal at the target node, and separates multiple signal components that are mixed due to the dispersion effect of the seawater medium based on the curved dispersion trajectory in the time-frequency domain of each autocorrelation function and cross-correlation function; filters out the mode order set of the dominant signal component, and converts all signal components into distance information from the sound source to the dual acoustic sensor vertical receiving array at the target node through the mode order set, and continues to determine the distance information from the sound source to the dual acoustic sensor vertical receiving arrays at the remaining nodes; and reconstructs the three-dimensional migration path of the Pacific cod population in the target sea area based on all the distance information and the time difference of the signals arriving at different nodes.

[0073] Therefore, this application constructs an information sensing network with spatial distribution and vertical structure by deploying a dual-hydrophone vertical receiving array with multiple nodes in the target sea area. This enhances the spatial dimension and reliability of the sound field information at the data acquisition level, providing a physical basis for distinguishing aliased modes. Secondly, by calculating the autocorrelation function of each channel signal at the target node and the cross-correlation function between channels, and separating multiple signal components aliased due to seawater dispersion effects based on their curved dispersion trajectories in the time-frequency domain, this scheme achieves decoupling and feature extraction of complex composite signals, overcoming the shortcomings of traditional methods in insufficient feature information extraction under aliasing conditions. Furthermore, this application selects the dominant signal component mode based on the constraint relationship of characteristic frequency peaks between all autocorrelation functions and cross-correlation functions. This method utilizes the inherent correlation of modal orders in the vertical space and cross-correlation domain of the sound field to construct a discrimination mechanism independent of prior assumptions about the sound source or environment. This enables robust and unique identification of dominant modal order combinations without requiring strong external assumptions, fundamentally eliminating the multi-valued ranging problem caused by modal ambiguity. Based on this, signal components are converted into distance information through theoretical dispersion relations, and combined with the orientation and time difference of arrival between multiple nodes, a high-confidence three-dimensional migration path is finally reconstructed. This forms a complete technical chain from array deployment, signal demixing, modal identification to positioning reconstruction, systematically improving the ability and reliability of continuous, accurate, and non-invasive three-dimensional monitoring of fish migration paths in complex shallow sea environments. In summary, the scheme proposed in this application can identify the order of the dominant cod individual acoustic signal propagation mode from aliased acoustic signals in the sea area.

[0074] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0075] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for monitoring the natural migration routes of Pacific cod populations, comprising pre-deploying a vertical receiving array of dual acoustic sensors with multiple nodes in the target sea area, characterized in that, The method includes: selecting a node in the target sea area as the target node; acquiring coded acoustic signals emitted by a cod population carrying acoustic tags at the target node using a dual acoustic sensor vertical receiving array; determining the autocorrelation function and cross-correlation function of each channel signal in the coded acoustic signal at the target node; separating multiple signal components aliased due to the dispersion effect of seawater medium based on the curved dispersion trajectory in the time-frequency domain of each autocorrelation function and cross-correlation function; selecting the mode order set of the dominant signal component; converting all signal components into distance information from the sound source to the dual acoustic sensor vertical receiving array at the target node using the mode order set; further determining the distance information from the sound source to the dual acoustic sensor vertical receiving arrays at the remaining nodes; and reconstructing the three-dimensional migration path of the Pacific cod population in the target sea area based on all distance information and the time difference of signal arrival at different nodes.

2. The method as described in claim 1, characterized in that, Before acquiring the coded acoustic signals emitted by a cod population carrying acoustic tags through a dual hydrophone vertical receiving array, the process also includes: pre-deploying multiple individual cod carrying acoustic tags into the target sea area.

3. The method as described in claim 1, characterized in that, Determining the autocorrelation function of each channel signal and the cross-correlation function between each channel signal in the coded acoustic signal at the target node specifically includes: acquiring a set of dual-channel signal data in the coded acoustic signal at the target node; preprocessing the signal data of each channel to obtain a stable data segment for each channel; determining the autocorrelation function of the stable data segment corresponding to each channel; and determining the cross-correlation function between the stable data segments corresponding to two different channels.

4. The method as described in claim 1, characterized in that, Separating multiple signal components aliased due to the dispersion effect of seawater medium from the curved dispersion trajectory in the time-frequency domain of each autocorrelation function and cross-correlation function specifically includes: a reference distance for sound wave propagation within a preset target sea area; converting the curved dispersion trajectory in each autocorrelation function and cross-correlation function into a straight line parallel to the frequency axis based on the preset reference distance, thereby obtaining the frequency domain signal of each autocorrelation function and cross-correlation function; extracting characteristic frequency peaks from each frequency domain signal, thereby obtaining multiple signal components aliased due to the dispersion effect of seawater medium.

5. The method as described in claim 1, characterized in that, The specific steps for selecting the modal order set of the dominant signal component include: extracting the main characteristic frequency peak with the largest amplitude from all autocorrelation functions and cross-correlation functions; obtaining the theoretical characteristic frequency set; matching all main characteristic frequency peaks with the theoretical characteristic frequency set to obtain a candidate modal order set that meets the frequency characteristics; and filtering the candidate modal order set through the constraint relationship between the characteristic frequency peaks of all autocorrelation functions and cross-correlation functions to obtain the modal order set of the dominant signal component.

6. The method as described in claim 1, characterized in that, Converting all signal components into distance information from the sound source to the target node's dual acoustic sensor vertical receiving array by means of the modal order set specifically includes: for each group of modal orders in the modal order set, obtaining the observed characteristic frequency value of the signal component corresponding to each group of modal orders; determining the theoretical characteristic frequency value corresponding to each group of modal orders; and determining the distance information from the sound source to the target node's dual acoustic sensor vertical receiving array based on the theoretical dispersion relation corresponding to each group of modal orders, the observed characteristic frequency value, the theoretical characteristic frequency value, and a preset reference distance.

7. The method as described in claim 1, characterized in that, The reconstruction of the three-dimensional migration path of the Pacific cod population in the target sea area based on the time differences of all distance information and signals arriving at different nodes specifically includes: obtaining all distance information and signal time differences arriving at different nodes; inputting all distance information and signal time differences arriving at different nodes into the positioning solution model to obtain the spatial position of individual cod in the target sea area; repeatedly monitoring to obtain the spatial position of individual cod in the target sea area at different times, and fitting all spatial positions into the movement trajectory of individual cod; continuing to determine the movement trajectory of other individual cod, and reconstructing all movement trajectories into the three-dimensional migration path of the cod population in the target sea area.

8. A monitoring system for the natural migration routes of Pacific cod populations, comprising a vertical receiving array of dual acoustic sensors with multiple nodes pre-deployed in the target sea area, characterized in that... The system includes: a data acquisition module, used to select a node in the target sea area as the target node, and acquire the coded acoustic signals emitted by the cod population carrying acoustic tags at the target node through a dual acoustic sensor vertical receiving array; a processing module, used to determine the autocorrelation function and cross-correlation function of each channel signal in the coded acoustic signal at the target node, and separate multiple signal components aliased due to the dispersion effect of the seawater medium based on the curved dispersion trajectory in the time-frequency domain of each autocorrelation function and cross-correlation function; the processing module is also used to filter out the mode order set of the dominant signal component, and convert all signal components into distance information from the sound source to the dual acoustic sensor vertical receiving array at the target node through the mode order set, and continue to determine the distance information from the sound source to the dual acoustic sensor vertical receiving arrays at the remaining nodes; and an execution module, used to reconstruct the three-dimensional migration path of the Pacific cod population in the target sea area based on all the distance information and the time difference of the signals arriving at different nodes.

9. A computer device, characterized in that, The computer device includes a memory and a processor, the memory storing code, and the processor being configured to acquire the code and execute the method for monitoring the natural migration routes of Pacific cod populations as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the method for monitoring the natural migration routes of Pacific cod populations as described in any one of claims 1 to 7.

Images