An electromagnetic pulse data transmission method

Abstract

The present application relates to the field of electromagnetic pulse technology, and discloses an electromagnetic pulse data transmission method. The method includes: collecting natural electromagnetic pulse signal data of a detection target area through a six-dimensional complementary electrode array structure, and performing Kalman filter peak extraction and characteristic frequency locking to obtain frequency-optimized signal data; performing first-order and second-order gradient difference matrix calculations and spatial vector synthesis to obtain signal source position coordinate data and signal intensity data; performing spectral domain reconstruction and multi-layer dynamic threshold analysis to obtain spatially clustered abnormal data packets; and selecting an optimal transmission channel to transmit the spatially clustered abnormal data packets to a cloud server. The present application effectively eliminates the influence of external interference sources, realizes the accurate extraction and characteristic frequency locking of signals in the effective frequency band, effectively suppresses power frequency and its harmonic interference, improves the signal-to-noise ratio of the effective frequency band of the signal, and ensures the reliability and timeliness of data transmission in a harsh field environment.

Description

Technical Field

[0001] This application relates to the field of electromagnetic pulse technology, and in particular to an electromagnetic pulse data transmission method. Background Art

[0002] As a non-invasive geological exploration method, the natural electromagnetic pulse signal technology has broad application prospects in the fields of geological disaster warning, mine safety monitoring, and underground structure analysis. Traditional electromagnetic pulse signal monitoring systems mainly rely on a single receiver or a simple array structure to collect data, with limited signal acquisition accuracy and insufficient anti-interference ability, making it difficult to meet the requirements of high-precision detection in complex geological environments. Especially during the monitoring process in areas with concentrated ground stress, due to the complex electromagnetic characteristics of geological bodies and multi-source electromagnetic interference, the signal-to-noise ratio of the collected signals is low and the characteristics are fuzzy, seriously affecting the accuracy of subsequent analysis and processing.

[0003] The electromagnetic pulse data transmission systems in the prior art generally have problems such as low data processing efficiency, large occupation of transmission bandwidth, and poor real-time performance. Traditional methods usually adopt filtering algorithms with fixed parameters and simple spectrum analysis techniques, which cannot effectively cope with the non-stationary and time-varying characteristics of geological electromagnetic signals. At the same time, due to the lack of effective spatial positioning means, most systems are difficult to accurately lock the position and intensity distribution of the signal source, resulting in insufficient accuracy in identifying abnormal areas. In addition, the communication conditions in the field exploration environment limit the timely and efficient transmission of a large amount of raw data to the processing center, restricting the real-time response ability of the geological disaster warning system. Summary of the Invention

[0004] This application provides an electromagnetic pulse data transmission method, which effectively eliminates the influence of external interference sources, realizes the precise extraction of signals in the effective frequency band and the locking of characteristic frequencies, effectively suppresses the power frequency and its harmonic interference, improves the signal-to-noise ratio of the effective frequency band of the signal, and ensures the reliability and timeliness of data transmission in harsh field environments.

[0005] In a first aspect, this application provides an electromagnetic pulse data transmission method, and the electromagnetic pulse data transmission method includes:

[0006] Collect natural electromagnetic pulse signal data of the detection target area through a six-dimensional complementary electrode array structure, and perform Kalman filter peak extraction and characteristic frequency locking to obtain frequency-optimized signal data;

[0007] Perform first-order and second-order gradient difference matrix calculations and spatial vector synthesis on the frequency-optimized signal data to obtain signal source position coordinate data and signal intensity data;

[0008] Perform spectral domain reconstruction and multi-layer dynamic threshold analysis on the signal source position coordinate data and the signal intensity data to obtain spatial clustering abnormal data packets;

[0009] Perform block-level differential compression on the spatially clustered abnormal data packets to obtain compressed data, and select the optimal transmission channel through the heterogeneous network self-organizing handover technology to transmit the compressed data to the cloud server.

[0010] In the technical solution provided by this application, the six-dimensional complementary electrode array structure adopts an equilateral triangle grid layout, achieving high-sensitivity acquisition of omnidirectional electromagnetic pulse components. Combined with high-precision GPS clock synchronization and reference station settings, the influence of external interference sources is effectively eliminated; the combination of spectrum adaptive tracking and Kalman filtering technology realizes the accurate extraction of signals in the effective frequency band and the locking of characteristic frequencies, effectively suppressing power frequency and its harmonic interference, and improving the signal-to-noise ratio of the effective frequency band of the signal; the combination of first-order and second-order gradient difference matrix calculations and spatial vector synthesis technology improves the signal source positioning accuracy, and realizes the accurate inversion of the signal source through principal component analysis and multilateration algorithms, making the stress field distribution characteristics clearer and more distinguishable; the signal enhancement method combining wavelet packet transform and singular value decomposition realizes the high-resolution reconstruction of the time-frequency characteristics of the signal, effectively removing random noise components while retaining key information, and enhancing the weak signal detection ability; the abnormal recognition scheme combining multi-layer dynamic threshold analysis and density clustering algorithm, by considering multiple threshold conditions of global statistics, regional relativity, and temporal changes, improves the accuracy of abnormal area recognition and reduces the false alarm rate; the block-level differential compression and heterogeneous network self-organizing transmission strategy achieve a high compression rate while keeping the key feature information intact, and combined with forward error correction and network automatic switching technology, ensure the reliability and timeliness of data transmission in harsh field environments. Brief Description of the Drawings

[0011] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0012] Figure 1 It is a schematic diagram of an embodiment of the electromagnetic pulse data transmission method in the embodiments of this application. Detailed Embodiments

[0013] The embodiment of the present application provides an electromagnetic pulse data transmission method. The terms "first", "second", "third", "fourth", etc. (if any) in the specification, claims and the above-mentioned drawings of the present application are used to distinguish similar objects, and do not have to be used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments described here can be implemented in an order other than that illustrated or described here. In addition, the term "comprising" or "having" and any variation thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0014] For ease of understanding, the specific process of the embodiment of the present application will be described below. Please refer to Figure 1 , an embodiment of the electromagnetic pulse data transmission method in the embodiment of the present application includes:

[0015] Step S101, collect the natural electromagnetic pulse signal data of the detection target area through a six-dimensional complementary electrode array structure, and perform Kalman filter peak extraction and characteristic frequency locking to obtain frequency-optimized signal data;

[0016] It can be understood that the execution subject of the present application can be an electromagnetic pulse data transmission system, or a terminal or a server, and no specific limitation is made here. The embodiment of the present application takes the server as the execution subject as an example for illustration.

[0017] Specifically, according to the terrain structure and geological target distribution of the detection area, an equilateral triangle arrangement pattern is adopted to divide the detection area into a grid, and based on this, the deployment scheme of the six-dimensional complementary electrode array is determined. At each set electrode receiving point, a six-dimensional complementary electrode array unit with a consistent structure is installed. This array structure consists of three pairs of mutually orthogonal vector electromagnetic induction coils, which are respectively used to collect the electric and magnetic field vector information in the X, Y, and Z axes, so as to achieve full-space, high-resolution, and multi-dimensional reception coverage of natural electromagnetic pulse signals. The induction coil structure is composed of high-sensitivity magnetoelectric materials and is equipped with silver-silver chloride composite electrodes to synchronously collect the electric field changes. At the same time, multiple electrode receiving devices arranged at equal intervals are embedded in each receiving unit to construct a gradient difference measurement structure to enhance the subsequent spatial positioning accuracy. In order to ensure the synchronization and coordination of all six-dimensional complementary electrode arrays in the large-area layout, a high-precision GPS clock synchronization module is integrated in each array unit. This module achieves a time alignment accuracy better than 10 nanoseconds, so that the subsequent phase difference calculation and gradient difference fitting are not interfered by time errors; at the same time, each electrode array node is equipped with a low-power wireless transmission module. This module supports the construction of a self-organizing mesh network architecture, realizes adaptive routing, data forwarding, and status coordination between nodes, constructs an edge-level data acquisition network through a distributed real-time communication system, and enables the original data of the entire detection area to be converged to the main control processing platform without delay. In order to enhance the system's ability to identify and suppress external environmental interference signals, E typical locations are selected on the periphery of the entire target area to deploy reference stations. Each reference station is equipped with a six-dimensional electrode unit equivalent to the detection node, which is used to record the background electromagnetic interference input from outside the area in real time, including broadband noise information generated by industrial sources, communication systems, meteorological changes, etc. The obtained interference data is used as the noise calibration benchmark in the subsequent electromagnetic signal analysis. On the above basis, the entire detection target area forms an original electromagnetic data acquisition system, which realizes real-time communication through a mesh network. Each node uploads the locally collected electromagnetic vector signals (including instantaneous electric field, current change rate, and spatial gradient) to the central processing unit to form a preliminary set of original electromagnetic pulse vector data. The background interference data collected by the reference station is introduced, and a correlation analysis is carried out channel by channel and frequency band by band with the original data of the target area nodes. Through the joint comparison in the frequency domain and time domain, the background interference components with high correlation are identified and removed, and the purified natural electromagnetic pulse signal data is calibrated.After calibration, the processed electromagnetic vector signal data is input into the spectrum adaptive tracking unit for fast Fourier transform with a fixed-length window (such as 2048 points) to obtain a complete spectrogram of the electromagnetic signal in the range of 0.1 Hz to 10 kHz. And the peak extraction algorithm based on Kalman filtering is applied in the spectrogram for automatic identification of the main frequency components. This algorithm constructs a dynamic state estimation model to continuously track the energy change trajectory of each frequency component and extracts the stable main frequency characteristics based on the sliding superposition of multiple windows. After identifying the stable main frequency, this frequency is matched with the preset geological feature frequency template, and a matching threshold (such as 85%) is set to ensure that the identified frequency has a clear geological directivity. When the matching is successful, the system automatically locks this frequency range and dynamically adjusts the subsequent data sampling frequency to eight times this characteristic frequency, thus achieving frequency optimization on the basis of satisfying the Nyquist sampling theorem. Output a set of signal data with optimized frequencies.

[0018] Preprocess the natural electromagnetic pulse signal data to enhance the recognizability of its spectral features and suppress noise interference. Since power frequency interference (such as 50Hz or 60Hz) is prevalent in the natural electromagnetic environment and has strong periodicity, the power frequency notch filtering technique is applied to the input signal data. A digital notch filter with adjustable center frequency is used to effectively remove the main power system interference frequency and its harmonic components from the signal. On this basis, to eliminate high-frequency noise and non-stationary background disturbances, the wavelet threshold denoising algorithm is used to perform multi-scale decomposition on the signal after notch processing. By selecting appropriate wavelet basis functions (such as Daubechies or Symlet), the signal is decomposed into multiple frequency bands and a soft threshold function is set to denoise the coefficients of each frequency band in an adaptive manner, and a group of denoised electromagnetic pulse signals with higher signal-to-noise ratio is reconstructed. The denoised signal is input into the spectral adaptive tracking unit and converted to the frequency domain. The discrete Fourier transform is used to expand the spectrum of the signal within the entire time window, calculate its power spectral density, and obtain the energy distribution of the signal at each frequency point. Based on the principle of local maximum, analyze the power spectrum curve, mark the frequency points with significant energy peaks in the spectrogram, and record their positions, amplitudes, and spectral widths to form the initial spectral feature data for subsequent modeling processing. To enhance the stability analysis ability of frequency components, a state space model is constructed with the extracted initial spectral features as the input. Its core includes a state transition matrix and an observation matrix, which are used to describe the evolution law of frequency components in the time series and the influence of observation noise. The model is embedded in the Kalman filtering algorithm to perform real-time state estimation and prediction, eliminating background disturbances and system errors, and obtaining a group of continuous and smooth filtered spectral trajectory data. Based on the filtered spectral trajectory data, calculate the drift rate of the spectral peak within three adjacent time windows, defined as the ratio of the difference in peak frequencies between adjacent windows to the time difference, and compare it with the preset stability target value. When the drift rate of a certain frequency point is lower than the threshold and stable features continuously appear, it is considered to have long-term trackability and physical representativeness. The system then marks this frequency point as a "locked frequency" and records its characteristic parameters in the frequency lock parameter set. This parameter set contains not only the frequency value but also its energy, stability coefficient, and bandwidth characteristics, which are used to guide subsequent analysis and processing. Based on the above frequency lock parameter set, perform the Hilbert transform processing with an adaptive window on the denoised signal. By constructing an analytic signal and calculating its instantaneous amplitude and instantaneous frequency, perform refined time-frequency joint analysis. Through the sliding window mechanism, perform local analysis on different time periods, so that the signal is analyzed in both the time and frequency dimensions, forming a group of enhanced time-frequency feature matrices containing energy, frequency, and time information, showing the evolution process of frequency components, and distinguishing the main part of the signal from the noise interference part.Implement a frequency-selective attenuation mechanism based on the enhanced time-frequency feature matrix, that is, uniformly implement 50% energy attenuation processing on non-feature frequency components, and at the same time completely retain the signal components marked as locked frequencies. Perform inverse processing on the adjusted time-frequency feature matrix, including inverse Hilbert transform and time-domain signal reconstruction, and output a set of electromagnetic pulse signal data with optimized frequencies.

[0019] Step S102: Perform first-order and second-order gradient difference matrix calculations and spatial vector synthesis on the signal data with optimized frequencies to obtain signal source position coordinate data and signal intensity data;

[0020] Specifically, based on the signal data after frequency optimization, a multi-point measurement framework is established in three-dimensional space. Based on the distribution characteristics of the six-dimensional complementary electrode array in the three axial directions of X, Y, and Z, according to the geometric structure between adjacent measurement points in each axial direction, the intensity difference of the signal at these measurement points is calculated. This intensity difference reflects the gradient change generated during the propagation of electromagnetic waves in different directions. A three-dimensional first-order gradient difference matrix composed of differences between multiple measurement points is constructed in each direction. Each element in the first-order gradient difference matrix is divided by the actual distance between the corresponding measurement points, that is, spatial normalization is performed to obtain the electromagnetic signal intensity change rate per unit distance, forming the standardized first-order gradient field distribution data. Based on the first-order gradient field distribution data, its spatial change rate is calculated, that is, the gradient of the gradient is calculated in each direction to form the second-order metric. The central difference method is used to perform differential operations on the first-order gradient field along each spatial dimension to construct a three-dimensional second-order gradient difference matrix, which is used to represent the degree of curvature change of the electromagnetic signal in space. The Gaussian smoothing filter algorithm is applied to the second-order gradient difference matrix, and spatial convolution operations are used to perform weighted averaging on adjacent points, thereby effectively suppressing local spikes and edge distortions, and obtaining a more continuous and physically reasonable second-order gradient field distribution data. The first-order gradient field distribution data and the second-order gradient field distribution data are combined into gradient field synthesis data, which reflects the entire process of the evolution of the electromagnetic signal from first-order linear change to second-order curvature in three-dimensional space. In order to extract the spatially varying features with physical dominant significance from the synthesized gradient field, principal component analysis is performed on this data. By calculating the covariance matrix of the data and solving its eigenvectors and eigenvalues, dimensionality reduction is achieved, and the spatial vector field characteristic parameters containing the main energy information are extracted. Based on the spatial vector field characteristic parameters, an inversion mathematical model of the electromagnetic signal source is constructed. According to the basic relationship of the spatial propagation of the electromagnetic field in Maxwell's equations, a mapping function between the signal intensity gradient and the signal source position is established, and the principal components obtained by principal component analysis are introduced as constraint terms into the inversion equation to form a constrained optimization problem. The model is solved by the least squares fitting algorithm to deduce the position coordinate data (X, Y, Z) of the signal source in three-dimensional space, and the spatial distance of each point in the entire measurement area relative to the signal source is calculated with this as the center. Using the attenuation model of electromagnetic wave propagation in different media, the signal intensity is correlated with parameters such as distance, dielectric constant, conductivity, and permeability, and the change function of the signal intensity in three-dimensional space is deduced to generate the signal intensity distribution map in the entire space range.

[0021] Step S103: Perform spectral domain reconstruction and multi-layer dynamic threshold analysis on the signal source position coordinate data and the signal intensity data to obtain spatially clustered abnormal data packets;

[0022] Specifically, the spatial positioning results of the signal source are fused with the corresponding electromagnetic pulse signal intensity data in terms of spatial correspondence. Taking each positioning moment as a reference, the corresponding three-dimensional coordinate data (X, Y, Z) are mapped one by one with the signal intensity value at that moment to construct a sequence of spatio-temporal domain signal blocks with clear geographical and spatial significance. The wavelet packet transform is performed on the sequence of spatio-temporal domain signal blocks to extract the local features of the signal in different frequency bands. The wavelet packet transform further decomposes both the high-frequency and low-frequency sub-bands simultaneously, enabling the entire signal to be unfolded in a fine-grained frequency domain structure and constructing a multi-level frequency band tree structure. After completing the multi-level decomposition, the calculation results of the coefficient energy of each sub-frequency band are summarized to generate a frequency band weight coefficient table, which reflects the energy distribution of the signal in different frequency bands. Based on the frequency band weight coefficient table, soft threshold shrinking processing is performed on all wavelet packet coefficients, that is, an adaptive soft threshold function is used to correct each coefficient, weakening the low-amplitude components while retaining the main information, to obtain a set of optimized wavelet packet coefficients. After reorganizing the optimized wavelet packet coefficients, a two-dimensional signal matrix is constructed. Each row or column of this matrix represents the spectral structure in different time or spatial dimensions, forming a high-dimensional data set reflecting spatio-temporal evolution characteristics. To extract the most representative principal component structure in this matrix, the singular value decomposition algorithm is introduced. The signal matrix is decomposed into three parts: a left singular matrix, a singular value diagonal matrix, and a right singular matrix. Each singular value in the singular value diagonal matrix represents the contribution degree of the corresponding principal component in the overall signal. These singular values are sorted and a retention threshold is set. For example, the first 5 to 10 principal singular value components that account for more than 90% of the total energy are retained, and the remaining small singular values are set to zero, thereby suppressing secondary noise information and only retaining the most significant principal component structure in the signal. By reconstructing the singular value diagonal matrix and combining the left singular matrix and the right singular matrix, the denoising and reconstruction process of the signal is completed, and a set of signal reconstruction data with dominant characteristics after denoising is obtained. On this basis, a frequency band decomposition operation is performed on the denoised signal data. The signal is remapped to each frequency sub-band according to the original decomposition structure in the wavelet packet tree and weighted and fused according to the energy distribution in the sub-band to form an enhanced time-frequency distribution signal with high signal-to-noise ratio and time-frequency resolution. A multi-layer dynamic threshold analysis mechanism is introduced to perform hierarchical screening and classification on the enhanced time-frequency distribution signal. Key statistical indicators such as the mean, standard deviation, kurtosis, and skewness of the enhanced signal at each spatio-temporal point are statistically analyzed, and a dynamic threshold system is set accordingly. The first-layer threshold corresponds to the anomaly detection of slightly deviating from the baseline, the second-layer threshold is used to identify significant anomaly signals, and the third-layer threshold focuses on capturing strongly abnormal signals with extremely strong energy mutations. Each layer of threshold is dynamically updated according to local statistical features and the global baseline, with environmental adaptability, and can automatically adapt to changes in background conditions such as seasonal changes, geological disturbances, or human interferences.After the multi-layer threshold analysis is completed, the signal points that meet the abnormal conditions are subjected to spatial clustering processing. The density clustering algorithm (such as DBSCAN) is used to divide the adjacent abnormal points into spatial aggregates, that is, spatial abnormal clusters. Each abnormal cluster contains key attributes such as its central position, average abnormal level, spatial expansion range, and duration. The description information of each abnormal cluster is encapsulated in a standardized format to construct a spatial clustering abnormal data packet, which contains spatial coordinates, time-frequency characteristics, abnormal levels, energy trends, initial appearance time, and duration, etc.

[0023] The enhanced time-frequency distribution signal is structured to accurately express the variation law of the signal in the three dimensions of space, time, and frequency. The enhanced time-frequency distribution signal is projected onto a three-dimensional coordinate system, which constructs a three-dimensional information body with spatial position, time index, and frequency component as the three dimensions. By setting the spatial resolution, time sampling interval, and frequency bandwidth step, the entire three-dimensional continuous signal field is divided into multiple discrete units of finite size. Each discrete unit is defined as a voxel, which contains the electromagnetic signal characteristic values such as instantaneous energy, spectral density, frequency stability coefficient, or coherent amplitude within this spatial position, this time period, and this frequency range, so as to systematically obtain the signal characteristic voxel data matrix, which is organized in the form of a regular three-dimensional grid. Based on this signal characteristic voxel data matrix, multi-layer threshold discrimination conditions are constructed. To ensure the accuracy and robustness of abnormal detection, global statistical analysis is performed on the entire voxel matrix, and key parameters such as the average value, standard deviation, kurtosis, and skewness of each type of signal characteristic are extracted. Based on these statistical indicators, a dynamic threshold model with multiple levels is constructed. This model consists of three layers, which are used to identify minor abnormalities, significant abnormalities, and strong abnormalities respectively; its threshold expression is formalized as , , , where μ is the average value of this feature and σ is the standard deviation. The feature values of each voxel are input into this multi-level threshold discrimination model, and point-by-point evaluation is carried out against each level of threshold to determine whether it meets the abnormal determination condition at a certain level. Through this process, all voxel units that meet the abnormal conditions are screened out in the entire three-dimensional voxel space to form a preliminary abnormal voxel set, which exists in the form of discontinuous discrete voxel point groups and represents the spatial electromagnetic abnormal area. Spatial density clustering processing is performed on the preliminary abnormal voxel set, and a density-based clustering algorithm, such as the DBSCAN algorithm, is used to classify the voxel set. In this process, two core clustering parameters are set, namely the scanning radius ε and the minimum number of points MinPts, and then density reachability analysis is performed on all preliminary abnormal voxels based on the relative position relationship and abnormal intensity weight of the voxels in the three-dimensional space. Voxels that are close in distance and similar in abnormal degree are aggregated into the same spatial clustering unit to obtain a spatial abnormal clustering set. Each clustering unit represents an electromagnetic abnormal structure that is continuous, locally concentrated, and significantly deviated from the normal background in terms of geophysical meaning, and has a clear spatial distribution pattern and signal intensity characteristics. On this basis, to improve the geometric expression ability and engineering applicability of the clustering results, the geometric features of each clustering result are calculated, including the boundary voxel envelope, spatial centroid position, maximum circumscribed boundary size, number of voxels, voxel density distribution, etc., and the boundary data of the target clustering is constructed accordingly. The boundary geometric data of each spatial abnormal clustering is matched and integrated with the signal feature parameters (such as average energy, peak frequency, duration, change trend, etc.) corresponding to its internal voxels, and a complete spatial clustering abnormal data packet is generated through a structured packaging method. Each data packet contains information such as the spatial boundary description of the abnormal clustering, three-dimensional position center, voxel feature statistics, multi-level threshold level, formation time and duration, etc., while keeping the data structure compact for efficient transmission and distributed storage.

[0024] Step S104: Perform block-level differential compression on the spatial clustering abnormal data packet to obtain compressed data, and select the optimal transmission channel through the heterogeneous network self-organizing switching technology to transmit the compressed data to the cloud server.

[0025] Specifically, a structured block division operation is performed on the spatially clustered abnormal data packets. This data packet contains multi-dimensional information such as the spatial coordinates, signal characteristics, boundary descriptions, temporal evolution, and anomaly levels of the abnormal regions. Therefore, based on the data attributes and data usage frequency, the entire data packet is divided into several logically complete but mutually independent data blocks according to the attribute categories. To improve the compression efficiency and reduce redundancy, these blocks are further optimized and rearranged to make the data storage structure more adaptable to the compression model, and an optimized set of data blocks is constructed. A spatio-temporal differential coding operation is performed on the optimized blocks, that is, according to the change trend of information in the time dimension and the numerical gradient between spatially adjacent voxels, the differences between consecutive blocks are encoded, so as to only retain the necessary incremental information. This process can significantly reduce the amount of data and form the first data stream. The first data stream is input into the adaptive quantizer module to perform the operation of grading data importance. This module assigns different quantization precision levels to different segments according to the key degree of each data segment in the overall signal expression, combined with multi-dimensional parameters such as its anomaly level, spatial position centrality, and correlation with other regions. Data with higher importance will retain higher precision, while data with lower importance will use a coarser quantization level, thus further compressing the amount of data while ensuring the integrity of key information. This operation generates the second data stream and introduces an obvious data hierarchical structure. A standardized data message format is constructed based on the second data stream, and complete message header information is added on this basis. The message header contains multiple key fields, including a high-precision timestamp indicating the acquisition time, a geographical location identifier output by the positioning module, a data priority flag bit generated by the aforementioned quantizer, and a checksum field for subsequent data integrity verification. These fields ensure the traceability and distribution scheduling ability of the data during transmission, and also improve the robustness and scheduling flexibility of the system in the face of heterogeneous network environments. To improve the reliable reception rate of the data in a low-quality network environment, Reed-Solomon forward error correction coding is added to the entire data message, and the error resistance ability of the data is improved by adding redundant symbols, so that the receiving end can still accurately restore the original data in the face of partial data loss or bit errors. After the above operations are completed, a compressed data unit with self-description ability, integrity guarantee ability, and redundancy and fault tolerance ability is formed. Before entering the actual transmission, the system real-time and parallelly detects various heterogeneous wireless communication networks available in the current environment, including 4G / 5G cellular communication networks, LoRa long-distance Internet of Things communication links, and long-distance wireless communication systems based on satellite links. For each communication channel, real-time parameters in multiple dimensions such as its signal strength, transmission delay, current available bandwidth, and frequent interruption probability are monitored respectively, and these parameters are organized into a network performance vector with a unified standard. Based on these vectors, a network quality scoring matrix is constructed, and each element in this matrix represents the performance of a certain network on a certain transmission index, with time dynamics and spatial correlation.The scoring matrix is comprehensively calculated using the fuzzy logic inference algorithm, converting multi-dimensional indicators into fuzzy membership functions. Through weighted synthesis and rule base evaluation, the comprehensive scores of each available network are output, and the optimal transmission channel at the current moment is selected accordingly. The compressed data is packetized according to the optimal transmission channel. Based on the maximum transmission unit size of the channel, the compressed data is divided into several target data packets, and each data packet contains a complete header and segmentation identification information. These data packets are scheduled and sent through the heterogeneous network self-organizing handover technology, which has the ability to dynamically switch between multiple links, can automatically detect the channel state during transmission and automatically re-route according to changes in network conditions, ensuring the reliability and continuity of data during transmission. After the transmission is completed, the cloud server completes data recombination, differential reduction, error correction decoding, and integrity verification through a preset parsing protocol, and finally obtains spatially clustered abnormal data with complete structure and content.

[0026] In the embodiment of the present application, the six-dimensional complementary electrode array structure adopts an equilateral triangle grid layout, achieving high-sensitivity acquisition of omnidirectional electromagnetic pulse components. With high-precision GPS clock synchronization and reference site settings, the influence of external interference sources is effectively eliminated; the combination of spectrum adaptive tracking and Kalman filtering technology realizes the accurate extraction of signals in the effective frequency band and the locking of characteristic frequencies, effectively suppressing power frequency and its harmonic interference, and improving the signal-to-noise ratio of the effective frequency band of the signal; the combination of first-order and second-order gradient difference matrix calculations and spatial vector synthesis technology improves the signal source positioning accuracy, and through principal component analysis and multi-lateral positioning algorithms, the accurate inversion of the signal source is achieved, making the stress field distribution characteristics clearer and more distinguishable; the signal enhancement method combining wavelet packet transform and singular value decomposition realizes the high-resolution reconstruction of the time-frequency characteristics of the signal, effectively removing random noise components while retaining key information, and enhancing the weak signal detection ability; the anomaly recognition scheme combining multi-layer dynamic threshold analysis and density clustering algorithm, by considering multiple threshold conditions such as global statistics, regional relativity, and temporal changes, improves the accuracy of anomaly region recognition and reduces the false alarm rate; the block-level differential compression and heterogeneous network self-organizing transmission strategy achieve a high compression rate while keeping the key feature information complete, combined with forward error correction and network automatic switching technology, ensuring the reliability and timeliness of data transmission in harsh field environments.

[0027] In a specific embodiment, the process of executing step S101 may specifically include the following steps:

[0028] Determine the electrode array deployment plan for the detection target area according to the equilateral triangle arrangement pattern, and install a six-dimensional complementary electrode array including three pairs of orthogonal vector electromagnetic induction coils at each set receiving point;

[0029] Install a high-precision GPS clock module on all six-dimensional complementary electrode arrays, and configure a low-power wireless transmission module to build a mesh network architecture to obtain a real-time data transmission system;

[0030] Set up E reference stations in the detection target area, and record the external electromagnetic interference signals of the area through each reference station to obtain noise calibration reference data;

[0031] Monitor the detection target area through the real-time data transmission system to obtain the original electromagnetic pulse vector data;

[0032] Perform correlation calculation and calibration based on the original electromagnetic pulse vector data and the noise calibration reference data to obtain the natural electromagnetic pulse signal data;

[0033] Input the natural electromagnetic pulse signal data into the spectrum adaptive tracking unit for Kalman filter peak extraction and characteristic frequency locking to obtain the frequency-optimized signal data.

[0034] Specifically, geometric modeling is performed on the area to be measured, and the detection points are deployed and optimized according to the geological structure complexity, target body distribution, and spatial size, adopting an equilateral triangle arrangement pattern to minimize blind spots and provide uniform coverage on the two-dimensional plane, thereby enhancing the spatial resolution ability of the electromagnetic field gradient change. Each set receiving point serves as a collection node, and a set of standardized six-dimensional complementary electrode array devices is installed. This array device consists of three pairs of vector electromagnetic induction coils arranged orthogonally to each other, which synchronously sense the electromagnetic field components in the three orthogonal spatial axes of X, Y, and Z respectively. At the same time, a high-sensitivity silver-silver chloride electrode and magnetic induction coil combination component is equipped to ensure the full-band response ability of low-frequency and medium-frequency electromagnetic pulse signals, thereby realizing high-resolution three-dimensional spatial electromagnetic vector field acquisition. Time synchronization and data communication configurations are carried out for each collection node. To ensure the timing consistency between multiple array nodes and reduce the subsequent signal alignment error, all six-dimensional complementary electrode array units are integrated with a high-precision GPS timing module, which has nanosecond-level time synchronization ability, and realizes unified timing control for all-region nodes through the satellite clock source, eliminating time drift problems caused by terrain differences, distances, or device asynchronization. At the same time, to ensure data intercommunication between long distances and multiple nodes, a low-power wireless transmission module is configured in each node. These modules support automatic networking, self-repair, and node discovery functions, and link all nodes into a distributed wireless transmission network with a flexible topology and scalable structure through a self-organizing mesh network architecture. This mesh architecture can automatically adjust the routing path according to the network environment, and has anti-interference and self-healing capabilities. It can automatically find alternative paths to maintain the overall network connection when any node fails, thereby constructing a real-time data acquisition and transmission system with edge computing preposition function. After completing the node layout and network initialization in the target area, several reference sites are set up outside the detection area for real-time collection of external background electromagnetic environment information. The number of reference sites is set to E, distributed on the four boundaries of the area, and the same electrode array configuration as the detection nodes is adopted to ensure consistent data structure and signal characteristics, which are used to record non-geological interference sources such as industrial interference, power system noise, communication base station signals, and natural background fields. After being collected by the reference sites, these interference signals form an independent noise reference library. As all nodes are put into operation, the mesh network starts the data link and maintains a stable communication channel with the edge processing server or the central collection unit. Each six-dimensional electrode array unit real-time collects the natural electromagnetic pulse signals distributed in the three-dimensional space. The obtained data contains multi-dimensional features such as electric field strength, magnetic induction intensity, and their gradient changes in vector form, and is digitized with high precision through a 24-bit analog-to-digital converter. The sampling rate is set between 1 kHz and 100 kHz to cover natural signals in different frequency bands, and then sent to the main control unit after local processing to form an original electromagnetic pulse vector data set.Meanwhile, E reference stations synchronously upload the noise data they collect via the network to form a background noise database that is spatio-temporally aligned with the subjective survey area. In the signal processing flow, the original electromagnetic vector data is dynamically compared and correlated with the reference noise data. Based on the joint correlation metric in the time domain and frequency domain, the signal coherence calculation, energy similarity matching, and frequency relative offset analysis are performed on the electromagnetic pulse signals collected by each detection node and the data of the reference stations to determine which frequency components or time periods the signals mainly originate from external interference, and which components have natural geological response characteristics. After the correlation calculation and noise rejection, the signal part that removes the influence of external background disturbances is extracted, and the natural electromagnetic pulse signal data with clear geophysical significance is constructed, and its time series characteristics, frequency component distribution, and spatial corresponding coordinates are retained in vector format. The natural electromagnetic pulse signal data is input into the spectrum adaptive tracking unit. Perform a fixed-length fast Fourier transform operation on the input signal, extract its power spectral density map, and identify the local peak characteristics in the spectrum. In order to dynamically extract the stable main frequency information from the non-stationary signal, a frequency tracking algorithm based on Kalman filter is used to establish a frequency drift state space model, and the frequency distribution is dynamically estimated and predicted and corrected. The Kalman filter continuously adjusts the error between the predicted value and the measured value, so that the main frequency tracking process has time continuity and mutation suppression ability, so as to lock a group of frequency points with high stability on the time axis as the main frequencies related to the geological target. Re-adjust the sampling parameters according to the locked frequency to ensure that the subsequent acquisition meets the Nyquist condition, and generate the finally frequency-optimized electromagnetic pulse signal data.

[0035] In a specific embodiment, the process of performing the steps of inputting the natural electromagnetic pulse signal data into the spectrum adaptive tracking unit for Kalman filter peak extraction and characteristic frequency locking to obtain the frequency-optimized signal data may specifically include the following steps:

[0036] Perform power frequency notch filtering and wavelet threshold denoising processing on the natural electromagnetic pulse signal data to obtain the denoised electromagnetic pulse signal;

[0037] Convert the denoised electromagnetic pulse signal to the frequency domain through the spectrum adaptive tracking unit, calculate the power spectral density by discrete Fourier transform, and mark the characteristic frequency points according to the local maximum principle to obtain the initial spectrum characteristic data;

[0038] Construct a state space model based on the initial spectrum characteristic data, and perform state estimation and prediction through the state space model to obtain the filtered spectrum trajectory data;

[0039] Based on the filtered spectrum trajectory data, calculate the drift rate of the spectrum peaks within three adjacent time windows, and mark the frequency points with a stability coefficient greater than the preset target value as the locked frequencies to obtain the frequency locking parameter set;

[0040] Perform adaptive window Hilbert transform based on the frequency-locked parameter set to obtain the instantaneous frequency and instantaneous amplitude, and conduct time-frequency joint analysis through the instantaneous frequency and instantaneous amplitude to obtain an enhanced time-frequency feature matrix;

[0041] Attenuate the non-feature frequency components in the enhanced time-frequency feature matrix by 50%, retain the signal integrity of the locked frequency, and obtain frequency-optimized signal data through inverse transformation.

[0042] Specifically, effective noise reduction operations are performed on the original signal to improve the spectral clarity of the signal and the accuracy of subsequent processing. In the actual natural electromagnetic environment, the signal is affected by a large amount of background interference. In particular, low-frequency power system interference signals represented by the power frequency are the most common, usually existing in the form of 50 Hz or 60 Hz and their higher harmonics. In the preprocessing stage, power frequency notch filtering technology is adopted to construct a band-stop filter with a center frequency precisely matching the power frequency interference frequency, filtering out the main frequency interference and its adjacent frequency band signals, thereby significantly reducing the impact of strong interference components on the subsequent analysis process. To suppress high-frequency oscillations, random noise, and local mutation interference, a wavelet threshold denoising method is adopted based on notch filtering. This method decomposes the signal into sub-signals of different frequency levels through multi-scale wavelet decomposition of the signal, and then performs shrinkage processing on the wavelet coefficients according to a preset soft threshold function, setting low-amplitude coefficients to zero or compressing them, so as to retain the main components and eliminate noise disturbances. Finally, the signal is reconstructed to obtain a group of electromagnetic pulse signals with higher signal-to-noise ratio and structural stability after denoising. The denoised electromagnetic pulse signal is input into the spectrum adaptive tracking unit to enter the frequency domain analysis stage. The task of this stage is to identify and extract the main frequency components in the signal and their changing trends over time. Therefore, the discrete Fourier transform is used to convert the time-domain signal to the frequency domain, calculate its power spectral density, and draw a complete spectrum diagram with frequency as the horizontal axis and energy as the vertical axis. In the spectrum diagram, according to the local maximum principle, the frequency points with significant energy prominence are identified, and these points correspond to the main frequency components of the electromagnetic pulse signal or their higher harmonic characteristics. Through point-by-point scanning and neighborhood comparison of the spectrum diagram, a series of frequency peaks are marked, and their frequency values, peak amplitudes, peak widths and other characteristic information are extracted to form an initial spectrum characteristic data set. Based on the initial spectrum characteristic data, a state space model is constructed to realize the dynamic estimation and trend prediction of the frequency trajectory. The model consists of a state transition equation and an observation equation, where the state variable represents the temporal evolution behavior of the frequency peak, and the observation variable is the frequency value and amplitude calculated within the current window. The Kalman filtering method is used to fuse the observed value and the predicted value, and the error covariance matrix is introduced to dynamically adjust the estimation accuracy, forming a spectrum tracking mechanism with noise suppression ability and time continuity. Under the action of this model, the original spectrum peak is smoothed, and a group of filtered spectrum trajectory data is output. This data not only shows continuous changes on the time axis, but also reflects the dominant trend of the real signal components on the frequency axis, and eliminates short-term disturbances or false peaks. Frequency stability analysis is carried out based on the filtered spectrum trajectory data. The change rate of each frequency peak within three consecutive time windows is analyzed in a sliding window manner, that is, the calculation of the drift rate. By calculating the ratio of the maximum change amplitude of the same frequency component in three adjacent windows to the time span, the drift rate of each peak is obtained, and the definition of the stability coefficient is introduced, that is, the frequency stability is represented by 1 minus the normalized value of the drift rate.If the stability coefficients of a certain frequency point in multiple time windows are all higher than the target threshold preset by the system, it is determined that the frequency has strong physical continuity and geological response stability. Mark it as a locked frequency, and form a frequency locking parameter set with the frequency value, average amplitude, main peak width, time stability range, etc. Perform an adaptive window Hilbert transform based on the frequency locking parameter set to extract fine-grained instantaneous features of the signal. The Hilbert transform constructs an analytic signal and directly extracts the instantaneous amplitude and instantaneous frequency from it to describe the non-stationary change trend of the signal on the micro time scale. The adaptive window mechanism automatically adjusts the window length according to the waveform period characteristics of the frequency band where the locked frequency is located, so that short-period high-frequency signals obtain sufficient frequency resolution, while long-period low-frequency signals obtain stronger time resolution. Traverse the entire signal sequence through a sliding window, extract the corresponding instantaneous frequency and amplitude in each window, and construct a two-dimensional matrix covering the entire time domain and frequency band, that is, an enhanced time-frequency feature matrix. The matrix takes time as the horizontal axis and frequency as the vertical axis, and the matrix value represents the instantaneous energy or amplitude, depicting the dynamic process of the signal evolving with time in different frequency channels, and highlighting the energy aggregation characteristics near the locked frequency. In order to enhance frequency selectivity and suppress the interference components in non-feature frequency bands, perform a frequency channel suppression operation on the enhanced time-frequency feature matrix. Uniformly attenuate all frequency channels in the matrix that are not included in the frequency locking parameter set, with a specific attenuation ratio of 50%, that is, halve the amplitude of the channel, while retain all signal intensities of the frequencies within a certain bandwidth near the locked frequency. The selective suppression operation makes the locked frequency show a relative advantage in the matrix, significantly improving the recognition degree of the main frequency component of the signal, and at the same time reducing the interference of redundant noise on the spectrum structure. After completing the feature retention and non-feature suppression, perform an inverse Hilbert transform and a time-frequency inverse transform operation on the time-frequency matrix to restore the processed matrix to a time-domain waveform, forming the final frequency-optimized electromagnetic pulse signal data.

[0043] In a specific embodiment, the process of performing step S102 may specifically include the following steps:

[0044] Calculate the signal intensity difference between adjacent measurement points in the X, Y, and Z three spatial directions for the frequency-optimized signal data, construct a three-dimensional first-order gradient difference matrix, and normalize the three-dimensional first-order gradient difference matrix by dividing it by the actual distance between the measurement points to obtain the first-order gradient field distribution data;

[0045] Based on the first-order gradient field distribution data, calculate the change rate of the gradient field in each direction, construct a three-dimensional second-order gradient difference matrix by the central difference method, and apply Gaussian smoothing filtering to the three-dimensional second-order gradient difference matrix to eliminate singular points to obtain the second-order gradient field distribution data;

[0046] Combine the first-order gradient field distribution data and the second-order gradient field distribution data into gradient field synthesis data, and perform principal component analysis on the gradient field synthesis data to obtain the spatial vector field characteristic parameters;

[0047] Calculate the signal source inversion equation based on the spatial vector field characteristic parameters to obtain the signal source position coordinate data, and calculate the distribution of the signal intensity in the three-dimensional space according to the signal source position coordinate data to obtain the signal intensity data.

[0048] Specifically, taking the three spatial directions of X, Y, and Z as the main axes, a directional analysis is performed on the three-dimensional measurement point grid deployed in the target area. Each direction contains multiple measurement point nodes with known physical coordinates and acquired frequency-optimized signals. To extract the changing trend of signal intensity in the spatial direction, a difference operation is performed on the amplitudes of the frequency-optimized signals between adjacent measurement points in each direction, obtaining the corresponding signal intensity differences, which reflect the intensity attenuation, interference enhancement, or spatial gradient change characteristics of the geological boundary caused by the propagation of electromagnetic signals in space. These intensity differences are combined into matrix structures in the X, Y, and Z directions according to the actual arrangement of the measurement points in the three-dimensional space, and a three-dimensional first-order gradient difference matrix is constructed and merged. The three-dimensional first-order gradient difference matrix is normalized by dividing it by the actual distance between the measurement points to eliminate the influence of different measurement point spacings on the gradient amplitude. According to the actual spatial distance between each pair of adjacent measurement points, the corresponding gradient difference is divided and normalized to obtain the gradient intensity per unit distance, forming the standardized first-order gradient field distribution data. Based on the first-order gradient field distribution data, the rate of change of the gradient field in each direction is calculated, that is, a second-order derivative structure is constructed. The degree of change of the first-order gradient field itself in space is detected, thereby revealing whether the signal intensity is accelerating or flattening in a certain direction, and further identifying the gradient mutations caused by geological interfaces, anomaly edges, or complex geological structures. The central difference method is used, that is, two points equidistant from the upstream and downstream of a measurement point are selected, and the difference between the first-order gradients of them divided by the distance between the two is used as the estimated value of the second-order gradient at that point. By performing central difference calculations on all valid points in the three-dimensional grid, a three-dimensional second-order gradient difference matrix is constructed. Noise reduction processing is performed on the second-order gradient matrix. The Gaussian smoothing filtering technology is introduced. By applying an isotropic Gaussian convolution kernel to the three-dimensional gradient data and performing a weighted average operation, the influence of singular points is effectively suppressed, and the continuity and physical rationality of the spatial curvature field are improved, obtaining the second-order gradient field distribution data after structure smoothing and anomaly point removal. The first-order gradient field distribution data is merged with the smoothed second-order gradient field data to construct gradient field composite data, which forms a high-dimensional vector tensor field containing first-order information (linear change) and second-order information (curvature change). To extract the most dominant change direction and spatial structure characteristics therein, principal component analysis is performed on the gradient field composite data. By constructing its covariance matrix and solving the eigenvalues and eigenvectors, the main directions with the most significant gradient changes and their weight distributions in the three-dimensional space are identified, providing the optimal coordinate projection direction and gradient change mapping structure for constructing a physical model in subsequent inversion calculations, and forming a set of spatially vector field characteristic parameters with a compact structure but clear physical meanings.Based on the characteristic parameters of the spatial vector field, an inversion mathematical model of the electromagnetic signal source is constructed. This model is based on Maxwell's equations. The main direction of the gradient field, the intensity change rate, and the spatial distribution function are introduced as the input variables of the model. By using the least squares fitting method or the Bayesian estimation method, a quantitative relationship between the signal intensity and the spatial position is established, so as to perform inversion calculations on the true position of the signal source in three-dimensional space. According to the estimation results output by this model, the spatial position coordinate data of the signal source is generated, in the specific form of three-dimensional coordinate values of X, Y, and Z, and this result is fed back into the overall model for iterative correction to ensure that its error converges within the preset range. Based on the known signal source position, combined with the spatial geometric relationship between measurement points and the electromagnetic wave propagation attenuation law, a signal intensity spatial diffusion model is constructed. This model calculates the energy loss situation during the diffusion process of the signal from the source point to each measurement point according to parameters such as the propagation path of the electromagnetic wave in the medium, the impedance matching coefficient, and the reflection and refraction boundary conditions, and generates a signal intensity distribution map in a three-dimensional space grid. This map shows the expanding characteristic that the signal decreases outward from the source point, and at the same time reveals the modulation effect of the geological structure on the propagation path, constituting the signal intensity data.

[0049] Among them, for the signal data optimized in frequency, the signal intensity differences between adjacent measurement points are calculated in the three spatial directions of X, Y, and Z respectively, and a three-dimensional first-order gradient difference matrix is constructed. The differences are normalized by dividing by the actual distance between the measurement points to obtain the spatial first-order gradient field distribution data, including: constructing a three-dimensional coordinate system for the signal data optimized in frequency according to the monitoring network topology structure, converting the geographical coordinates of each measurement point into standard rectangular coordinates, and creating a four-dimensional data index table in combination with the measurement point ID and the acquisition timestamp to obtain spatially structured signal data; performing neighborhood measurement point recognition on the spatially structured signal data, constructing a measurement point adjacency relationship graph based on the Delaunay triangulation algorithm, determining the nearest neighbor measurement points in the X, Y, and Z directions for each measurement point, and calculating the Euclidean distance between each measurement point to obtain the measurement point neighborhood relationship matrix;

[0050] Based on the measurement point neighborhood relationship matrix, by calculating the signal intensity difference ΔS x (i,j) for adjacent measurement point pairs (i,j) in the X direction, where S(i) and S(j) respectively represent the signal intensity values at measurement points i and j, and all the differences in the X direction are combined into a matrix ΔS x , to obtain the signal intensity gradient matrix in the X direction; using the same method as calculating in the X direction, calculate the signal intensity differences ΔS y (i,j) for adjacent measurement point pairs in the Y direction and the signal intensity differences ΔS z (i,j) for adjacent measurement point pairs in the Z direction, and combine all the differences in the Y direction and Z direction into matrices ΔS y and ΔS z, obtain the signal intensity gradient matrices in the Y and Z directions; for the signal intensity gradient matrix ΔS in the X direction x , the signal intensity gradient matrix ΔS in the Y direction y and the signal intensity gradient matrix ΔS in the Z direction z are integrated into a three-dimensional tensor, where i, j, and k respectively represent the measurement point indices in the X, Y, and Z directions, obtaining an initial three-dimensional first-order gradient difference matrix; for each element in the initial three-dimensional first-order gradient difference matrix, divide it by the actual physical distance d(i,j,k) between the corresponding measurement point pairs, perform normalization processing, and apply Gaussian smoothing filtering to eliminate noise interference, obtaining the spatial first-order gradient field distribution data.

[0051] Among them, based on the first-order gradient field distribution data, calculate the change rate of the gradient field in each direction, construct a three-dimensional second-order gradient difference matrix through the central difference method, apply Gaussian smoothing filtering to eliminate singular points, and obtain stable second-order gradient field distribution data, including:

[0052] Perform boundary extension processing on the first-order gradient field distribution data, add virtual grid points outside the calculation domain boundary through the mirror reflection method, maintain the continuity of the first-order gradient field at the boundary, set boundary condition constraints, and obtain the extended first-order gradient field calculation grid; for the first-order gradient values in the X direction in the extended first-order gradient field calculation grid, apply the five-point central difference format to calculate the second-order derivative, obtaining the second-order gradient component in the X direction; adopt the same central difference method as calculating the second-order gradient in the X direction to calculate the second-order derivative of the first-order gradient in the Y direction and the second-order derivative of the first-order gradient in the Z direction, and calculate the cross second-order derivative at the same time, forming a complete second-order gradient tensor, obtaining the original three-dimensional second-order gradient difference matrix; perform singular point detection on the original three-dimensional second-order gradient difference matrix, calculate the spatial aggregation degree and the surrounding gradient field continuity index of each abnormal point by setting threshold conditions, obtaining the singular point distribution map; based on the singular point distribution map, generate a three-dimensional Gaussian smoothing convolution kernel, adaptively adjust the threshold according to the severity of the gradient field change, set different smoothing intensities for different regions, obtaining the adaptive Gaussian filter parameters; apply the adaptive Gaussian filter parameters to the original three-dimensional second-order gradient difference matrix, perform three-dimensional convolution operation, and perform physical consistency verification on the processed result to ensure that the gradient field satisfies the continuity equation and boundary conditions, obtaining stable second-order gradient field distribution data.

[0053] In a specific embodiment, the process of executing step S103 may specifically include the following steps:

[0054] Integrate the signal source position coordinate data and the signal intensity data according to the spatial correspondence relationship to obtain a spatio-temporal domain signal block sequence;

[0055] Perform wavelet packet transform on the spatio-temporal domain signal block sequence to obtain a frequency band weight coefficient table, and based on the frequency band weight coefficient table, perform soft threshold shrinkage processing on the wavelet packet coefficients to obtain optimized wavelet packet coefficients;

[0056] Recombine the optimized wavelet packet coefficients into a signal matrix, and decompose the signal matrix into a left singular matrix, a singular value diagonal matrix, and a right singular matrix through the singular value decomposition algorithm;

[0057] Sort and reconstruct the singular values in the singular value diagonal matrix to obtain the denoised signal reconstruction data, and perform frequency band decomposition on the denoised signal reconstruction data to obtain the enhanced time-frequency distribution signal;

[0058] Perform multi-layer dynamic threshold analysis based on the enhanced time-frequency distribution signal to obtain spatial clustering abnormal data packets.

[0059] Specifically, the signal source position coordinate data is paired with the corresponding signal strength data. Based on the uniqueness of each coordinate point in the three-dimensional space, it is used as an index to associate the strength value, and a spatial point array signal representation model is established. On this basis, the time dimension is introduced, and the signal change sequences of each spatial point at consecutive moments are combined according to the measurement time step length to form a spatio-temporal domain signal block sequence with a three-element structure of spatial position, time sequence, and signal strength value. Each signal block is the electromagnetic response waveform of a certain spatial point within a certain time period, and this sequence reflects the evolution process of the electromagnetic field in the entire monitoring area in terms of time and space. Perform wavelet packet transform processing on the spatio-temporal domain signal block sequence. Use a compactly supported wavelet basis function suitable for electromagnetic pulse analysis (such as Symlet or Daubechies) to perform multi-level, wavelet packet-structured frequency domain expansion on each group of signal blocks, and completely decompose the original time-domain signal at different scales and frequency resolutions. Wavelet packet transform not only decomposes the low-frequency components but also recursively disassembles the high-frequency parts, so it can more comprehensively capture the local energy change characteristics of the signal in multiple frequency sub-bands. By traversing all wavelet sub-bands, calculate the sum of the squares of the wavelet coefficients in each frequency band, that is, the energy value of this frequency band, construct a frequency band weight coefficient table, and record the contribution degree of each frequency band in the entire signal segment. According to the frequency band weight coefficient table, implement a soft threshold shrinkage strategy to perform energy adjustment processing on all coefficients obtained after wavelet packet decomposition. Introduce a higher shrinkage intensity in the frequency bands with smaller weights to achieve weak information suppression and background noise removal; in the high-weight frequency bands corresponding to the main frequency bands, more of the original energy structure is retained, so as to optimize the entire wavelet packet coefficient set and obtain a set of optimized wavelet packet coefficients that retain key features and filter out unstructured disturbances. Reorganize the optimized wavelet packet coefficients structurally to generate a two-dimensional signal matrix, where each row represents the signal feature vector of a spatial point, and each column represents the response characteristics at a certain frequency scale, thus constructing a signal matrix integrating spatio-temporal-frequency triple features. In order to extract the most representative structural components in the signal and eliminate residual redundant information, apply the singular value decomposition algorithm to this signal matrix and decompose it into three parts: a left singular matrix, a singular value diagonal matrix, and a right singular matrix. The left singular matrix represents the main direction of the spatial structure, the right singular matrix reflects the main components in the frequency domain, and each singular value in the singular value diagonal matrix represents the contribution degree of the corresponding main component to the overall signal energy. By sorting these singular values in descending order and setting an energy retention threshold (such as retaining the first 90% of the cumulative energy), identify the secondary components with high redundancy and low signal-to-noise ratio, and set their corresponding singular values to zero, thus completing the singular value reconstruction of the signal. Use the updated singular value diagonal matrix and the left and right singular matrices to perform matrix multiplication to generate the denoised signal reconstruction data.Perform band decomposition once again on the denoised signal, decompose the reconstructed signal to a multi-band scale using the same wavelet packet structure as described above, and construct a time-frequency distribution map based on the frequency center, bandwidth, and local amplitude information of each sub-band. This map has the dual characteristics of time continuity and frequency resolution, and can clearly reflect the instantaneous energy change trajectory of each spatial point in different frequency bands, thereby generating an enhanced time-frequency distribution signal. Based on the enhanced time-frequency distribution signal, introduce a multi-layer dynamic threshold analysis mechanism to achieve automatic discrimination and clustering attribution of potential electromagnetic anomalies. Conduct statistical analysis on the signal amplitudes of each frequency channel, time window, and spatial point, calculate statistical indicators such as the mean μ, standard deviation σ, skewness, and kurtosis, and establish a three-layer progressive dynamic threshold model accordingly. The first-layer threshold is defined as μ + 2σ, which is used to identify mildly abnormal regions; the second-layer threshold is μ + 3σ, which is used to judge significant abnormal behaviors; and the third-layer threshold μ + 4σ is used to locate strongly mutated characteristic signals. Apply this multi-layer discrimination model point by point on the entire time-frequency map, label the signal points that meet any threshold condition as abnormal, and form a high-dimensional abnormal point set. Perform a density clustering algorithm, such as DBSCAN, on all abnormal points, and aggregate the abnormal points that are close to each other in space and have similar spectral characteristics into independent abnormal region clusters using parameters such as the scanning radius ε and the minimum number of neighboring points MinPts. Each abnormal cluster represents an abnormal event with physical unity and geological response consistency, and its characteristics include the spatial coordinate range, abnormal energy center, dominant frequency band, duration, etc. Describe, encode, and package each abnormal clustering region, and finally form a structured spatial clustering abnormal data packet. Each data packet contains key features such as the boundary description of the abnormal region, abnormal level, corresponding frequency band, first occurrence time, maximum duration, abnormal energy density, etc., and is stored in a unified data format.

[0060] In a specific embodiment, the process of performing multi-layer dynamic threshold analysis based on the enhanced time-frequency distribution signal to obtain the spatial clustering abnormal data packet may specifically include the following steps:

[0061] Perform three-dimensional voxel division on the enhanced time-frequency distribution signal to obtain a signal feature voxel data matrix;

[0062] Based on the signal feature voxel data matrix, construct multi-layer threshold discrimination conditions, and input the signal feature voxel data matrix into the multi-layer threshold discrimination conditions for point-by-point evaluation to obtain a preliminary abnormal voxel set;

[0063] Perform density clustering on the preliminary abnormal voxel set to obtain a spatial abnormal clustering set, and based on the spatial abnormal clustering set, calculate the geometric features of each cluster to obtain target clustering boundary data;

[0064] Pack and integrate the target clustering boundary data with the corresponding abnormal voxel signal features to obtain a spatial clustering abnormal data packet.

[0065] Specifically, perform a structured reconstruction on the enhanced time-frequency distribution signal to make it have three-dimensional dividable characteristics. The enhanced signal data is derived from the time-frequency matrix enhanced by methods such as frequency locking, Hilbert transform, and wavelet analysis in the previous module. This matrix itself forms a three-dimensional domain with time, frequency, and spatial position as the three axes, and can be naturally mapped to three-dimensional space and perform discretization processing. According to the preset time step Δt, frequency resolution Δf, and spatial grid scale Δs, perform equally spaced segmentation along the time axis, frequency axis, and spatial axis respectively, and divide the entire enhanced signal space into several three-dimensional voxel units with fixed sizes. Each voxel unit corresponds to a specific time window, a frequency bandwidth, and a geographical area, and stores multiple statistics such as the average signal intensity, main frequency feature, spectral energy density, or instantaneous change rate within this range, thereby constructing a signal feature voxel data matrix. This matrix is essentially a multi-dimensional array with position indices, where the indices are voxel coordinates (x, y, z), and the values are the corresponding feature vectors or scalar values. Based on the signal feature voxel data matrix, construct multi-layer threshold discrimination conditions. Conduct a global statistical analysis on the entire voxel data matrix, extract statistical parameters such as the mean μ, standard deviation σ, kurtosis K, and skewness S of all voxels, and combine historical samples and regional electromagnetic background empirical data to determine a triple abnormal determination interval. The first layer of discrimination conditions is used to identify slight perturbations, and its threshold is set to ; the second layer of conditions is used to mark moderate abnormalities, and the threshold is set to ; while the third layer specifically identifies high-intensity mutation regions, and its threshold is . Introduce a dynamic adjustment mechanism to slightly compensate and correct the above thresholds according to the background drift amount of the electromagnetic environment in different time periods to adapt to the influence of external factors such as seasonal changes, weather disturbances, or power frequency interference. Take the signal feature voxel data matrix as the input and perform point-by-point evaluation with each layer of threshold conditions in turn. For each voxel, judge whether its eigenvalue exceeds the judgment standard of a certain level. If any condition is met, it is marked as an abnormal voxel, and its abnormal level and the discriminant layer it belongs to are recorded to form a preliminary abnormal voxel set. Perform density clustering processing on the preliminary abnormal voxel set to identify abnormal clusters with concentrated distribution characteristics in three-dimensional space. Use a density-based spatial clustering algorithm (such as DBSCAN). This algorithm does not require presetting the number of clusters, but automatically identifies the clustering structure based on the spatial distance and density distribution between voxels. Set the scanning radius ε, that is, any voxel within this radius can be regarded as a "neighboring point", and at the same time set the minimum number of points MinPts, which means that when several voxels in a region simultaneously meet the neighboring conditions, it can be determined that they form an aggregation region. Perform DBSCAN clustering on the abnormal voxel set according to the voxel spatial position. Expand the neighborhood of each voxel identified as a core point to form multiple spatial abnormal clustering sets. Each clustering set physically represents a spatially continuous and energy-concentrated electromagnetic abnormal region, which is caused by geological fractures, tectonic boundaries, aquifer activities, ore body changes, or external electromagnetic sources and has clear abnormality. In order to make each cluster have recognizable boundary attributes and geometric morphological characteristics, extract its geometric parameters for the clustering result. The extraction process of geometric parameters includes calculating key indicators such as the spatial convex hull volume, maximum boundary length, voxel density distribution function, spatial centroid, and boundary irregularity (i.e., surface area to volume ratio) of the clustering voxels, so as to describe the morphological structure, boundary range, and diffusion ability of each abnormal cluster in three-dimensional space. At the same time, according to the statistical average value and peak value of the abnormal levels of each voxel in the cluster, generate the energy level evaluation index of the cluster, and together with the boundary parameters, form a complete clustering boundary description structure. The boundary data is stored in the form of a polygon boundary or a boundary box, which is convenient for subsequent visualization drawing and access to the GIS platform, and at the same time supports subsequent data transmission and packaging. Associate and integrate the generated target clustering boundary data with the corresponding abnormal voxel signal feature data inside it to complete the construction of the spatial clustering abnormal data packet. Each data packet is encapsulated in a structured manner, including core fields such as clustering number, spatial boundary information, abnormal level statistics, main frequency characteristics, number of voxels, time span, energy density, average amplitude change rate, clustering center coordinates, and first appearance time. The size of the data packet is automatically adjusted according to the clustering scale and controlled within a few KB to more than a dozen KB. The system supports packing the data packet in JSON, XML, or a custom binary structure and directly transmitting it to the upper-layer data fusion platform or the cloud abnormal recognition system through an interface to achieve automatic recognition, spatial clustering analysis, and efficient summary of large-scale, multi-source heterogeneous electromagnetic monitoring data.

[0066] In a specific embodiment, the process of executing step S104 may specifically include the following steps:

[0067] Perform block division on the spatially clustered abnormal data packets to obtain an optimized set of data blocks, and perform spatio-temporal differential coding on the optimized set of data blocks to obtain a first data stream;

[0068] Input the first data stream into an adaptive quantizer for data importance grading to obtain a second data stream;

[0069] Based on the second data stream, construct a data packet header including a timestamp, a geographical location identifier, a data priority flag, and an integrity check code, and add Reed-Solomon forward error correction coding to obtain compressed data;

[0070] For the compressed data, detect the signal strength, transmission delay, and bandwidth resources of 4G / 5G cellular networks, LoRa long-range Internet of Things, and satellite communication networks in parallel, establish a network quality scoring matrix, and use a fuzzy decision algorithm to select the optimal transmission channel;

[0071] Packetize the compressed data according to the optimal transmission channel to obtain multiple target data packets, and send the multiple target data packets to the cloud server through heterogeneous network self-organizing handover technology.

[0072] Specifically, the multi-dimensional structured information contained within the spatially clustered anomaly data packets is subjected to functional partitioning. Considering the high-dimensional parameters such as the geospatial coordinates, anomaly level, voxel statistics, time-frequency characteristics of the anomaly signal, duration, first occurrence time, spatial extent, and trend of change in the data packets, the content of the data packets is partitioned into blocks based on the logical associations and access frequencies between the data. The partitioning dimensions include time-related information blocks, spatial structure information blocks, signal feature information blocks, statistical analysis information blocks, and redundancy check information blocks. By storing each type of data centrally and encapsulating it in a fixed format, an optimized set of data blocks is generated. To reduce data redundancy in the time and space dimensions, a spatio-temporal differential encoding operation is performed on the optimized set of data blocks. A reference value is selected between or within the data blocks, and each target value is replaced with the difference between it and the reference value, forming a data representation format dominated by the change amount. Temporal difference is used to express the trend of change in anomaly data between consecutive time instants, while spatial difference is used to express the attribute differences between adjacent regions at the same time instant. Through these two differential methods, the storage bits of repetitive or gradual information are effectively reduced. The output result of the spatio-temporal differential encoding is the first data stream, which consists of a data sequence composed of differences and has a high data compression ratio and good structural sparsity in most scenarios. Taking the first data stream as the input, an adaptive quantizer module is introduced to evaluate the importance and compress the precision of the data units therein. The quantizer assigns an importance level to each data segment in combination with multiple indicators such as the anomaly level, signal strength, spatial centrality, and rarity of the time-frequency structure of the data. High-importance data will be given a higher quantization precision to retain the original information to the greatest extent, while low-importance data will be mapped to a low-precision interval to save bit-width overhead. This process outputs a second data stream, whose content structure is basically the same as that of the first data stream, but the values have been compressed according to the quantization level and carry additional importance flag bits, forming a data content structure optimized for transmission. After the second data stream is constructed, a data packet header is constructed for it to support cross-network transmission, packet integrity verification, and priority scheduling. The header contains multiple key fields: one is the timestamp field, which is used to mark the time when the data is generated, facilitating the receiving end to sort the data and perform timeliness analysis; the second is the geographical location identification field, which encodes the spatial region or grid number to which the data belongs to support spatial distribution modeling and regional analysis; the third is the data priority flag field, which directly comes from the grading result of the aforementioned adaptive quantizer and is used to guide the resource allocation of the network scheduling strategy; the fourth is the integrity check code field, which is generated by the system using cyclic redundancy check or hash function to ensure that the data integrity of the packet can be verified by the receiving end during transmission. To enhance the system's error resistance ability in a weak network environment, the Reed-Solomon forward error correction coding algorithm is adopted to embed redundant information in the entire data segment, enabling the receiving end to automatically recover lost or damaged data segments within a certain range.Through this step, a data compression result with structural compressibility, importance hierarchy, location identification, error correction and fault tolerance, and integrity protection is obtained. Before data transmission, the optimal transmission channel is dynamically selected to adapt to the variability and heterogeneity of the current communication environment. The system parallelly detects three different types of link resources, including 4G / 5G cellular networks, LoRa long-distance low-power Internet of Things communication networks, and satellite communication networks, and real-time obtains key indicators such as signal strength, bandwidth resources, transmission delay, link stability, and data congestion degree of each channel, and constructs a network quality scoring matrix. This scoring matrix is a three-dimensional table, where each row corresponds to an available transmission channel, each column corresponds to a network performance dimension, and each cell stores the performance indicators of the network at the current moment. The system uses a fuzzy decision-making algorithm to comprehensively evaluate the scoring matrix, converts each network performance dimension into a membership function, and selects the channel with the largest comprehensive membership degree as the current optimal transmission path through weighted average and rule reasoning. After determining the transmission path, the compressed data is sub-packaged according to the maximum transmission unit standard of the selected channel, and each data packet contains a complete header, data body segment, and sub-packet identification information to ensure that the cloud server reorganizes the received data and restores the original structure. These target data packets are sent to the cloud server through the heterogeneous network self-organizing handover technology, which supports dynamic switching between multiple networks. For example, when the packet loss rate of the LoRa network increases or the 5G network is congested, the data transmission link is automatically migrated without interrupting the transmission process, ensuring the continuity and integrity of the data. In addition, an intelligent caching mechanism is introduced, which can temporarily cache data packets when the network is interrupted and automatically retransmit them after the network is restored; it supports preemptive scheduling when high-priority data arrives, inserts data with higher importance at the front of the current transmission queue for transmission, ensuring that the system has high availability and data grading response capabilities. All data packets are safely and efficiently uploaded to the cloud server, and after a series of processes such as decoding, dequantization, inverse differential reduction, singular value reconstruction, and voxel fusion, they are restored to complete spatial anomaly data.

[0073] Among them, the optimal transmission channel is selected through the heterogeneous network self-organizing handover technology to transmit the compressed data to the cloud server, including: classifying the transmission priorities of the compressed data, and dividing the data into three priorities: urgent, important, and regular based on the stress intensity, change rate, and spatial range of the abnormal clustering data, and assigning transmission resource weight coefficients to each priority to obtain a data transmission priority list; deploying a solar power supply system energy management module at the monitoring node, collecting current energy data through a light intensity sensor and a battery status monitor, combining historical energy consumption records and meteorological forecast data, establishing an energy availability prediction model, and obtaining future 24-hour energy distribution prediction data; fusing the energy distribution prediction data with the data transmission priority list, constructing a multi-level transmission management architecture including environmental perception, energy prediction, task scheduling, and execution control, setting an initial transmission strategy through heuristic rules, and obtaining a set of transmission parameter configurations; applying the DSAC-CAL (Distributed Soft Actor-Critic Conservative Actor-Lagrangian) algorithm to the set of transmission parameter configurations, integrating the energy constraint conditions and the data transmission objectives into the deep reinforcement learning framework, training the transmission policy network by setting the state space, action space, and reward function, and obtaining an optimal transmission policy model under energy constraints; based on the optimal transmission policy model, dynamically adjusting the parameters of the wireless transmission module, including transmission power, data sending interval, compression rate adjustment, and processor frequency, setting a four-level energy management mode according to the battery power status, automatically starting the extreme power-saving mode when the power is lower than 20%, and obtaining adaptive transmission control parameters; according to the adaptive transmission control parameters, performing real-time performance evaluation on various network interfaces, calculating the transmission efficiency indexes of the current 4G / 5G, LoRa, and satellite communication networks, combining the transmission data volume and the real-time network quality, and selecting the current optimal transmission channel combination through a multi-objective optimization algorithm to obtain a multi-channel collaborative transmission scheme; intelligently subcontracting the compressed data according to the multi-channel collaborative transmission scheme, implementing a redundant transmission strategy for the urgent priority data, sending the same data packet in parallel through multiple channels, adopting a primary and backup channel mechanism for the important priority data, and selecting the most economical channel for the regular data according to the bandwidth cost to obtain differentiated transmission data packets; monitoring the transmission status of the differentiated transmission data packets, recording the sending status, confirmation reception situation, and transmission delay of each data packet, triggering an automatic retransmission mechanism when a transmission anomaly is detected, and using the transmission quality feedback information as a reward signal for the reinforcement learning system for online learning to continuously optimize the transmission strategy to achieve high-reliability data transmission under energy-constrained conditions.

[0074] Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the above-described systems, systems, and units can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0075] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to enable an electromagnetic pulse data transmission device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs.

[0076] As described above, the above embodiments are only used to illustrate the technical solution of the present invention and are not intended to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. An electromagnetic pulse data transmission method, characterized in that: include: The natural electromagnetic pulse signal data of the detection target area is collected through a six-dimensional complementary electrode array structure, and Kalman filter peak extraction and characteristic frequency locking are performed to obtain frequency-optimized signal data; Performing first-order and second-order gradient difference matrix calculation and space vector synthesis on the frequency-optimized signal data to obtain signal source position coordinate data and signal strength data; Performing spectral domain reconstruction and multi-layer dynamic threshold analysis on the signal source position coordinate data and the signal strength data to obtain spatial clustering abnormal data packets; Block-level differential compression is performed on the spatial clustering abnormal data packets to obtain compressed data, and the optimal transmission channel is selected through the heterogeneous network self-organizing switching technology to transmit the compressed data to the cloud server.

2. The electromagnetic pulse data transmission method according to claim 1, characterized in that: The method collects the natural electromagnetic pulse signal data of the detection target area through the six-dimensional complementary electrode array structure, performs Kalman filter peak extraction and characteristic frequency locking, and obtains frequency-optimized signal data, including: Determine the electrode array deployment plan for the detection target area according to an equilateral triangle arrangement pattern, and install a six-dimensional complementary electrode array including three pairs of orthogonal vector electromagnetic induction coils at each set receiving point; Install high-precision GPS clock modules on all six-dimensional complementary electrode arrays and configure low-power wireless transmission modules to build a mesh network architecture to obtain a real-time data transmission system; Setting E reference sites in the detection target area, and recording the external electromagnetic interference signal of the area through each reference site to obtain noise calibration benchmark data; The detection target area is monitored by the real-time data transmission system to obtain original electromagnetic pulse vector data; Perform correlation calculation and calibration based on the original electromagnetic pulse vector data and the noise calibration reference data to obtain natural electromagnetic pulse signal data; The natural electromagnetic pulse signal data is input into a spectrum adaptive tracking unit to perform Kalman filtering peak extraction and characteristic frequency locking to obtain frequency-optimized signal data.

3. The electromagnetic pulse data transmission method according to claim 2, characterized in that: The step of inputting the natural electromagnetic pulse signal data into a spectrum adaptive tracking unit for Kalman filtering peak extraction and characteristic frequency locking to obtain frequency-optimized signal data includes: Performing power frequency notch filtering and wavelet threshold denoising processing on the natural electromagnetic pulse signal data to obtain a denoised electromagnetic pulse signal; The electromagnetic pulse signal after noise reduction is converted into the frequency domain by a spectrum adaptive tracking unit, the power spectrum density is calculated by discrete Fourier transform, and the characteristic frequency points are marked according to the local maximum principle to obtain initial spectrum characteristic data; Constructing a state space model according to the initial spectrum feature data, and performing state estimation and prediction through the state space model to obtain filtered spectrum trajectory data; Based on the filtered spectrum trajectory data, the drift rate of the spectrum peaks in three adjacent time windows is calculated, and the frequency points whose stability coefficients are greater than a preset target value are marked as locked frequencies according to the drift rate to obtain a frequency locking parameter set; Performing adaptive window Hilbert transform based on the frequency locking parameter set to obtain instantaneous frequency and instantaneous amplitude, and performing time-frequency joint analysis through the instantaneous frequency and the instantaneous amplitude to obtain an enhanced time-frequency feature matrix; The non-characteristic frequency components in the enhanced time-frequency characteristic matrix are attenuated by 50%, the signal integrity of the locked frequency is retained, and frequency-optimized signal data is obtained through inverse transformation.

4. The electromagnetic pulse data transmission method according to claim 1, characterized in that: The performing first-order and second-order gradient difference matrix calculation and space vector synthesis on the frequency-optimized signal data to obtain signal source position coordinate data and signal strength data includes: Calculating the signal strength differences between adjacent measuring points in the three spatial directions of X, Y and Z for the frequency-optimized signal data, constructing a three-dimensional first-order gradient difference matrix, and dividing the three-dimensional first-order gradient difference matrix by the actual distance between the measuring points for normalization processing to obtain first-order gradient field distribution data; Based on the first-order gradient field distribution data, the rate of change of the gradient field in each direction is calculated, a three-dimensional second-order gradient difference matrix is ​​constructed by a central difference method, and a Gaussian smoothing filter is applied to the three-dimensional second-order gradient difference matrix to eliminate singular points, thereby obtaining second-order gradient field distribution data; Combining the first-order gradient field distribution data with the second-order gradient field distribution data into gradient field synthesis data, and performing principal component analysis on the gradient field synthesis data to obtain spatial vector field characteristic parameters; The signal source inversion equation is calculated based on the spatial vector field characteristic parameters to obtain signal source position coordinate data, and the distribution of signal strength in three-dimensional space is calculated according to the signal source position coordinate data to obtain signal strength data.

5. The electromagnetic pulse data transmission method according to claim 1, characterized in that: The performing spectral domain reconstruction and multi-layer dynamic threshold analysis on the signal source position coordinate data and the signal strength data to obtain a spatial clustering abnormal data packet includes: The signal source position coordinate data and the signal strength data are merged according to a spatial correspondence to obtain a time-space domain signal block sequence; Performing wavelet packet transformation on the time-space domain signal block sequence to obtain a frequency band weight coefficient table, and performing soft threshold shrinkage processing on the wavelet packet coefficients based on the frequency band weight coefficient table to obtain optimized wavelet packet coefficients; Reorganizing the optimized wavelet packet coefficients into a signal matrix, and decomposing the signal matrix into a left singular matrix, a singular value diagonal matrix and a right singular matrix by a singular value decomposition algorithm; Sorting and reconstructing the singular values ​​in the singular value diagonal matrix to obtain denoised signal reconstruction data, and performing frequency band decomposition on the denoised signal reconstruction data to obtain an enhanced time-frequency distribution signal; Multi-layer dynamic threshold analysis is performed according to the enhanced time-frequency distribution signal to obtain spatially clustered abnormal data packets.

6. The electromagnetic pulse data transmission method according to claim 5, characterized in that: The performing of multi-layer dynamic threshold analysis according to the enhanced time-frequency distribution signal to obtain spatially clustered abnormal data packets includes: Performing three-dimensional voxel division on the enhanced time-frequency distribution signal to obtain a signal feature voxel data matrix; Based on the signal feature voxel data matrix, a multi-layer threshold discrimination condition is constructed, and the signal feature voxel data matrix is ​​input into the multi-layer threshold discrimination condition for point-by-point evaluation to obtain a preliminary abnormal voxel set; Density clustering is performed on the preliminary abnormal voxel set to obtain a spatial abnormal cluster set, and based on the spatial abnormal cluster set, geometric features of each cluster are calculated to obtain target cluster boundary data; The target cluster boundary data and the corresponding abnormal voxel signal features are packaged and integrated to obtain a spatial cluster abnormal data packet.

7. The electromagnetic pulse data transmission method according to claim 1, characterized in that: The block-level differential compression is performed on the spatial clustering abnormal data packets to obtain compressed data, and the optimal transmission channel is selected through the heterogeneous network self-organizing switching technology to transmit the compressed data to the cloud server, including: Dividing the spatially clustered abnormal data packets into blocks to obtain an optimized data block set, and performing spatiotemporal differential coding on the optimized data block set to obtain a first data stream; Inputting the first data stream into an adaptive quantizer to perform data importance classification to obtain a second data stream; Based on the second data stream, construct a data message header including a timestamp, a geographic location identifier, a data priority flag, and an integrity check code, and add Reed-Solomon forward error correction coding to obtain compressed data; For the compressed data, a network quality scoring matrix is ​​established by parallelly detecting the signal strength, transmission delay and bandwidth resources of 4G / 5G cellular networks, LoRa long-distance IoT and satellite communication networks, and a fuzzy decision algorithm is used to select the optimal transmission channel; The compressed data is packetized according to the optimal transmission channel to obtain multiple target data packets, and the multiple target data packets are sent to the cloud server through heterogeneous network self-organizing switching technology.

Images