Complex environment resource intelligent matching method based on electromagnetic characteristics

Abstract

The invention discloses a complex environment resource intelligent matching method based on electromagnetic characteristics, and the method comprises the steps: S1, collecting electromagnetic signal data in a target environment, and monitoring and recording the change of a variable resource environment in real time, S2, processing the collected electromagnetic signal data based on an electromagnetic characteristic analysis model, extracting electromagnetic characteristic parameters, and carrying out the processing of the electromagnetic signal data based on an electromagnetic characteristic analysis model. S3, constructing a resource matching model, and combining the extracted electromagnetic characteristic parameters to generate an optimal resource allocation scheme in real time, and S4, according to the generated optimal resource allocation scheme, rapidly adjusting and matching resources, and improving real-time matching efficiency. According to the complex environment resource intelligent matching method based on the electromagnetic characteristics, the problem of low real-time matching efficiency caused by fast dynamic change in a variable resource environment in the prior art is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent resource matching and allocation, and particularly relates to an intelligent resource matching method for complex environments based on electromagnetic characteristics. Background Art

[0002] In modern complex environments, resource matching is a key issue and is widely applied in fields such as communication networks, logistics scheduling, and intelligent manufacturing. Traditional resource matching methods usually rely on static rules or limited heuristic algorithms and attempt to allocate resources according to predefined conditions. However, in practical applications, due to the dynamic changes in the resource environment, such as frequent updates of resource status, rapid fluctuations in environmental conditions, etc., the applicability of traditional methods is greatly limited.

[0003] In the prior art, most methods often cannot quickly adjust the allocation strategy when dealing with dynamic resource environments, which will lead to low resource matching efficiency. Especially in scenarios where resources change frequently, traditional methods need to recalculate and verify the matching scheme multiple times, increasing the calculation time and complexity. In addition, these methods usually lack the full utilization of the electromagnetic characteristics of resources and are difficult to accurately describe the physical properties and dynamic behaviors of resources, which further reduces the matching efficiency and accuracy. Technologies for resource description and analysis based on electromagnetic characteristics have achieved initial results, but in the prior art, this method is mainly used for static resource allocation in single scenarios or specific conditions. For the dynamic real-time matching problem in a changing resource environment, there are still many deficiencies. Traditional technologies are difficult to effectively integrate various electromagnetic characteristic information and adjust the matching strategy in real time to adapt to the rapid changes in the environment. To sum up, the prior art has disadvantages such as low real-time matching efficiency, poor adaptability, and insufficient utilization of the electromagnetic characteristics of resources when facing the problem of dynamic resource allocation in complex environments. Summary of the Invention

[0004] The purpose of the present invention is to provide an intelligent resource matching method for complex environments based on electromagnetic characteristics to solve the problem of low real-time matching efficiency in the prior art due to rapid dynamic changes in a changing resource environment.

[0005] To achieve the above purpose, the present invention provides the following technical solution: an intelligent resource matching method for complex environments based on electromagnetic characteristics, the method comprising: S1. Collect electromagnetic signal data in the target environment, and monitor and record the changes in the changing resource environment in real time; S2. Process the collected electromagnetic signal data based on an electromagnetic characteristic analysis model to extract electromagnetic characteristic parameters; S3. Construct a resource matching model, combine the extracted electromagnetic characteristic parameters, and generate an optimal resource allocation scheme in real time, including modeling the electromagnetic characteristic parameters and calculating the costs of different allocation schemes. The specific formula is: ; Among them, F(x) represents the total allocation time required for the current resource allocation plan, t0 represents the time required to complete the basic allocation under the condition that the resources are not restricted, x represents the amount of resources already allocated in the current environment, c represents the maximum allocable amount of resources in a specific area or system, k represents the non-linear adjustment factor in the resource allocation process, and n represents the degree of non-linear growth of the resource allocation cost with the amount of resource usage; S4. According to the generated optimal resource allocation plan, quickly adjust and match resources to improve the real-time matching efficiency.

[0006] Preferably, the said S1 includes: Arrange signal receivers in the target environment, record the electromagnetic signal intensities at different points, calculate the attenuation parameters according to the signal attenuation formula, and estimate the signal propagation range. The specific formula is: ; Among them, A represents the electromagnetic signal intensity at a specific point in the target environment, A0 represents the initial intensity of the signal at the source point, α represents the attenuation rate of the electromagnetic signal affected by environmental factors during the propagation process, d represents the propagation path length of the signal from the source point to the target point, and e represents the base of the natural logarithm; Record the signal changes in real time, construct a regional signal intensity distribution map, and dynamically monitor the environmental changes.

[0007] Preferably, the said S2 includes: Record the time series of the electromagnetic signal changes, fit the changes of the electromagnetic signal, and extract parameters. The specific formula is: ; Among them, I represents the input intensity of the electromagnetic signal, C m represents the response ability to the changes of the electromagnetic signal, V represents the instantaneous intensity of the signal or the current state of the electromagnetic signal, V rest represents the electromagnetic signal background value in the target environment when there is no signal input, g represents the transmission efficiency of the electromagnetic signal, t represents time, and m represents the response attribute of the environment to the electromagnetic signal.

[0008] Preferably, the said S4 includes: Calculate the utilization amount of the current resources, calculate the resource allocation efficiency, and guide the adjustment of the resource allocation plan. The specific formula is: ; Among them, E represents the resource allocation efficiency, P represents the maximum allocation efficiency that the resources may reach under the optimal conditions, L represents the current actual amount of resources used, and K represents the saturation point in the resource allocation.

[0009] Preferably, S2 includes obtaining the time series data of the electromagnetic signal, extracting the amplitude values of multiple frequency bands based on spectrum analysis, calculating the power spectral density of different frequency bands, performing feature fusion according to the preset frequency band weights, and generating the final electromagnetic feature parameters.

[0010] Preferably, the extracting the amplitude values of multiple frequency bands based on spectrum analysis includes converting the time series data into a frequency domain signal by using Fourier transform, extracting the amplitude values within multiple predefined frequency band ranges in the frequency domain signal, and performing normalization processing based on the amplitude values of the frequency components to eliminate the amplitude differences. Set the frequency range [f1, f2]. If the frequency f falls within the specified frequency band range, i.e., f ∈ [f1, f2], then extract the amplitude value of this frequency band B(f) = |F(f)|, where F(f) represents the spectral amplitude value at frequency f, |F(f)| represents the absolute value of the spectral amplitude, and f1 and f2 respectively represent the lower and upper limits of the frequency band.

[0011] Preferably, the calculating the power spectral density of different frequency bands includes squaring the amplitude values of each extracted frequency band to obtain the corresponding power spectral values, integrating the power spectral values of each frequency band to obtain the total power of each frequency band, calculating the power spectral density of each frequency band based on the total power, setting the power threshold T1. If the total power Q of the frequency band is greater than the threshold T1, then this frequency band is considered an important frequency band, otherwise it is considered an unimportant frequency band. Here, T1 is the preset threshold for the total power of the frequency band, and Q is the total power of the frequency band.

[0012] Preferably, the obtaining the time series data of the electromagnetic signal includes collecting the electromagnetic signal data from multiple sensors, preprocessing the collected data, removing noise and interference signals, arranging the preprocessed data in chronological order into a time series, setting the time series length threshold T2. If the length H of the time series is greater than the set threshold T2, then use the sliding window technique to extract a subsequence of a fixed length. Here, H is the length of the time series, and T2 is the preset threshold for the length of the time series.

[0013] Preferably, the preprocessing the collected data includes performing digital filtering to remove high-frequency noise, performing signal denoising through wavelet transform to further improve the signal quality, and performing normalization processing to make the data distribution uniform. Set the threshold T3 for the amplitude value of the filtered signal. If the amplitude value S(f) of the filtered signal is less than the threshold T3, then regard this amplitude value as an interference signal and remove it. Here, S(f) represents the amplitude value of the filtered signal, and T3 is the preset threshold for the amplitude value of the filtered signal.

[0014] Preferably, S2 further includes calculating the time-frequency joint characteristics of the electromagnetic signal and extracting the time-domain and frequency-domain characteristic parameters of the signal through short-time Fourier transform.

[0015] As can be seen from the above technical solutions, the present invention has the following beneficial effects: The intelligent matching method for complex environment resources based on electromagnetic characteristics collects electromagnetic signal data in the target environment, monitors and records the changes in the variable resource environment in real time, processes the collected electromagnetic signal data based on the electromagnetic characteristic analysis model, extracts electromagnetic characteristic parameters, constructs a resource matching model, combines the extracted electromagnetic characteristic parameters, generates an optimal resource allocation scheme in real time, adjusts and matches resources quickly according to the generated optimal resource allocation scheme, improves the real-time matching efficiency, can quickly respond to the changes in the resource state, effectively shortens the allocation decision time, greatly improves the real-time performance and efficiency of resource matching, can maintain a high matching efficiency and allocation stability, significantly improves the adaptability, can not only accurately describe the resource characteristics, but also use these characteristics to optimize the allocation scheme, thereby improving the accuracy and precision of matching, greatly reducing the computational complexity, and at the same time ensuring that the resource allocation is always in an efficient state, reducing the computational cost, and solving the problem of low real-time matching efficiency in the prior art due to the fast dynamic changes in the variable resource environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a flowchart of the method of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0018] As Figure 1 shown, the present invention provides a technical solution: an intelligent matching method for complex environment resources based on electromagnetic characteristics, the method comprising: S1. Collect electromagnetic signal data in the target environment, and monitor and record the changes in the variable resource environment in real time; S2. Process the collected electromagnetic signal data based on the electromagnetic characteristic analysis model, and extract electromagnetic characteristic parameters; S3. Construct a resource matching model, combine the extracted electromagnetic characteristic parameters, and generate an optimal resource allocation scheme in real time, including modeling the electromagnetic characteristic parameters and calculating the costs of different allocation schemes. The specific formula is: ; Among them, F(x) represents the total allocation time required for the current resource allocation plan, t0 represents the time required to complete the basic allocation under the condition that resources are not restricted, x represents the amount of resources already allocated in the current environment, c represents the maximum allocable amount of resources in a specific area or system, k represents the non-linear adjustment factor in the resource allocation process, and n represents the degree of non-linear growth of the resource allocation cost with the resource usage amount; S4. According to the generated optimal resource allocation plan, quickly adjust and match resources to improve the real-time matching efficiency.

[0019] The working principle of the above method is that by collecting electromagnetic signal data in the target environment, the resource changes in the complex environment are monitored in real time. Using the electromagnetic feature analysis model, the key feature parameters in the electromagnetic signal can be accurately extracted, providing data support for subsequent resource matching. The resource matching model calculates the costs of different allocation plans according to the extracted electromagnetic feature parameters, determines the total allocation time through the formula F(x), and thus obtains the optimal allocation plan. Finally, the generated optimal resource allocation plan guides the resource adjustment and matching, ensuring that the resource allocation is efficient and dynamically adapts to environmental changes. Using the electromagnetic feature parameters to optimize the allocation plan in real time significantly shortens the resource matching time. By monitoring the electromagnetic signal data in real time, the dynamic changes in the resource environment are quickly captured, optimizing the matching accuracy. Through the resource matching model and the non-linear allocation formula, the utilization rate of resources is optimized and the allocation conflicts are reduced, realizing the accurate modeling of the resource allocation process and improving the allocation reliability.

[0020] S1 includes arranging signal receivers in the target environment, recording the electromagnetic signal intensities at different points, calculating the attenuation parameters according to the signal attenuation formula, and estimating the signal propagation range. The specific formula is: ; Among them, A represents the electromagnetic signal intensity at a specific point in the target environment, A0 represents the initial intensity of the signal at the source point, α represents the attenuation rate of the electromagnetic signal affected by environmental factors during the propagation process, d represents the propagation path length of the signal from the source point to the target point, and e represents the base of the natural logarithm; Record the signal changes in real time, construct a regional signal intensity distribution map, and dynamically monitor the environmental changes.

[0021] In this embodiment, by arranging signal receivers in the target environment, the electromagnetic signal intensity data at each point in the environment is collected in real time. According to the collected data, combined with the formula Model the attenuation law of the signal, where the attenuation rate a and the propagation path length d are the core parameters affecting the signal strength. By analyzing the change of signal strength with position, generate a regional signal strength distribution map to dynamically monitor the resource changes in the target environment and provide basic data support for subsequent intelligent resource matching. Record and calculate the change of electromagnetic signal strength in real time, accurately describe the propagation range and distribution of signals in the environment through the signal attenuation model, provide accurate signal strength data for the subsequent steps, improve the reliability and intelligence level of the resource matching model, quickly capture environmental changes through the regional signal strength distribution map, adapt to the changing resource environment conditions, avoid relying on complex measurement equipment, and reduce the system implementation cost.

[0022] S2 includes recording the time series of electromagnetic signal changes, fitting the changes of electromagnetic signals, and extracting parameters. The specific formula is: ; Among them, I represents the input intensity of the electromagnetic signal, C m represents the response ability to the change of the electromagnetic signal, V represents the instantaneous intensity of the signal or the current state of the electromagnetic signal, V rest represents the electromagnetic signal background value in the target environment when there is no signal input, g represents the transmission efficiency of the electromagnetic signal, t represents time, and m represents the response attribute of the environment to the electromagnetic signal.

[0023] In this embodiment, by recording the time series of electromagnetic signal changes, the dynamic change situation of the signal in the target environment can be obtained. Using the formula to model the input intensity of the electromagnetic signal, where C m represents the response characteristic of the environment to the signal change, g represents the transmission efficiency of the signal, and V and V rest are used to describe the current state of the signal and the background characteristics. By fitting the change trend of the electromagnetic signal over time, extract the characteristic parameters of the signal under different environmental conditions, and provide key input data for the resource matching model. This method can dynamically track signal changes and adapt to complex and changeable environmental characteristics. By recording the time series and fitting the signal change curve, more signal characteristics can be extracted, improving the accuracy and integrity of signal analysis. This method combines the time change characteristics and environmental response characteristics of electromagnetic signals, can better adapt to dynamic complex environments, and provides high-quality data support for the optimization of subsequent resource allocation schemes, suitable for signal analysis requirements in various scenarios, especially scenarios involving complex electromagnetic environments.

[0024] S4 includes calculating the utilization amount of the current resources, calculating the resource allocation efficiency, and guiding the adjustment of the resource allocation scheme. The specific formula is: ; Among them, E represents the resource allocation efficiency, P represents the maximum allocation efficiency that resources can reach under optimal conditions, L represents the current actual amount of resources used, and K represents the saturation point in resource allocation.

[0025] In this embodiment, based on the formula the allocation efficiency of resources is modeled and calculated. Among them, the maximum allocation efficiency P of resources describes the potential of resource utilization in an ideal environment, while the actual amount of resources used L and the allocation saturation point K reflect the resource utilization status in the current environment. By calculating the allocation efficiency E, the existing resource allocation scheme can be optimized in real time, and the overall efficiency can be improved by adjusting the allocation ratio. At the same time, this formula provides a dynamic regulation mechanism for balancing resource demand and environmental constraints, avoiding overload and waste in resource allocation. By accurately calculating the resource allocation efficiency, guiding the adjustment of the resource allocation scheme, maximizing the resource utilization rate, dynamically optimizing the allocation ratio according to the changes in the resource utilization status and the allocation saturation point, adapting to different environmental conditions, identifying resource overload or idle situations in a timely manner by calculating the allocation efficiency, avoiding unnecessary resource waste, and realizing intelligent management of complex resource environments through the allocation efficiency model, reducing manual intervention.

[0026] S2 includes obtaining the time series data of electromagnetic signals, extracting the amplitude values of multiple frequency bands based on spectrum analysis, calculating the power spectral density of different frequency bands, and performing feature fusion according to the preset frequency band weights to generate the final electromagnetic feature parameters.

[0027] In this embodiment, by obtaining the time series data of electromagnetic signals in the target environment, the changes of the signals can be recorded over time. Using the spectrum analysis method, the time-domain signal is converted into a frequency-domain signal, the amplitude values of different frequency bands are extracted, and the power spectral density of these frequency bands is calculated to describe the energy distribution of the electromagnetic signals. Subsequently, according to the preset frequency band weights set for the application scenario, the features of different frequency bands are weighted and fused, so as to generate important parameters that can comprehensively reflect the electromagnetic characteristics. This electromagnetic feature parameter provides an accurate data basis for subsequent resource matching. By extracting the amplitude values and power spectral density of the signals, the characteristics of the electromagnetic signals can be captured more accurately. Combining the feature fusion method of frequency band weights can flexibly adapt to the requirements under different application scenarios, optimize the effect of feature parameter extraction. Using the combination of spectrum analysis and weighted fusion can quickly process large-scale time series data, meet the real-time processing requirements, and the generated electromagnetic feature parameters can be applied to the resource matching requirements in a variety of electromagnetic environments, improving the generality of the method.

[0028] Extracting the amplitude values of multiple frequency bands based on spectral analysis includes using Fourier transform to convert time series data into a frequency domain signal, extracting the amplitude values within multiple predefined frequency band ranges from the frequency domain signal, and performing normalization processing on the amplitude values of the frequency components to eliminate amplitude differences. Set the frequency range [f1, f2]. If the frequency f falls within the specified frequency band range, i.e., f ∈ [f1, f2], then extract the amplitude value of this frequency band B(f) = |F(f)|, where F(f) represents the spectral amplitude value at frequency f, and |F(f)| represents the absolute value of the spectral amplitude. f1 and f2 represent the lower and upper limits of the frequency band respectively.

[0029] In this embodiment, first perform Fourier transform on the time series data to convert it from the time domain to the frequency domain and generate a frequency domain signal. Subsequently, in the frequency domain signal, extract the amplitude value B(f) of the frequency band according to the preset frequency band range [f1, f2]. Specifically, for each frequency f, if f is within the predefined frequency band range, i.e., satisfies f ∈ [f1, f2], then extract the amplitude value |F(f)| of this frequency from the spectrum of the frequency domain signal. To eliminate the influence of amplitude differences on subsequent processing, the extracted amplitude values will be normalized, for example, by maximum value normalization (B′(f) = B(f) / B max or mean normalization (B′(f) = (B(f)−μ) / σ. This method enables the amplitude values of different frequency bands to be compared and fused on a unified dimension, ensuring the accuracy and consistency of feature extraction. Using Fourier transform to accurately separate the frequency components in the time series signal ensures that the extracted amplitude values can accurately reflect the characteristics of different frequency bands. The normalization processing eliminates the dimensional differences in amplitude, making the features of different frequency bands more comparable and conducive to subsequent feature fusion. By setting different frequency band ranges [f1, f2], the corresponding frequency band feature information can be extracted for specific application scenarios, improving the flexibility of the method. By taking the absolute value processing of the frequency components (|F(f)|), the interference of phase information on amplitude feature extraction is avoided, further improving the stability and robustness of the feature parameters.

[0030] Calculating the power spectral density of different frequency bands includes squaring the amplitude values of each extracted frequency band to obtain the corresponding power spectral values, integrating the power spectral values of each frequency band to obtain the total power of each frequency band, and calculating the power spectral density of each frequency band based on the total power. Set the power threshold T1. If the total power Q of the frequency band is greater than the threshold T1, then this frequency band is considered an important frequency band, otherwise it is considered an unimportant frequency band. Here, T1 is the preset threshold of the total power of the frequency band, and Q is the total power of the frequency band.

[0031] In this embodiment, first square the amplitude value B(f) extracted from each frequency band and calculate the corresponding power spectral value P(f)=[B(f)] 2, to reflect the energy distribution of the signals within the frequency band. Subsequently, the power spectral values P(f) within each frequency band are integrated to calculate the total power of the frequency band , where [f1, f2] represents the range of the frequency band.

[0032] Next, based on the total power value Q, the power spectral density of this frequency band is calculated (for example, normalized by the frequency band width, PSD = Q / (f2 - f1)). Finally, by setting a power threshold T1, it is determined whether the frequency band is an important frequency band: when Q > T1, this frequency band is marked as an important frequency band; otherwise, it is marked as an unimportant frequency band. This process can effectively distinguish the frequency bands that play a key role in the characteristics of the electromagnetic environment and provide a basis for subsequent feature fusion. By calculating the total power of the frequency band through square summation and integration, the energy distribution characteristics of the signal in different frequency bands are comprehensively reflected. Based on the total power and the preset threshold T1, the frequency bands that make important contributions to the description of the environmental characteristics can be quickly screened out, improving the efficiency of electromagnetic feature extraction. By calculating the power spectral density, it can adapt to environments with different frequency band ranges and signal intensities, improving the universality of feature extraction. By using a simple discrimination method with a power threshold, the processing of unimportant frequency bands can be reduced, optimizing the utilization of computing resources.

[0033] Obtaining the time series data of electromagnetic signals includes collecting electromagnetic signal data from multiple sensors, preprocessing the collected data to remove noise and interference signals, arranging the preprocessed data in chronological order into a time series, setting a time series length threshold T2. If the length H of the time series is greater than the set threshold T2, the sliding window technique is used to extract subsequences of a fixed length, where H is the length of the time series and T2 is the preset threshold of the time series length.

[0034] In this embodiment, first, electromagnetic signals are collected by multiple sensors to obtain multi-source signal data, and preprocessing is performed on it, including removing noise, interference signals, and outliers to ensure the accuracy and reliability of the data. Subsequently, the processed signal data is arranged in chronological order to form a complete time-series signal. When the length H of the time series exceeds the preset threshold T2, to avoid the impact of an overly long time series on subsequent analysis, a sliding window technique is introduced. The sliding window slices the time series with a fixed window length, extracting a subsequence of a fixed length each time, and the step size of the window sliding can be set according to requirements. The sliding window method can not only retain the dynamic change characteristics in the time series but also ensure the consistency and processing efficiency of the segmented data. Through the signal collection of multiple sensors, the electromagnetic characteristics of the target environment can be comprehensively reflected, improving the comprehensiveness and accuracy of the data. The preprocessing steps can significantly reduce the impact of noise and interference signals, providing high-quality data for subsequent analysis. The sliding window technique effectively avoids the computational complexity problem caused by a long time series, making subsequent processing more efficient. The subsequences extracted by the sliding window can retain the time dynamic change characteristics of the signal, providing complete data for subsequent spectrum analysis and feature extraction. The time series length threshold T2 and the size of the sliding window can both be adjusted according to the specific scenario to meet different application requirements.

[0035] The preprocessing of the collected data includes performing digital filtering to remove high-frequency noise, signal denoising through wavelet transform to further improve the signal quality, standardization processing to make the data distribution uniform, setting a threshold T3 for the amplitude of the filtered signal. If the amplitude S(f) of the filtered signal is less than the threshold T3, then this amplitude is regarded as an interference signal and removed, where S(f) represents the amplitude of the filtered signal and T3 is the preset threshold for the amplitude of the filtered signal.

[0036] In this embodiment, the data preprocessing is mainly divided into the following steps: Use a low-pass filter to remove high-frequency noise interference to retain the main characteristic frequency band of the signal; Adopt the wavelet transform method to decompose the signal into wavelet components of different scales, denoise the high-frequency components containing noise, and at the same time retain the effective characteristics of the signal; Standardize the denoised signal to adjust the signal values to a unified scale to ensure the uniformity and consistency of the data distribution; Set the threshold T3 of the amplitude of the filtered signal. For signals with an amplitude S(f) less than T3, they are considered interference signals and are excluded. Through this process, the quality of the data can be significantly improved, and low-amplitude interference signals that are invalid for subsequent analysis can be removed. Combining digital filtering and wavelet transform technologies can effectively remove high-frequency noise and interference signals, improve the quality and reliability of the signal. Through standardization processing, the distribution of the signal data becomes more uniform, providing high-quality input data for subsequent analysis. Through filtering and threshold screening, important signal characteristics with high amplitudes are retained, and low-amplitude interference signals are excluded to ensure the effectiveness of feature extraction. The preprocessing stage simplifies the signal data to high-quality parts, effectively reducing the complexity of subsequent calculations.

[0037] S2 also includes calculating the time-frequency joint characteristics of the electromagnetic signal and extracting the time-domain and frequency-domain characteristic parameters of the signal through short-time Fourier transform. In this embodiment, in order to comprehensively characterize the dynamic characteristics of the electromagnetic signal, short-time Fourier transform is used to calculate the time-frequency joint characteristics of the signal. The specific process is as follows: Divide the electromagnetic signal into several short-time segments. Assume that the signal is a stationary signal within each time segment, and then perform Fourier transform on each segment to extract its frequency components. Through the sliding window technique, the time-domain signal is gradually transformed into a time-frequency diagram with the common distribution of time and frequency. Extract the characteristic parameters reflecting the time variation of the signal from the time-frequency diagram, such as short-time energy, peak amplitude, signal duration, etc. Extract the characteristic parameters reflecting the frequency distribution of the signal from the time-frequency diagram, such as main frequency, bandwidth, power spectral density, etc. Integrate the time-domain and frequency-domain characteristic parameters to generate comprehensive time-frequency characteristic parameters for comprehensively describing the dynamic characteristics of the electromagnetic signal. This method can capture both the time variation information and the frequency distribution information of the signal, making up for the deficiencies of using only time-domain analysis or frequency-domain analysis alone, and providing more comprehensive data support for subsequent resource intelligent matching. Through time-frequency joint analysis, the time and frequency characteristic parameters of the signal can be extracted simultaneously, providing richer information than single time-domain or frequency-domain analysis. Utilizing the time-frequency localization characteristic of short-time Fourier transform, the detailed characteristics of the signal changing with time can be captured, meeting the requirements of signal feature extraction in a dynamic and complex environment. The time-frequency joint characteristics can provide more accurate input parameters for the resource matching model, improving the accuracy and robustness of intelligent matching. This method is applicable to various analysis scenarios of non-stationary signals, especially the processing of complex electromagnetic signals in a dynamic environment.

[0038] Although embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent matching method for complex environment resources based on electromagnetic characteristics, characterized in that The method includes: S1. Collect electromagnetic signal data in the target environment, and monitor and record the changes in the variable resource environment in real time; S2. Process the collected electromagnetic signal data based on the electromagnetic feature analysis model to extract electromagnetic feature parameters; S3. Build a resource matching model, and in combination with the extracted electromagnetic feature parameters, generate an optimal resource allocation plan in real time, including modeling the electromagnetic feature parameters and calculating the costs of different allocation plans. The specific formula is as follows: ; Among them, F(x) represents the total allocation time required for the current resource allocation scheme, t0 represents the time required to complete the basic allocation under the condition that the resources are not restricted, x represents the amount of resources already allocated in the current environment, c represents the maximum allocable amount of resources in a specific area or system, k represents the non-linear adjustment factor in the resource allocation process, and n represents the degree of non-linear growth of the resource allocation cost with the amount of resource use; S4. According to the generated optimal resource allocation scheme, quickly adjust and match resources to improve the real-time matching efficiency.

2. The intelligent matching method for complex environment resources based on electromagnetic characteristics according to claim 1, characterized in that: The S1 includes: Deploy signal receivers in the target environment, record the electromagnetic signal strengths at different points, calculate the attenuation parameters according to the signal attenuation formula, and estimate the signal propagation range. The specific formula is as follows: ; Among them, A represents the electromagnetic signal intensity at a specific point in the target environment, A0 represents the initial intensity of the signal at the source point, α represents the attenuation rate of the electromagnetic signal affected by environmental factors during the propagation process, d represents the propagation path length of the signal from the source point to the target point, and e represents the base of the natural logarithm; Record the signal changes in real time, construct a regional signal intensity distribution map, and dynamically monitor the environmental changes.

3. An intelligent matching method for complex environment resources based on electromagnetic characteristics according to claim 1, characterized in that: The S2 includes: Record the time series of the electromagnetic signal change, fit the change of the electromagnetic signal, and extract parameters. The specific formula is as follows: ; Where, I represents the input intensity of the electromagnetic signal, C m represents the response ability to the change of the electromagnetic signal, V represents the instantaneous intensity of the signal or the current state of the electromagnetic signal, V rest represents the electromagnetic signal background value in the target environment when there is no signal input, g represents the transmission efficiency of the electromagnetic signal, t represents time, and m represents the response attribute of the environment to the electromagnetic signal.

4. A method for intelligent matching of complex environment resources based on electromagnetic characteristics according to claim 1, characterized in that: The S4 includes: Calculate the utilization of the current resources, calculate the resource allocation efficiency, and guide the adjustment of the resource allocation plan. The specific formula is as follows: ; Among them, E represents the resource allocation efficiency, P represents the maximum allocation efficiency that the resources can reach under the optimal conditions, L represents the amount of resources actually used currently, and K represents the saturation point in the resource allocation.

5. The intelligent matching method for complex environment resources based on electromagnetic characteristics according to claim 1, wherein: The S2 includes obtaining the time series data of the electromagnetic signal, extracting the amplitude values of multiple frequency bands based on the spectrum analysis, calculating the power spectral density of different frequency bands, and performing feature fusion according to the preset frequency band weights to generate the final electromagnetic feature parameters.

6. The intelligent matching method for complex environment resources based on electromagnetic characteristics according to claim 5, wherein: The extracting the amplitude values of multiple frequency bands based on the spectrum analysis includes using the Fourier transform to convert the time series data into a frequency domain signal, extracting the amplitude values within multiple predefined frequency band ranges from the frequency domain signal, performing normalization processing based on the amplitude values of the frequency components to eliminate the amplitude differences, setting the frequency range [f1, f2], if the frequency f falls within the specified frequency band range, that is, f ∈ [f1, f2], then extract the amplitude value of this frequency band B(f)=|F(f)|, where F(f) represents the spectrum amplitude value at the frequency f, |F(f)| represents the absolute value of the spectrum amplitude, and f1 and f2 respectively represent the lower and upper limits of the frequency band.

7. An intelligent matching method for complex environment resources based on electromagnetic characteristics according to claim 5, characterized in that: The calculating the power spectral density of different frequency bands includes squaring the amplitude values of each extracted frequency band to obtain the corresponding power spectral values, integrating the power spectral values of each frequency band to obtain the total power of each frequency band, calculating the power spectral density of each frequency band based on the total power, setting the power threshold T1, if the total power Q of the frequency band is greater than the threshold T1, then this frequency band is considered an important frequency band, otherwise this frequency band is considered an unimportant frequency band, where T1 is the preset threshold of the total power of the frequency band, and Q is the total power of the frequency band.

8. An intelligent matching method for complex environment resources based on electromagnetic characteristics according to claim 5, characterized in that: The acquisition of the time series data of the electromagnetic signal includes collecting the electromagnetic signal data from multiple sensors, preprocessing the collected data to remove noise and interference signals, arranging the preprocessed data in chronological order into a time series, setting a time series length threshold T2. If the length H of the time series is greater than the set threshold T2, the sliding window technique is used to extract a subsequence of a fixed length, where H is the length of the time series and T2 is the preset threshold of the length of the time series.

9. The intelligent matching method for complex environment resources based on electromagnetic characteristics according to claim 8, characterized in that: The preprocessing of the collected data includes digital filtering to remove high-frequency noise, signal denoising through wavelet transform to further improve the signal quality, and normalization processing to make the data distribution uniform. A threshold T3 of the amplitude of the filtered signal is set. If the amplitude S(f) of the filtered signal is less than the threshold T3, the amplitude is regarded as an interference signal and removed, where S(f) represents the amplitude of the filtered signal and T3 is the preset threshold of the amplitude of the filtered signal.

10. The intelligent matching method for complex environment resources based on electromagnetic characteristics according to claim 1, characterized in that: The S2 also includes calculating the time-frequency joint characteristics of the electromagnetic signal and extracting the time-domain and frequency-domain characteristic parameters of the signal through short-time Fourier transform.

Images