Intrusion Detection Method Based on Low-Frequency and Microwave Composite Induction

Abstract

This invention provides an intrusion detection method based on low-frequency and microwave composite sensing, belonging to the field of intrusion detection technology. The invention determines the distribution distance of vibration sensors based on the low-frequency signal intensity of the warning target type and installs them accordingly. It performs mode decomposition on the low-frequency signal and identifies the above-ground and underground mode components based on the mode energy ratio. A low-frequency feature set is formed based on the vibration location of the received low-frequency signal and low-frequency frequency domain characteristics. A microwave feature set is formed based on the detection location of the microwave target and after decomposing and denoising the echo signal. An intrusion target identification model is established, and the warning target type is identified based on the low-frequency and microwave feature sets. The mode energy ratio judgment threshold is adjusted based on soil moisture and surface water, and the noise threshold is adjusted based on rainfall and air humidity, thereby improving the accuracy of warning target type identification.

Description

Technical Field

[0001] This invention relates to the field of intrusion detection technology, specifically to an intrusion detection method based on a combination of low-frequency and microwave induction. Background Technology

[0002] In modern security and intrusion detection systems, traditional monitoring methods typically rely on a single type of sensor, such as infrared sensors, ground vibration sensors, or microwave radar. However, these single sensors often have limitations. For example, infrared sensors are susceptible to weather conditions, microwave radar can be affected by physical obstacles, and vibration sensors exhibit significant differences in response to different soil types. These traditional methods often fail to provide sufficient accuracy and robustness in complex environments. Therefore, combining the advantages of multiple sensors, especially under varying environmental factors such as soil type, humidity, and water accumulation, to achieve effective intrusion detection has become a pressing technical challenge.

[0003] The existing technology, disclosed in CN 112687068 B, discloses an intrusion detection method based on microwave and vibration sensor data. The main steps include: Step 1: Testing different intrusion behaviors to obtain different characteristic waveforms representing different event types for different scenarios, serving as feature wave templates; Step 2: Acquiring sensor signal output signals and counting the number of times the sensor voltage intensity exceeds a threshold per unit time, obtaining data feature waves reflecting the intensity of the sensor output signal change over "number of times - time"; Step 3: Matching the similarity between the feature wave templates and the data feature waves, and determining the event type based on the similarity; Step 4: Outputting different alarm results based on combinations of event types. However, this method only detects intrusion behavior through changes in sensor output voltage intensity and identifies event types by matching feature wave templates, without considering the impact of environmental changes.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide an intrusion detection method based on low-frequency and microwave composite induction to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] An intrusion detection method based on low-frequency and microwave combined induction, comprising the following steps:

[0008] Step 1: Obtain soil type, warning target type, and environmental data for the area to be alarmed. Using the fixed variable method, obtain the low-frequency signal intensity of the vibration source in different soil types for different environmental data for the warning target type. Determine the distribution distance of the vibration sensor based on the low-frequency signal intensity and install it.

[0009] Step 2: Acquire real-time low-frequency signals, perform modal decomposition on the low-frequency signals using the modal decomposition method, identify the above-ground and underground modal components based on the modal energy ratio, reconstruct the above-ground and underground modal components respectively, acquire the low-frequency reconstructed signals from the above-ground and underground, and perform time-frequency analysis on the reconstructed signals;

[0010] Step 3: Determine the vibration location of the low-frequency signal source based on the time difference of the received low-frequency signal, and form a low-frequency feature set with the low-frequency frequency domain features. Based on the detection location of the microwave detection target, and after decomposing and denoising the echo signal, form a microwave feature set.

[0011] Step 4: Establish an intrusion target identification model. Use low-frequency feature sets and microwave feature sets as input data for the model, and the warning target type and coordinate location as output data to train the model. Based on real-time input data, use the trained model to identify the warning target type.

[0012] Step 5: Adjust the threshold for judging the modal energy ratio based on soil moisture and surface water to remove the influence of water accumulation on the low-frequency signal offset. Adjust the noise threshold based on rainfall and air humidity to remove the influence of the environment on microwave echo characteristics.

[0013] Furthermore, the soil types include: hardened soil, sandy soil, and loose soil;

[0014] The types of targets for early warning include ground intrusion: personnel, vehicles, and animals;

[0015] Underground intrusion: drilling, subsidence;

[0016] The environmental data includes soil moisture, water depth, rainfall, and air humidity;

[0017] The method for obtaining low-frequency signals of vibration sources of different types of warning targets in different environments and with different soil types using the fixed variable method is as follows:

[0018] At the vibration source and at a distance of , respectively Vibration sensors are installed at the location. By sequentially changing one set of data types—the warning target type, environmental data, and soil type—while keeping the remaining two data types unchanged, the low-frequency signal strength of the warning target type is detected.

[0019] Furthermore, the method for determining the distribution distance of the vibration sensor based on the low-frequency signal intensity is as follows:

[0020] The distribution distance is calculated by calculating the attenuation coefficient of the low-frequency signal intensity of each group of data, and based on the attenuation coefficient, the low-frequency signal of each group of early warning target types, and the detection threshold of the vibration sensor.

[0021] ;

[0022] ;

[0023] in, The vibration attenuation coefficient is... The distance detected by the vibration sensor is the distance to the vibration source. Low-frequency signal strength at that location The intensity of the low-frequency signal at the vibration source detected by the vibration sensor. The distribution distance of the vibration sensors;

[0024] The installation method is as follows: Install sensors according to the distribution distance so that at least 3 vibration sensors can detect any position on the boundary of the area to be detected intrusion.

[0025] Furthermore, the specific steps for performing mode decomposition on the low-frequency signal using the mode decomposition method are as follows:

[0026] Set the number of modal components Broadband constraints initial value of center frequency For each modal component , =1,2,…, and angular frequency Update: Determine whether the low-frequency signal decomposition is complete based on the convergence condition, and output the modal components;

[0027] Iterative update of modal components:

[0028] ;

[0029] in, For the first The first modal component in the frequency domain Frequency domain value of the next iteration For the number of iterations, It is a low-frequency signal. Angular frequency, for Modal classification in The center frequency of the next iteration For the first The frequency domain values ​​of each modal component in the frequency domain. For the frequency domain representation of Lagrange multipliers, ;

[0030] Update center frequency:

[0031] ;

[0032] in, for Modal classification in The center frequency of the next iteration;

[0033] Determine whether to terminate the update based on the convergence criteria:

[0034] ;

[0035] in, To determine accuracy.

[0036] Furthermore, based on the modal energy ratio, the modal components above ground and below ground are identified. The formula for calculating the modal energy ratio is:

[0037] ;

[0038] in, The modal energy ratio, For the first The frequency representation of each modal component For frequency integral, These are the threshold values ​​for dividing the frequencies of above-ground and underground modal components, respectively.

[0039] when At that time, it was determined that the modal components originated from the underground portion:

[0040] when When determining that the modal components originate from the ground surface:

[0041] in, The threshold for determining the modal energy ratio.

[0042] Furthermore, the method for recombining the above-ground and underground modal components to obtain the low-frequency recombined signals from the above-ground and underground is as follows:

[0043] ;

[0044] ;

[0045] in, , These are the recombined signals from underground and above ground, respectively. To determine the number of modal components in the underground part, To determine the number of modal components in the aboveground portion, for The first moment One modal component.

[0046] Furthermore, the method for determining the vibration location of the low-frequency signal source based on the time difference of the received low-frequency signal is as follows:

[0047] Based on the time difference in the reception of low-frequency signals by the vibration sensor, the vibration location of the low-frequency signal source can be determined by using the time difference and the propagation speed of the low-frequency signal.

[0048] ;

[0049] ;

[0050] in, , , , These are vibration sources, the first one. , , The coordinates of each sensor For the propagation speed of low-frequency signals, For the first , The time difference between the detection of low-frequency signals by each sensor For the first , The time difference between the detection of low-frequency signals by the sensors.

[0051] Furthermore, the specific steps for forming a microwave feature set after decomposing and denoising the echo signal are as follows:

[0052] The detection location of the target is identified based on the echo time of the microwave detection target, and the echo signal is decomposed and denoised by wavelet transform to obtain detail coefficient features. A microwave feature set is formed based on the detection location and detail coefficient features.

[0053] First, the echo signal is subjected to wavelet transform, which decomposes it into a series of approximation coefficients and detail coefficients of different frequencies;

[0054] ;

[0055] in, For echo signal, These are approximate coefficients. For detail coefficients, For scaling function, For wavelet functions, The time point for acquiring communication signals. These are the scale parameters and displacement parameters, respectively.

[0056] Noise signals in the echo signal can be removed by setting a noise threshold.

[0057] ;

[0058] in, The detail coefficient after noise reduction. The noise threshold is set.

[0059] Furthermore, the environmental data includes soil moisture, water depth, rainfall, and air humidity;

[0060] The steps for adjusting the modal energy ratio based on soil moisture and surface water are as follows:

[0061] The dielectric constant of the soil is corrected based on soil moisture:

[0062] ;

[0063] in, The corrected dielectric constant is... To detect the basic dielectric constant of the soil in the region, For soil moisture, The dielectric constant of moisture-saturated soil. The dielectric constant of dry soil;

[0064] Determine the water accumulation attenuation coefficient based on the accumulated water:

[0065] ;

[0066] in, The water accumulation attenuation coefficient is... The basic water accumulation attenuation coefficient for soil type. This refers to the depth of the accumulated water. The center frequency of the low-frequency signal;

[0067] The modal energy ratio judgment threshold is adaptively adjusted based on the corrected dielectric constant and water accumulation attenuation coefficient. The calculation formula is as follows:

[0068] ;

[0069] in, The threshold for determining the corrected modal energy ratio is... The threshold for determining the modal energy ratio;

[0070] The specific steps for adjusting the noise threshold based on rainfall and air humidity are as follows:

[0071] ;

[0072] in, The corrected noise threshold. To set a basic noise threshold, This is the fundamental correction constant for the transmitted microwave frequency. For rainfall, This refers to air humidity.

[0073] Compared with the prior art, the beneficial effects of the present invention are as follows: the distribution distance of vibration sensors is determined according to the low-frequency signal intensity of the warning target type, and the sensors are installed; the low-frequency signal is decomposed into modes, and the above-ground and underground mode components are identified according to the mode energy ratio; a low-frequency feature set is formed based on the vibration position of the received low-frequency signal and the low-frequency frequency domain features; a microwave feature set is formed based on the detection position of the microwave detection target and after decomposing and denoising the echo signal; an intrusion target identification model is established; the warning target type is identified based on the low-frequency feature set and the microwave feature set; the mode energy ratio judgment threshold is adjusted according to soil moisture and surface water; and the noise threshold is adjusted according to rainfall and air humidity, thereby improving the identification accuracy of the warning target type.

[0074] This invention, based on the composite induction of low-frequency vibration signals and microwave echo signals, fully utilizes the complementarity of the two signals to provide more accurate intrusion detection in complex environments. First, it uses mode decomposition and time-frequency analysis of the low-frequency vibration signals to detect both above-ground and underground targets separately. Second, the microwave echo signals undergo wavelet transform noise reduction processing, effectively removing noise caused by environmental changes, further improving the accuracy and stability of the microwave signals.

[0075] This invention further improves the adaptability of low-frequency signals and microwave echo signals under different conditions by combining the influence of environmental factors (such as soil moisture and water depth) on signal characteristics and adopting a dynamic adjustment method. Through the fusion of multimodal features, higher-precision target recognition and intrusion detection can be achieved. Especially in harsh weather and complex environments, this method can significantly improve the reliability and accuracy of intrusion detection. Attached Figure Description

[0076] Figure 1 This is a schematic diagram of the overall method flow of the present invention. Detailed Implementation

[0077] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0078] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0079] Example:

[0080] Please see Figure 1 The present invention provides a technical solution:

[0081] An intrusion detection method based on low-frequency and microwave combined induction, comprising the following steps:

[0082] Step 1: Obtain soil type, warning target type, and environmental data for the area to be alarmed. Using the fixed variable method, obtain the low-frequency signal intensity of the vibration source in different soil types for different environmental data for the warning target type. Determine the distribution distance of the vibration sensor based on the low-frequency signal intensity and install it.

[0083] In this embodiment, the soil types include: hardened soil, sandy soil, and loose soil;

[0084] The types of targets for early warning include ground intrusion: personnel, vehicles, and animals;

[0085] Underground intrusion: drilling, subsidence;

[0086] The environmental data includes soil moisture, water depth, rainfall, and air humidity;

[0087] The method for obtaining low-frequency signals of vibration sources of different types of warning targets in different environments and with different soil types using the fixed variable method is as follows:

[0088] At the vibration source and at a distance of , respectively Vibration sensors are installed at the location. By sequentially changing one set of data types—the warning target type, environmental data, and soil type—while keeping the remaining two data types unchanged, the low-frequency signal strength of the warning target type is detected.

[0089] Environmental factors such as soil type, moisture content, water depth, and rainfall significantly influence the propagation characteristics of vibration signals. For example, the vibration propagation characteristics of hardened soil differ from those of loose or sandy soil, potentially leading to variations in signal attenuation, propagation range, and sensitivity. Soil moisture and water depth also affect soil density and viscosity, thus influencing the intensity of low-frequency signals. Therefore, the fixed-variable method can eliminate interference from unstable environmental factors, enabling accurate calculation of the impact of each soil type and different early warning targets on low-frequency signals under varying environmental conditions.

[0090] By using the fixed-variable method, the intensity of low-frequency vibration signals caused by different types of early warning targets (such as personnel, vehicles, animals, and wells) can be determined under known environmental conditions such as soil type, humidity, and water accumulation. This helps to establish a stable benchmark model for precise vibration sensor deployment based on different soil types and environmental conditions. Knowing the signal intensity under each soil type and environmental factor allows for a more reasonable determination of sensor distribution distances, enabling the system to effectively detect target vibration signals in various complex environments.

[0091] Soil type and environmental conditions significantly impact the performance and detection range of vibration sensors. Different soil types exhibit varying vibration signal attenuation rates, resulting in different sensor sensitivities at different locations. Therefore, by employing a fixed-variable method, the signal intensity of the vibration source can be accurately calculated under different environmental data, thereby determining the optimal installation location and spacing of the vibration sensors. This optimized sensor layout not only improves vibration signal acquisition but also enhances the overall system efficiency and accuracy, reducing unnecessary false alarms or missed alarms.

[0092] In this embodiment, the method for determining the distribution distance of the vibration sensor based on the low-frequency signal intensity is as follows:

[0093] The distribution distance is calculated by calculating the attenuation coefficient of the low-frequency signal intensity of each group of data, and based on the attenuation coefficient, the low-frequency signal of each group of early warning target types, and the detection threshold of the vibration sensor.

[0094] ;

[0095] ;

[0096] in, The vibration attenuation coefficient is... The distance detected by the vibration sensor is the distance to the vibration source. Low-frequency signal strength at that location The intensity of the low-frequency signal at the vibration source detected by the vibration sensor. The distribution distance of the vibration sensors;

[0097] The installation method is as follows: Install sensors according to the distribution distance so that at least 3 vibration sensors can detect any position on the boundary of the area to be detected intrusion.

[0098] Step 2: Acquire real-time low-frequency signals, perform modal decomposition on the low-frequency signals using the modal decomposition method, identify the above-ground and underground modal components based on the modal energy ratio, reconstruct the above-ground and underground modal components respectively, obtain the reconstructed low-frequency signals from the above-ground and underground, and perform time-frequency analysis on the reconstructed signals.

[0099] Mode decomposition (MD) is a signal processing method that decomposes a raw signal into different modal components, thereby extracting different features from the signal. Different modal components correspond to different vibration modes, each with a specific frequency, amplitude, and phase. MMD can separate vibration signals from different sources, both above and below ground, eliminating environmental noise and irrelevant signal components. This makes the useful information in low-frequency signals clearer. This is because vibration signals above ground (e.g., from pedestrians, vehicles, and machinery) typically have higher frequencies and faster propagation speeds, with their propagation paths influenced by both the ground and air. Underground vibrations, on the other hand, typically have lower frequencies, and their propagation is primarily affected by the underground soil, resulting in slower propagation speeds and different attenuation patterns compared to surface signals. Due to these physical differences, above-ground and underground signals exhibit different characteristics in the frequency and time domains, particularly in terms of frequency characteristics, energy distribution, and propagation modes.

[0100] Mode decomposition allows signals to be divided into different modal components based on their frequency composition. Signals from different sources have varying energy distributions, and mode decomposition can effectively separate these signals, allowing surface and underground signals to have different frequency domain characteristics, thus enabling differentiation.

[0101] In this embodiment, the specific steps for performing mode decomposition on the low-frequency signal using the mode decomposition method are as follows:

[0102] Set the number of modal components Broadband constraints initial value of center frequency For each modal component , =1,2,…, and angular frequency Update: Determine whether the low-frequency signal decomposition is complete based on the convergence condition, and output the modal components;

[0103] Iterative update of modal components:

[0104] ;

[0105] in, For the first The first modal component in the frequency domain Frequency domain value of the next iteration For the number of iterations, It is a low-frequency signal. Angular frequency, for Modal classification in The center frequency of the next iteration For the first The frequency domain values ​​of each modal component in the frequency domain. For the frequency domain representation of Lagrange multipliers, ;

[0106] Update center frequency:

[0107] ;

[0108] in, for Modal classification in The center frequency of the next iteration;

[0109] Determine whether to terminate the update based on the convergence criteria:

[0110] ;

[0111] in, To determine accuracy.

[0112] In this embodiment, the modal components above ground and below ground are identified based on the modal energy ratio. The formula for calculating the modal energy ratio is as follows:

[0113] ;

[0114] in, The modal energy ratio, For the first The frequency representation of each modal component For frequency integral, These are the threshold values ​​for dividing the frequencies of above-ground and underground modal components, respectively.

[0115] when At that time, it was determined that the modal components originated from the underground portion:

[0116] when When determining that the modal components originate from the ground surface:

[0117] in, The threshold for determining the modal energy ratio.

[0118] By recombining the modal components from above-ground and underground sources, we can obtain low-frequency signals associated with each type of intrusion target. This signal separation allows the system to more accurately identify different types of intrusion targets. Above-ground and underground vibration signals may overlap or interfere with each other in certain frequency bands. The recombining step effectively reduces interference between different signal sources, making the characteristics of each signal clearer and more independent, which is helpful for subsequent processing and identification.

[0119] Time-frequency analysis (TFA) refers to the simultaneous analysis of a signal in terms of both time and frequency. It typically uses the short-time Fourier transform. Intrusion signals (such as those involving people, vehicles, and drilling) are usually non-stationary, meaning their frequency and amplitude change over time. TFA can capture the spectral characteristics of a signal over time, which is crucial for identifying and monitoring these non-stationary intrusion signals. Furthermore, TFA provides the instantaneous frequency and temporal characteristics of a signal, accurately reflecting frequency changes at different points in time, further improving the accuracy of intrusion detection. For example, TFA clearly shows the start, duration, and end times of an intrusion event, helping to make more accurate alarm decisions.

[0120] In this embodiment, the method for recombining the above-ground and underground modal components to obtain the low-frequency recombined signals from the above-ground and underground is as follows:

[0121] ;

[0122] ;

[0123] in, , These are the recombined signals from underground and above ground, respectively. To determine the number of modal components in the underground part, To determine the number of modal components in the aboveground portion, for The first moment One modal component.

[0124] The method for performing time-frequency analysis on the recombined signal is as follows:

[0125] By using a sliding window function on the reconstructed signal through short-time Fourier transform, the reconstructed signal is divided into shorter segments. Then, a Fourier transform is performed on each segment, thereby simultaneously obtaining the time-frequency characteristics of the reconstructed signal. The specific calculation formula is as follows:

[0126] ;

[0127] in, In order to be in The time-frequency characteristics of the reconstructed signal at a given time. This is a recombination signal. For window functions, in The center position of the time window at any given moment is , For the frequency variable of the reconstructed signal, The number of samples within the window. It is the imaginary unit.

[0128] Step 3: Determine the vibration location of the low-frequency signal source based on the time difference of the received low-frequency signal, and form a low-frequency feature set with the low-frequency frequency domain features. Based on the detection location of the microwave detection target, and after decomposing and denoising the echo signal, form a microwave feature set.

[0129] A significant advantage of time-difference positioning (TDRP) is its ability to pinpoint the location of vibration sources with high precision. By coordinating multiple sensors, the position of the vibration source relative to these sensors can be calculated based on the time difference. This is extremely helpful for accurately locating low-frequency signal sources, especially in complex environments (such as underground or obstructed areas). It can effectively eliminate interference sources and accurately identify vibration sources. When different sensors receive low-frequency signals, the signals themselves may be distorted due to noise, interference, or different signal propagation paths. However, through time-difference analysis, we can effectively filter out some irrelevant noise sources. This is because genuine low-frequency signal sources will exhibit consistent time-difference changes across multiple sensors, while noise or interference signals often lack this consistency. Therefore, time-difference-based positioning can enhance the system's noise suppression and improve the accuracy of low-frequency signal identification.

[0130] A planar coordinate system is established with the center of the intrusion detection area as the origin, the south direction as the positive direction of the vertical axis, the east direction as the positive direction of the horizontal axis, and the vertical upward direction as the positive direction of the vertical axis.

[0131] Using a fixed, standardized coordinate system makes data processing more unified across the entire system. Whether it's multiple sensors, monitoring equipment, or other detection systems, as long as a unified coordinate origin and coordinate axis definition are used, all measurement results can be directly compared and merged, avoiding the conversion problems between different coordinate systems. When a fixed coordinate system is used and the directions of each axis are set, describing the target position becomes more concise and standardized. For example, the position of a vibration source can be described by three coordinate values ​​(x, y, z), where x and y are directly related to the target's relative position in the plane, and the z coordinate describes the displacement in the vertical direction. This approach simplifies the analysis and judgment in the target positioning process, improving the accuracy and operability of the positioning system.

[0132] In this embodiment, the method for determining the vibration position of the low-frequency signal source based on the time difference of the received low-frequency signal is as follows:

[0133] Based on the time difference in the reception of low-frequency signals by the vibration sensor, the vibration location of the low-frequency signal source can be determined by using the time difference and the propagation speed of the low-frequency signal.

[0134] ;

[0135] ;

[0136] in, , , , These are vibration sources, the first one. , , The coordinates of each sensor For the propagation speed of low-frequency signals, For the first , The time difference between when each sensor detects a low-frequency signal For the first , The time difference between the detection of low-frequency signals by the sensors.

[0137] Microwave echo signals are often affected by various noise sources, such as environmental noise and interference from electronic devices. This noise degrades signal quality and affects the accuracy of intrusion detection. Wavelet transform can localize signals through multi-scale analysis. It decomposes the signal into approximate and detail coefficients. By selecting appropriate wavelet functions and thresholding methods, wavelet transform can decompose the signal at different scales, suppressing high-frequency noise components while preserving useful signal information. This effectively separates noise components from the signal, thus achieving effective noise reduction.

[0138] The detail coefficients after noise reduction contain high-frequency information from the signal, reflecting the signal characteristics and variation features across different frequency ranges. This is particularly important for capturing local changes in microwave signals. In microwave echo signals, detail coefficients are typically closely related to changes in the motion, structure, or shape of the target object. Therefore, extracting detail coefficient features can yield valuable target information.

[0139] In this embodiment, the specific steps for forming a microwave feature set after decomposing and denoising the echo signal are as follows:

[0140] The detection location of the target is identified based on the echo time of the microwave detection target, and the echo signal is decomposed and denoised by wavelet transform to obtain detail coefficient features. A microwave feature set is formed based on the detection location and detail coefficient features.

[0141] First, the echo signal is subjected to wavelet transform, which decomposes it into a series of approximation coefficients and detail coefficients of different frequencies;

[0142] ;

[0143] in, For echo signal, These are approximate coefficients. For detail coefficients, For scaling function, For wavelet functions, This refers to the time point when the communication signal was acquired. These are the scale parameters and displacement parameters, respectively.

[0144] Noise signals in the echo signal can be removed by setting a noise threshold.

[0145] ;

[0146] in, The detail coefficient after noise reduction. The noise threshold is set.

[0147] Step 4: Establish an intrusion target identification model. Use low-frequency feature sets and microwave feature sets as input data for the model, and the warning target type and coordinate location as output data to train the model. Based on real-time input data, use the trained model to identify the warning target type.

[0148] Backpropagation (BP) neural networks possess powerful nonlinear mapping capabilities. In intrusion target identification, signal features are often complex and exhibit nonlinear relationships. For example, the relationship between the features of microwave and low-frequency signals and target types is not linear. BP neural networks, through a multi-layered hierarchical structure of neurons, can learn the complex nonlinear mapping relationship between input data and the target type.

[0149] Simultaneously, the nonlinear activation function of the hidden layer can effectively capture the complex patterns between microwave echo signals and low-frequency features, thereby improving recognition accuracy. Furthermore, in intrusion detection, target changes are highly dynamic. Based on the excellent self-learning ability of the BP neural network, in a practical intrusion target recognition system, the model will be updated as new data enters. The BP neural network can continuously optimize and adjust its parameters through incremental learning, improving the model's adaptability as new data is added.

[0150] The established intrusion target identification model is based on a BP neural network, specifically including an input layer, a hidden layer, and an output layer:

[0151] Input layer:

[0152] ;

[0153] in, Input data into the model, The amount of extinguishing agent used. This is a fire assessment value. Fire level;

[0154] Hidden layer:

[0155] ;

[0156] ;

[0157] in, For the weighted input of the hidden layer, The output of the hidden layer, Here is the weight matrix of the hidden layer. For the bias term of the hidden layer, The activation function for the hidden layer;

[0158] Output layer:

[0159] ;

[0160] ;

[0161] in, For the weighted input of the output layer, The output should include information on the target type for the early warning. This is the weight matrix of the output layer. For the bias term of the output layer, This is the activation function for the output layer.

[0162] Firstly, for ground intrusions, the microwave and low-frequency feature sets are fused together using an intrusion target identification model for identification. Microwave signals can sensitively capture dynamic changes in ground targets, especially fast-moving targets. Detail coefficients help extract high-frequency information from the signal, reflecting the target's fine features. By combining detection location, microwave features can provide more accurate ground target identification, especially in complex environments or under obstructed conditions, effectively distinguishing different types of ground targets, such as personnel and vehicles. The advantage of forming a low-frequency feature set by combining vibration location and low-frequency domain features lies in the fact that low-frequency signals can effectively describe the interaction between the target and the ground, especially the stability and persistence characteristics of ground targets. Low-frequency domain features can reflect the long-term vibration patterns generated by the target, which is very effective for identifying stationary or slowly moving ground targets. When the low-frequency and microwave feature sets are used in combination, more comprehensive information is provided. Through the fusion analysis of signals from different frequency bands, the accuracy and robustness of ground target identification are further improved, ensuring the stability and reliability of the early warning system in complex environments.

[0163] For underground intrusions, low-frequency signals can reflect the vibration or motion characteristics of underground targets, especially when the target is not exposed on the surface. Low-frequency frequency domain characteristics can capture subtle vibrations caused by the target's activity or structural changes. These low-frequency characteristics are generally unaffected by surface interference and can effectively distinguish underground targets from other background noise, ensuring that the early warning system can accurately monitor the activity of underground targets in complex environments. Another major advantage of low-frequency frequency domain characteristics in identifying underground targets is their ability to identify the long-term stability of underground targets and their interaction with the surrounding environment. This is particularly suitable for areas with complex terrain or poor signal propagation conditions. By combining the analysis of vibration location and frequency domain characteristics, underground targets can be identified more efficiently and accurately, improving the performance and response speed of the early warning system.

[0164] Step 5: Adjust the threshold for judging the modal energy ratio based on soil moisture and surface water to remove the influence of water accumulation on the low-frequency signal offset. Adjust the noise threshold based on rainfall and air humidity to remove the influence of the environment on microwave echo characteristics.

[0165] Adjusting the modal energy ratio (MORR) threshold based on soil moisture and surface water levels can effectively eliminate the influence of water accumulation or wet soil on low-frequency signals. When soil moisture is too high or water depth is significant, the presence of water may cause attenuation or distortion of low-frequency signals, affecting the MORR determination and thus the accuracy of the detection results. Real-time monitoring and dynamic adjustment of these environmental factors ensure that the MORR more accurately reflects the vibration characteristics of underground targets, eliminating errors caused by water accumulation or wet soil, and improving the reliability and accuracy of signal processing. On the other hand, changes in rainfall and air humidity affect the propagation environment of echo signals. Especially in cases of high humidity or heavy rainfall, the noise level of the signal may increase, affecting the extraction of echo features and target identification. Adjusting the noise threshold based on changes in rainfall and air humidity can dynamically remove environmental noise interference to the echo signal, thereby improving the wavelet transform denoising effect, ensuring clearer target features in the microwave echo signal, reducing the influence of environmental factors, and improving the precision of signal processing and the accuracy of target identification.

[0166] In this embodiment, the environmental data includes soil moisture, water depth, rainfall, and air humidity;

[0167] The steps for adjusting the modal energy ratio based on soil moisture and surface water are as follows:

[0168] The dielectric constant of the soil is corrected based on soil moisture:

[0169] ;

[0170] in, The corrected dielectric constant is... To detect the basic dielectric constant of the soil in the region, For soil moisture, The dielectric constant of moisture-saturated soil. The dielectric constant of dry soil;

[0171] Determine the water accumulation attenuation coefficient based on the accumulated water:

[0172]

[0173] in, The water accumulation attenuation coefficient is... The basic water accumulation attenuation coefficient for soil type. This refers to the depth of the accumulated water. The center frequency of the low-frequency signal;

[0174] The modal energy ratio judgment threshold is adaptively adjusted based on the corrected dielectric constant and water accumulation attenuation coefficient. The calculation formula is as follows:

[0175] ;

[0176] in, The threshold for determining the corrected modal energy ratio is... The threshold for determining the modal energy ratio;

[0177] The specific steps for adjusting the noise threshold based on rainfall and air humidity are as follows:

[0178] ;

[0179] in, The corrected noise threshold. To set a basic noise threshold, This is the fundamental correction constant for the transmitted microwave frequency. For rainfall, This refers to air humidity.

[0180] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0181] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0182] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0183] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. An intrusion detection method based on low-frequency and microwave composite induction, characterized in that, The specific steps include: Step 1: Obtain soil type, warning target type, and environmental data for the area to be alarmed. Using the fixed variable method, obtain the low-frequency signal intensity of the vibration source in different soil types for different environmental data for the warning target type. Determine the distribution distance of the vibration sensor based on the low-frequency signal intensity and install it. Step 2: Acquire real-time low-frequency signals, perform modal decomposition on the low-frequency signals using the modal decomposition method, identify the above-ground and underground modal components based on the modal energy ratio, reconstruct the above-ground and underground modal components respectively, acquire the low-frequency reconstructed signals from the above-ground and underground, and perform time-frequency analysis on the reconstructed signals; Step 3: Determine the vibration location of the low-frequency signal source based on the time difference of the received low-frequency signal, and form a low-frequency feature set with the low-frequency frequency domain features. Based on the detection location of the microwave detection target, and after decomposing and denoising the echo signal, form a microwave feature set. Step 4: Establish an intrusion target identification model. Use low-frequency feature sets and microwave feature sets as input data for the model, and the warning target type and coordinate location as output data to train the model. Based on real-time input data, use the trained model to identify the warning target type. Step 5: Adjust the threshold for judging the modal energy ratio based on soil moisture and surface water to remove the influence of water accumulation on the low-frequency signal offset. Adjust the noise threshold based on rainfall and air humidity to remove the influence of the environment on microwave echo characteristics.

2. The intrusion detection method based on low-frequency and microwave composite induction according to claim 1, characterized in that: The soil types include: hardened soil, sandy soil, and loose soil; The types of targets for early warning include ground intrusion: personnel, vehicles, and animals; Underground intrusion: drilling, subsidence; The environmental data includes soil moisture, water depth, rainfall, and air humidity; The method for obtaining low-frequency signals of vibration sources of different types of warning targets in different environments and with different soil types using the fixed variable method is as follows: At the vibration source and at a distance of , respectively Vibration sensors are installed at the location. By sequentially changing one set of data types—the warning target type, environmental data, and soil type—while keeping the remaining two data types unchanged, the low-frequency signal strength of the warning target type is detected.

3. The intrusion detection method based on low-frequency and microwave composite induction according to claim 1, characterized in that: The method for determining the distribution distance of vibration sensors based on low-frequency signal intensity is as follows: The distribution distance is calculated by calculating the attenuation coefficient of the low-frequency signal intensity of each group of data, and based on the attenuation coefficient, the low-frequency signal of each group of early warning target types, and the detection threshold of the vibration sensor. ； ； in, The vibration attenuation coefficient is... The distance detected by the vibration sensor is the distance to the vibration source. Low-frequency signal strength at that location The intensity of the low-frequency signal at the vibration source detected by the vibration sensor. This represents the distribution distance of the vibration sensors; The installation method is as follows: Install sensors according to the distribution distance so that at least 3 vibration sensors can detect any position on the boundary of the area to be detected intrusion.

4. The intrusion detection method based on low-frequency and microwave composite induction according to claim 1, characterized in that: The specific steps for performing mode decomposition on low-frequency signals using the mode decomposition method are as follows: Set the number of modal components Broadband constraints initial value of center frequency For each modal component , =1,2,…, The angular frequency ω is updated, and the low-frequency signal decomposition is determined by the convergence condition, and the modal components are output. Iterative update of modal components: ； in, For the first The first modal component in the frequency domain Frequency domain value of the next iteration For the number of iterations, It is a low-frequency signal. Angular frequency, for Modal classification in The center frequency of the next iteration For the first The frequency domain values ​​of each modal component in the frequency domain. For the frequency domain representation of Lagrange multipliers, ; Update center frequency: ； in, for Modal classification in The center frequency of the next iteration; Determine whether to terminate the update based on the convergence criteria: ； in, To determine accuracy.

5. The intrusion detection method based on low-frequency and microwave composite induction according to claim 1, characterized in that: Based on the modal energy ratio, the modal components above ground and below ground are identified. The formula for calculating the modal energy ratio is as follows: ； in, The modal energy ratio, For the first The frequency representation of each modal component For frequency integral, These are the threshold values ​​for dividing the frequencies of above-ground and underground modal components, respectively. when At that time, it was determined that the modal components originated from the underground portion: when When determining that the modal components originate from the ground surface: in, The threshold for determining the modal energy ratio.

6. The intrusion detection method based on low-frequency and microwave composite induction according to claim 1, characterized in that: The method for recombining the above-ground and underground modal components to obtain the low-frequency recombined signals from the above-ground and underground is as follows: ； ； in, , These are the recombined signals from underground and above ground, respectively. To determine the number of modal components in the underground part, To determine the number of modal components in the aboveground portion, for The first moment One modal component.

7. The intrusion detection method based on low-frequency and microwave composite induction according to claim 1, characterized in that: The method for determining the vibration location of a low-frequency signal source based on the time difference of the received low-frequency signal is as follows: Based on the time difference in the reception of low-frequency signals by the vibration sensor, the vibration location of the low-frequency signal source can be determined by using the time difference and the propagation speed of the low-frequency signal. ； ； in, , , , These are vibration sources, the first one. , , The coordinates of each sensor For the propagation speed of low-frequency signals, For the first , The time difference between when each sensor detects a low-frequency signal For the first , The time difference between the detection of low-frequency signals by the sensors.

8. The intrusion detection method based on low-frequency and microwave composite induction according to claim 1, characterized in that: The specific steps for forming a microwave feature set after decomposing and denoising the echo signal are as follows: The detection location of the target is identified based on the echo time of the microwave detection target, and the echo signal is decomposed and denoised by wavelet transform to obtain detail coefficient features. A microwave feature set is formed based on the detection location and detail coefficient features. First, the echo signal is subjected to wavelet transform, which decomposes it into a series of approximation coefficients and detail coefficients of different frequencies; ； in, For echo signal, These are approximate coefficients. For detail coefficients, For scaling function, For wavelet functions, The time point for acquiring communication signals. These are the scale parameter and the displacement parameter, respectively. Noise signals in the echo signal can be removed by setting a noise threshold. ； in, The detail coefficient after noise reduction. The noise threshold is set.

9. The intrusion detection method based on low-frequency and microwave composite induction according to claim 1, characterized in that: The environmental data includes soil moisture, water depth, rainfall, and air humidity; The steps for adjusting the modal energy ratio based on soil moisture and surface water are as follows: The dielectric constant of the soil is corrected based on soil moisture: ； in, The corrected dielectric constant is... To detect the basic dielectric constant of the soil in the region, For soil moisture, The dielectric constant of moisture-saturated soil. The dielectric constant of dry soil; Determine the water accumulation attenuation coefficient based on the accumulated water: ； in, The water accumulation attenuation coefficient is... The basic water accumulation attenuation coefficient for soil type. This refers to the depth of the accumulated water. The center frequency of the low-frequency signal; The modal energy ratio judgment threshold is adaptively adjusted based on the corrected dielectric constant and water accumulation attenuation coefficient. The calculation formula is as follows: ； in, The threshold for determining the corrected modal energy ratio is... The threshold for determining the modal energy ratio; The specific steps for adjusting the noise threshold based on rainfall and air humidity are as follows: ； in, The corrected noise threshold. To set a basic noise threshold, This is the fundamental correction constant for the transmitted microwave frequency. For rainfall, This refers to air humidity.

Images