Spectrum analysis device and method for radio monitoring system, and storage medium

Abstract

The invention relates to the technical field of radio monitoring, in particular to a spectrum analysis device and method for a radio monitoring system and a storage medium, and the method comprises the steps: obtaining a corresponding environment noise level in a monitoring process, adjusting a sensitivity threshold value corresponding to signal detection based on the environment noise level, receiving a corresponding to-be-analyzed radio frequency signal, and preprocessing the received to-be-analyzed radio frequency signal based on an adaptive filter; calling a pre-trained spectrum analysis model, inputting the preprocessed to-be-analyzed radio frequency signal into the spectrum analysis model, and obtaining a corresponding spectrum classification result; and calling frequency information and time sequence data corresponding to a radio frequency signal to be analyzed, obtaining corresponding spectrum analysis data based on the spectrum classification result, the frequency information and the time sequence data, and displaying key parameters in the spectrum analysis data through a graphical interface. According to the invention, the effectiveness of signal detection and identification can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of radio monitoring, and in particular to a spectrum analysis device, method and storage medium for a radio monitoring system. Background Art

[0002] With the rapid development of wireless communication technology, spectrum resources are becoming increasingly scarce. In order to effectively manage and utilize limited spectrum resources, radio monitoring systems play a vital role. Traditional radio monitoring systems usually rely on static or semi-static methods to perform spectrum analysis. Although this method meets basic needs to a certain extent, it is unable to cope with dynamically changing signal environments.

[0003] Especially in high-density and complex electromagnetic environments, existing methods find it difficult to effectively detect and identify low-power signals, and also have shortcomings in real-time performance.

[0004] From the above, it can be seen that how to improve the effectiveness of signal detection and recognition remains to be solved. Summary of the invention

[0005] In order to improve the effectiveness of signal detection and identification, the present application provides a spectrum analysis device, method and storage medium for a radio monitoring system.

[0006] In a first aspect, the present application provides a spectrum analysis method for a radio monitoring system, which adopts the following technical solution: A spectrum analysis method for a radio monitoring system, comprising: during the monitoring process, obtaining a corresponding environmental noise level, adjusting a sensitivity threshold corresponding to signal detection based on the environmental noise level, receiving a corresponding radio frequency signal to be analyzed, and preprocessing the received radio frequency signal to be analyzed based on an adaptive filter; calling a pre-trained spectrum analysis model, inputting the preprocessed radio frequency signal to be analyzed into the spectrum analysis model, and obtaining a corresponding spectrum classification result; calling frequency information and time series data corresponding to the radio frequency signal to be analyzed, obtaining corresponding spectrum analysis data based on the spectrum classification result, the frequency information and the time series data, and displaying key parameters in the spectrum analysis data through a graphical interface, wherein the key parameters include frequency occupancy and a signal strength distribution diagram.

[0007] By adopting the above technical solution, the environmental noise level is dynamically obtained during the monitoring process, and the sensitivity threshold of signal detection is adjusted based on this, thereby effectively improving the accuracy of signal detection in different noise environments. The received RF signal is preprocessed by an adaptive filter to remove noise and enhance useful signals, and then the preprocessed signal is input into a pre-trained spectrum analysis model to obtain spectrum classification results. By combining frequency information with time series data, detailed spectrum analysis data is obtained, and key parameters such as frequency occupancy and signal strength distribution diagram are intuitively displayed through a graphical interface, thereby achieving an improvement in the effectiveness of signal detection and identification; in complex and changing electromagnetic environments, RF signals can be more accurately identified and classified, thereby improving the reliability and accuracy of signal detection and identification.

[0008] Optionally, in the process of obtaining the corresponding spectrum classification results, the method also includes: if the spectrum analysis model identifies an abnormal signal that deviates from the normal mode; calculate the deviation data between the abnormal signal and the normal mode signal; classify the abnormal signal based on the deviation data, obtain the corresponding target abnormality type, and generate a corresponding abnormality report.

[0009] By adopting the above technical solution, by calculating the deviation data between these abnormal signals and normal mode signals and obtaining the corresponding target abnormal type, and then generating an abnormal report, it is helpful to timely discover and classify abnormal signals, thereby improving the monitoring accuracy and reliability of the system and ensuring the safety and standardization of spectrum use.

[0010] Optionally, after obtaining the corresponding target abnormality type, the method also includes: detecting the target abnormality type to determine whether the target abnormality type is an illegal or unauthorized frequency band; if so, performing an interference source clearing operation on the target abnormality type, the interference source clearing operation including using interference suppression technology, frequency band switching, and signal blocking; obtaining the corresponding actual processing effect according to the interference source clearing operation, and optimizing the spectrum clearing strategy based on the actual processing effect.

[0011] By adopting the above technical solution, when the exception type is an illegal or unauthorized frequency band, the interference source clearing operation is performed on these target exception types. According to the actual processing effect, the spectrum clearing strategy can be further optimized, and the illegal or unauthorized frequency band can be effectively identified and processed, thereby improving the security and efficiency of spectrum management.

[0012] Optionally, the method also includes: obtaining current location information of the radio frequency signal to be analyzed, and determining the corresponding specific area based on the current location information; collecting historical spectrum usage data within the specific area, the historical spectrum usage data including frequency occupancy, signal strength distribution, interference conditions, and spectrum utilization; generating optimal spectrum usage recommendations within the specific area based on the current location information and historical spectrum usage data, the optimal spectrum usage recommendations including frequency band selection, power control, time slot allocation, and multi-band combination.

[0013] By adopting the above technical solution, by obtaining the current location information of the RF signal to be analyzed, determining the corresponding specific area, and collecting the historical spectrum usage data in the area, and then generating the best spectrum usage recommendations in the specific area based on this information, it can help users optimize the use of spectrum resources and improve communication efficiency and quality.

[0014] Optionally, after obtaining the radio frequency signal to be analyzed, the method also includes: constructing a virtual spectrum simulation environment, wherein the spectrum simulation environment can simulate the real-world radio spectrum usage; generating a plurality of corresponding spectrum simulation model options for the spectrum simulation environment, and sending all of the spectrum simulation model options to the user for selection, wherein the spectrum simulation environment allows the user to configure model parameters according to specific needs; obtaining corresponding feedback data based on the user's options, determining the corresponding target spectrum simulation model based on the feedback data, controlling the target spectrum simulation model to run simulation in real time, and displaying the simulation results through a graphical interface.

[0015] By adopting the above technical solutions and building a virtual spectrum simulation environment, users can select and configure appropriate spectrum simulation models, run simulations in real time and display results, helping users test and verify spectrum usage strategies before actual deployment, thereby improving the accuracy and reliability of spectrum management.

[0016] Optionally, the method also includes: collecting different types of multimodal analysis data from pre-installed sensors, the analysis data including radio frequency signals, video signals, and audio signals; preprocessing the collected multimodal analysis data; extracting corresponding key features from the preprocessed multimodal analysis data, and fusing the key features of different modalities, wherein the key features include radio frequency features, video features, and audio features; and inputting the fused key features into a spectrum analysis model to obtain corresponding spectrum analysis results.

[0017] By adopting the above technical solution, various types of multimodal data such as RF signals, video signals and audio signals are collected, preprocessed and feature extracted, and these features are fused and input into the spectrum analysis model, thereby improving the accuracy and reliability of spectrum analysis.

[0018] Optionally, in the process of fusing the key features of different modalities, the method further includes: the fusion method includes early fusion, mid-term fusion and late fusion.

[0019] In a second aspect, the present application provides a spectrum analysis device for a radio monitoring system, which adopts the following technical solution: A spectrum analysis device for a radio monitoring system, comprising: The RF signal receiving module to be analyzed obtains the corresponding environmental noise level during the monitoring process, adjusts the sensitivity threshold corresponding to the signal detection based on the environmental noise level, is used to receive the corresponding RF signal to be analyzed, and pre-processes the received RF signal to be analyzed based on the adaptive filter; The spectrum classification result acquisition module calls the pre-trained spectrum analysis model and inputs the pre-processed RF signal to be analyzed into the spectrum analysis model to obtain the corresponding spectrum classification result; The spectrum analysis data acquisition module retrieves the frequency information and time series data corresponding to the radio frequency signal to be analyzed, and uses the spectrum classification result, the frequency information and the time series data to acquire the corresponding spectrum analysis data, and displays the key parameters in the spectrum analysis data through a graphical interface, wherein the key parameters include frequency occupancy and signal strength distribution diagram.

[0020] In a third aspect, the present application provides a spectrum analysis system for a radio monitoring system, which adopts the following technical solution: A spectrum analysis system for a radio monitoring system comprises a processor, wherein the processor runs a program of any one of the above-mentioned spectrum analysis methods for a radio monitoring system.

[0021] In a fourth aspect, the present application provides a storage medium, which adopts the following technical solution: A storage medium stores a program of any one of the above-mentioned spectrum analysis methods for a radio monitoring system.

[0022] In summary, the present application includes at least one of the following beneficial technical effects: by dynamically adjusting the sensitivity threshold of signal detection, using adaptive filters to preprocess signals, applying pre-trained spectrum analysis models, and combining frequency information with time series data, accurate detection and classification of radio frequency signals are achieved, and the reliability and accuracy of spectrum analysis are improved. In addition, by identifying abnormal signals, optimizing spectrum clearing strategies, generating optimal spectrum usage recommendations for specific areas, building a virtual spectrum simulation environment, and fusing multimodal data, the system's monitoring accuracy and the efficiency and safety of spectrum management are further enhanced. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1The present invention is a flow chart showing a spectrum analysis method for a radio monitoring system according to an exemplary embodiment.

[0024] Figure 2 The present invention is a structural block diagram of a spectrum analysis method and device for a radio monitoring system according to an exemplary embodiment. DETAILED DESCRIPTION

[0025] Embodiments of the present application are described in detail below, examples of which are shown in the accompanying drawings.

[0026] In the description of this specification, the description with reference to the terms "certain embodiments", "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiments or examples are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0027] The present application embodiment discloses a spectrum analysis method for a radio monitoring system, referring to Figure 1 ,include: S100, during the monitoring process, obtaining a corresponding environmental noise level, adjusting a sensitivity threshold corresponding to signal detection based on the environmental noise level, receiving a corresponding radio frequency signal to be analyzed, and preprocessing the received radio frequency signal to be analyzed based on an adaptive filter.

[0028] The noise level in the environment is collected by radio monitoring equipment, which is usually done using a special noise measurement device or a spectrum analyzer. The noise level in the current environment is collected using the noise measurement module in the radio monitoring equipment or an external noise measurement device. The spectrum characteristics of the background noise are extracted by performing spectrum analysis on the signals received within a certain period of time, and the noise level can be quantified by calculating statistics such as the mean and standard deviation of the signal strength.

[0029] Based on the noise level obtained, the sensitivity threshold in the signal detection algorithm is adjusted. This means that when the noise level is high, the sensitivity threshold is increased to reduce false positives (false alarms), and when the noise level is low, the sensitivity threshold is decreased to reduce false negatives (missed alarms). In addition, the adjustment of the sensitivity threshold can be set manually or automatically, the latter being more common in modern radio monitoring systems.

[0030] In addition, the radio monitoring equipment receives radio frequency signals from the air, which may come from multiple directions and frequency bands. The signals are converted into digital form and are ready for further processing. These radio frequency signals to be analyzed may be collected from multiple monitoring sites or antenna arrays. Then, an adaptive filter is used to pre-process the collected radio frequency signals to be analyzed, with the aim of removing noise and other interference from the signals, thereby improving the quality of useful signals. In addition, the adaptive filter can dynamically adjust its filtering parameters according to the characteristics of the input signal to achieve the best denoising effect. It should be pointed out here that the adaptive filter may include a minimum mean square error (LMS) filter, a recursive least squares (RLS) filter, or other types of filters.

[0031] By adjusting the sensitivity threshold and using adaptive filters, the signal-to-noise ratio can be effectively improved to ensure accurate signal detection in different noise environments.

[0032] S110, calling a pre-trained spectrum analysis model, inputting the pre-processed radio frequency signal to be analyzed into the spectrum analysis model, and obtaining a corresponding spectrum classification result.

[0033] Among them, a pre-trained spectrum analysis model is loaded. The spectrum analysis model is usually trained based on a large amount of training data and deep learning algorithms, and can effectively identify and classify different RF signals. Then the RF signal preprocessed by the adaptive filter is input into the spectrum analysis model. The spectrum analysis model processes the input signal and outputs the classification result. The classification result indicates the type of signal, such as Wi-Fi signal, Bluetooth signal, 4G / 5G signal, etc. In addition, the classification result may include the probability distribution of the signal type or a clear category label.

[0034] Through the pre-trained model, RF signals can be automatically identified and classified, which can improve the accuracy of signal classification and reduce the possibility of misjudgment and missed detection, thus providing a basis for subsequent spectrum analysis.

[0035] S120, retrieve frequency information and time series data corresponding to the radio frequency signal to be analyzed, obtain corresponding spectrum analysis data based on the spectrum classification result, frequency information and time series data, and display key parameters in the spectrum analysis data through a graphical interface.

[0036] Among them, frequency information (such as spectrum diagram) and time series data (such as the change of signal strength over time) are extracted from the preprocessed RF signal. In addition, frequency information can be obtained by fast Fourier transform (FFT) or other frequency domain transformation methods; time series data can be directly extracted from the preprocessed signal.

[0037] The generated spectrum analysis data can include key parameters such as frequency occupancy and signal strength distribution. Specifically, frequency occupancy: calculate the occupancy of each frequency point over a period of time. This is usually obtained by statistical analysis of signal strength. The occupancy of frequency points over a period of time can be used to generate a frequency occupancy map; signal strength distribution: record the signal strength at different frequency points and draw a signal strength distribution map. These maps can show which frequency points have stronger signals and which have weaker signals; other parameters: in addition to frequency occupancy and signal strength distribution, other parameters can also be calculated, such as signal bandwidth, signal modulation method, signal duration, etc.

[0038] In addition, for display through a graphical interface, different parameters can be displayed in different graphical ways, for example, frequency occupancy graph: the frequency occupancy data is displayed in a two-dimensional graph, the horizontal axis represents the frequency range, the vertical axis represents the time, and different colors or shadows represent different occupancy levels; signal strength distribution graph: the signal strength distribution data is displayed in a two-dimensional graph, the horizontal axis represents the frequency range, the vertical axis represents the signal strength, and different colors or line thicknesses can be used to represent different levels of signal strength; other graphs: other graphs can also be included, such as a trend graph of signal strength over time, a graph of signal type over time, etc. And the spectrum analysis data is intuitively displayed through a graphical interface, so that users can quickly understand the use of the spectrum.

[0039] In the process of obtaining the corresponding spectrum classification result, in the embodiment of the present application, the method further includes: S131: If the spectrum analysis model identifies an abnormal signal that deviates from the normal mode.

[0040] When the spectrum analysis model classifies the pre-processed RF signals, the model identifies the signal type based on the features in the training data. If the model finds that some signals do not conform to the known normal pattern (such as abnormally high or low signal strength, abnormal frequency occupancy pattern, etc.), these signals are considered abnormal signals. This can promptly detect signals that do not conform to the normal usage pattern, indicating possible illegal or unauthorized frequency band usage.

[0041] S132, calculating deviation data between the abnormal signal and the normal mode signal.

[0042] Among them, feature extraction is performed on the abnormal signal, including but not limited to signal strength, frequency occupancy, signal duration, etc., and the features of the abnormal signal are compared with the features of the normal mode signal. The difference between the two is calculated, such as the difference in signal strength, the difference in frequency occupancy, etc. By calculating the deviation data between the abnormal signal and the normal mode signal, the difference between the features of the abnormal signal and the features of the normal signal can be quantified.

[0043] S133, classify the abnormal signal based on the deviation data, obtain the corresponding target abnormal type, and generate a corresponding abnormal report.

[0044] Among them, the abnormal signal is classified according to the calculated deviation data using the predefined abnormal type standard, and the abnormal type may include but is not limited to illegal signals, interference signals, unknown signals, etc. Through the deviation data, it is determined which abnormal type the abnormal signal belongs to. For example, if the signal strength is much higher than the normal level, it may be classified as an interference signal; if the signal occupies an unauthorized frequency band, it may be classified as an illegal signal.

[0045] At the same time, a detailed abnormality report is generated based on the type and characteristics of the abnormal signal and its deviation data from the normal mode signal. The report should include the basic information of the abnormal signal (such as frequency, intensity, duration, etc.), deviation data and classification results. The abnormality report can be in text form or in the form of prompt information in the graphical interface. There should be clear identification and warnings in the report to remind users of the existence of abnormal signals and their potential impact. The report can be sent to relevant personnel via email, SMS, application push, etc.

[0046] In addition, after obtaining the corresponding target exception type, the method further includes: S1331, detect the target abnormal type and determine whether the target abnormal type is an illegal or unauthorized frequency band.

[0047] Among them, checking whether the target anomaly type is inconsistent with the known legal frequency band can determine whether the target anomaly type is an illegal or unauthorized frequency band by querying the spectrum allocation database or referring to the legal frequency band list. If the characteristics of the target anomaly type (such as frequency range, signal strength, duration, etc.) do not match the legal frequency band, the target anomaly type is considered to be an illegal or unauthorized frequency band.

[0048] By detecting and judging whether the target anomaly type is an illegal or unauthorized frequency band, illegal spectrum occupation can be discovered and handled in a timely manner, thereby ensuring the compliance of spectrum use and preventing illegal occupation from interfering with legitimate users.

[0049] S1332: If yes, then perform an interference source clearing operation on the target abnormal type, where the interference source clearing operation includes using interference suppression technology, frequency band switching, and signal blocking.

[0050] Among them, adaptive filtering: applying adaptive filtering technology, dynamically adjusting filter parameters to weaken or eliminate interference signals, and using smart antenna technology to avoid interference sources by adjusting the directionality of the antenna; frequency band switching: when illegal or unauthorized use of a frequency band is detected, the system can automatically switch to another available frequency band. In some cases, users or administrators can manually choose to switch to other frequency bands to avoid interference; signal blocking: the system can send specific interference signals to block the transmission of illegal signals. In extreme cases, illegal signals can be blocked by physical means (such as shutting down the device, cutting off the power supply, etc.).

[0051] S1333: Obtain a corresponding actual processing effect according to the interference source clearing operation, and optimize the spectrum clearing strategy based on the actual processing effect.

[0052] After executing the interference source clearing operation, continue to monitor the spectrum usage of the target frequency band and record the impact before and after the clearing operation. By comparing key parameters such as spectrum occupancy and signal strength before and after the clearing operation, evaluate the actual effect of the clearing operation.

[0053] Then, the spectrum clearing strategy is adjusted according to the actual processing effect, such as changing the parameter settings of the interference suppression technology, optimizing the timing of frequency band switching, etc. Through continuous experiments and optimization, the efficiency and reliability of the spectrum clearing strategy can be improved.

[0054] By executing steps S131 to S1333, abnormal signals can be effectively identified and classified, and detailed abnormality reports can be provided, thereby helping users to promptly discover and deal with potential problems and improve the security and reliability of the system; at the same time, it can effectively detect and deal with illegal or unauthorized frequency bands, eliminate interference sources, and continuously optimize spectrum clearing strategies based on actual processing results, thereby further improving the security and reliability of the system.

[0055] In addition, after obtaining the radio frequency signal to be analyzed, in the embodiment of the present application, the method further includes: S141, construct a virtual spectrum simulation environment.

[0056] Among them, the spectrum simulation environment can simulate the real-world radio spectrum usage; in addition, a software-defined radio (SDR) platform, network simulation tools (such as NS-3, OMNeT++) or other dedicated spectrum simulation software can be used to build a virtual environment, and the environment includes simulated real-world elements such as geographical layout, obstacles (such as buildings), user equipment (UE) location and movement trajectory, etc.

[0057] At the same time, the simulation of real-world spectrum usage, including but not limited to signal propagation models, interference models, communication protocol models, etc., can be customized according to the actual spectrum usage to ensure that the simulation environment is as close to the actual situation as possible.

[0058] S142, generating a plurality of spectrum simulation model options corresponding to the spectrum simulation environment, and sending all spectrum simulation model options to the user for selection.

[0059] Among them, according to different application scenarios and requirements, a variety of spectrum simulation model options are generated, for example, there may be different signal propagation models, interference models, communication protocol models, etc.; each model option can be customized for different frequency bands, environmental conditions, user equipment and other factors. Then all the generated spectrum simulation model options are sent to the user through a graphical interface or other means, and the user can select a suitable model option according to his or her specific needs. After the user selects the model option, the user is allowed to configure the model parameters according to specific needs, and the parameters may include but are not limited to signal strength, device location, environmental conditions, communication protocol, etc.

[0060] S143, obtaining corresponding feedback data based on the user's options, determining a corresponding target spectrum simulation model based on the feedback data, controlling the target spectrum simulation model to run simulation in real time, and displaying the simulation results through a graphical interface.

[0061] Among them, based on the user's options, the model selected by the user and its configuration parameters are obtained, and the feedback data includes the model type selected by the user and the specific parameters configured by the user. According to the obtained feedback data, the target spectrum simulation model selected by the user is determined to ensure that the model configuration is correct and prepare to start the simulation.

[0062] Control the selected target spectrum simulation model to run the simulation in real time, update the simulation results in real time, and ensure the accuracy and timeliness of the data in the simulation process. And display the simulation results to the user through a graphical interface, which can include charts, spectrum diagrams, signal strength distribution diagrams, etc., so that users can intuitively view and analyze the simulation results through the graphical interface.

[0063] By executing the above steps S141 to S143 and running the simulation in real time, the simulation results can be obtained instantly, thereby improving the efficiency and accuracy of the simulation process; by displaying the simulation results through a graphical interface, the user can intuitively view and analyze the simulation data, which can help the user better understand and make decisions.

[0064] In the embodiment of the present application, factors affecting the radio frequency signal to be analyzed in different places are also taken into consideration, so the method further includes: S151, obtaining current location information of the radio frequency signal to be analyzed, and determining a corresponding specific area based on the current location information.

[0065] Among them, the current location information of the radio frequency signal to be analyzed is obtained using GPS, cellular network positioning or other geographic positioning technology, and the location information may include latitude and longitude coordinates, address, etc.; according to the obtained current location information, a specific geographical area is determined, and the specific area may be a city, a cell, or a smaller geographical range, depending on the needs of spectrum management. By obtaining the current location information of the radio frequency signal, the geographical area where the signal is located can be accurately determined.

[0066] S152, collecting historical spectrum usage data in a specific area.

[0067] The historical spectrum usage data in a specific area is collected from a historical database or real-time monitoring system. The data source can be a radio monitoring device, a sensor network or other data acquisition system. It should be pointed out here that the historical spectrum usage data specifically includes: frequency occupancy rate: statistics on the occupancy of each frequency band in a specific area; signal strength distribution: record the strength distribution of signals in each frequency band; interference situation: analyze the interference sources and their interference levels in a specific area; spectrum utilization rate: calculate the overall utilization rate of the spectrum in a specific area.

[0068] S153: Generate optimal spectrum usage suggestions in a specific area based on the current location information and historical spectrum usage data.

[0069] Among them, by generating the best spectrum usage recommendation in a specific area, the use of spectrum resources can be optimized and the spectrum utilization rate can be improved. The best spectrum usage recommendation specifically includes: frequency band selection: based on the frequency occupancy rate and spectrum utilization rate in the historical spectrum usage data, it is recommended to use the relatively idle frequency band in the specific area. For example, if the historical occupancy rate of a certain frequency band is low, it can be recommended to use this frequency band; power control: based on the signal strength distribution and interference situation, it is recommended to recommend appropriate transmission power. For example, if the signal strength of a certain frequency band is low and the interference is less, it can be recommended to increase the transmission power; time slot allocation: based on the spectrum utilization rate in the historical spectrum usage data, it is recommended to use the frequency band in the time period when the frequency band is relatively idle. For example, if a certain frequency band is relatively idle at night, it can be recommended to use the frequency band at night; multi-band combination: based on the historical spectrum usage data, it is recommended to use a combination of multiple frequency bands to achieve more efficient spectrum utilization. For example, it can be recommended to switch to the backup frequency band when the main frequency band is busy, or to use multiple frequency bands at the same time to disperse the load.

[0070] By executing the above steps S151 to S153, it is possible to generate optimal spectrum usage recommendations in a specific area based on the current location information of the RF signal to be analyzed and historical spectrum usage data, thereby helping users optimize the use of spectrum resources and improve communication efficiency and quality.

[0071] In the solution of the embodiment of the present application, in order to further improve the accuracy of radio frequency signal recognition and classification, the method also includes: S161, collecting different types of multimodal analysis data from pre-installed sensors, the analysis data including radio frequency signals, video signals, and audio signals; and pre-processing the collected multimodal analysis data.

[0072] Among them, RF signal collection: use radio monitoring equipment (such as SDR) to collect RF signals. These signals can be broadband signals or signals in a specific frequency band, and the data can include information such as signal strength, frequency, and phase; video signal collection: use cameras or other visual sensors to collect video signals, and video data can include information such as frame rate and resolution; audio signal collection: use microphones or other audio sensors to collect audio signals, and audio data can include information such as sampling rate and number of channels.

[0073] By collecting various types of multimodal data such as RF signals, video signals, and audio signals, richer contextual information can be obtained, which helps to improve the accuracy and reliability of spectrum analysis.

[0074] The preprocessing of the collected multimodal analysis data includes: RF signal preprocessing: Use filters to remove noise from RF signals; enhance useful signals through signal processing techniques (such as gain adjustment); convert RF signals into a unified data format for subsequent processing.

[0075] Video signal preprocessing: Improve image quality by adjusting parameters such as brightness and contrast; use image processing algorithms to remove noise in the video; convert video signals into a unified data format for subsequent processing.

[0076] Audio signal preprocessing: Use audio processing algorithms to remove background noise; improve audio quality by adjusting volume and other parameters; convert audio signals into a unified data format for subsequent processing.

[0077] S162, extracting corresponding key features from the preprocessed multimodal analysis data, and fusing the key features of different modalities.

[0078] Key features include RF features, video features, and audio features. Specifically, RF feature extraction: extracting frequency domain features of the signal, such as frequency distribution, occupancy, etc.; extracting time domain features of the signal, such as signal strength, pulse width, etc. Video feature extraction: extracting image features, such as color histogram, edge detection results, etc.; extracting motion features in the video, such as optical flow field, motion vector, etc. Audio feature extraction: extracting acoustic features of audio, such as Mel-frequency cepstral coefficients (MFCC), fundamental frequency, etc.; extracting semantic features in audio, such as speech recognition results, etc.

[0079] In the process of fusing the key features of different modalities, it can be specifically: early fusion: directly fusing data of different modalities in the feature extraction stage to generate a comprehensive feature representation; mid-term fusion: fusing features of different modalities after feature extraction but before the final decision, for example, a multi-layer perceptron or a multimodal convolutional neural network can be used to fuse features; late fusion: after classification or identification of each modality, the results are fused, for example, using voting mechanism, weighted average and other methods for fusion.

[0080] By extracting the key features of different modes and fusing them, more comprehensive information can be obtained and the accuracy and robustness of spectrum analysis can be improved.

[0081] S163, inputting the fused key features into a spectrum analysis model to obtain corresponding spectrum analysis results.

[0082] The spectrum analysis model processes the input fusion features and outputs spectrum analysis results, which may include spectrum classification, anomaly detection and other information.

[0083] The notification executes steps S161 to S163, by collecting various types of multimodal data, preprocessing, feature extraction and fusion thereof, and finally inputting the spectrum analysis model to obtain the spectrum analysis results, thereby improving the accuracy and reliability of the spectrum analysis.

[0084] The present application embodiment discloses a spectrum analysis device for a radio monitoring system, referring to Figure 2 , including but not limited to: The RF signal receiving module 200 to be analyzed obtains the corresponding environmental noise level during the monitoring process, adjusts the sensitivity threshold corresponding to the signal detection based on the environmental noise level, is used to receive the corresponding RF signal to be analyzed, and pre-processes the received RF signal to be analyzed based on the adaptive filter; The spectrum classification result acquisition module 210 retrieves a pre-trained spectrum analysis model and inputs the pre-processed radio frequency signal to be analyzed into the spectrum analysis model to obtain a corresponding spectrum classification result; The spectrum analysis data acquisition module 220 retrieves the frequency information and time series data corresponding to the RF signal to be analyzed, and uses the spectrum classification results, frequency information and time series data to obtain the corresponding spectrum analysis data, and displays the key parameters in the spectrum analysis data through a graphical interface, wherein the key parameters include frequency occupancy and signal strength distribution diagram.

[0085] Further, including but not limited to: If the spectrum analysis model identifies an abnormal signal that deviates from the normal pattern; A deviation data calculation module, used to calculate the deviation data between the abnormal signal and the normal mode signal; The target anomaly type acquisition module classifies the abnormal signal based on the deviation data to obtain the corresponding target anomaly type and generate the corresponding anomaly report.

[0086] Further, including but not limited to: A target abnormal type judgment module detects the target abnormal type to determine whether the target abnormal type is an illegal or unauthorized frequency band; The clearing module, if yes, is used to perform interference source clearing operation on the target abnormal type, and the interference source clearing operation includes using interference suppression technology, frequency band switching, and signal blocking; The spectrum clearing strategy optimization module obtains the corresponding actual processing effect according to the interference source clearing operation, and optimizes the spectrum clearing strategy based on the actual processing effect.

[0087] Further, including but not limited to: A current location information acquisition module, used to acquire the current location information of the radio frequency signal to be analyzed, and determine the corresponding specific area based on the current location information; A historical spectrum usage data collection module is used to collect historical spectrum usage data in a specific area. The historical spectrum usage data includes frequency occupancy, signal strength distribution, interference, and spectrum utilization. The optimal spectrum usage suggestion generation module is used to generate the optimal spectrum usage suggestion in a specific area based on the current location information and historical spectrum usage data. The optimal spectrum usage suggestion includes frequency band selection, power control, time slot allocation, and multi-band combination.

[0088] Further, including but not limited to: A spectrum simulation environment construction module is used to construct a virtual spectrum simulation environment, wherein the spectrum simulation environment can simulate the use of radio spectrum in the real world; A spectrum simulation model option generation module is used to generate a variety of spectrum simulation model options corresponding to the spectrum simulation environment, and send all spectrum simulation model options to the user for selection, wherein the spectrum simulation environment allows the user to configure model parameters according to specific needs; The target spectrum simulation model determination module obtains corresponding feedback data based on user options, determines the corresponding target spectrum simulation model based on the feedback data, controls the target spectrum simulation model to run simulation in real time, and displays the simulation results through a graphical interface.

[0089] Further, including but not limited to: A multimodal analysis data collection module is used to collect different types of multimodal analysis data from pre-installed sensors, the analysis data including radio frequency signals, video signals, and audio signals; and pre-process the collected multimodal analysis data; A fusion module extracts corresponding key features from the preprocessed multimodal analysis data to fuse the key features of different modalities, where the key features include radio frequency features, video features, and audio features; The spectrum analysis result acquisition module inputs the fused key features into the spectrum analysis model to obtain the corresponding spectrum analysis results.

[0090] Further, including but not limited to: Fusion methods include early fusion, mid-term fusion and late fusion.

[0091] An embodiment of the present application further discloses a spectrum analysis system for a radio monitoring system, comprising a processor, wherein a program of any one of the above-mentioned spectrum analysis methods for a radio monitoring system is running on the processor.

[0092] The embodiment of the present application further discloses a storage medium storing a program of any one of the above-mentioned spectrum analysis methods for a radio monitoring system.

[0093] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A spectrum analysis method for a radio monitoring system, characterized in that: include: During the monitoring process, a corresponding environmental noise level is obtained, a sensitivity threshold corresponding to signal detection is adjusted based on the environmental noise level, a corresponding radio frequency signal to be analyzed is received, and the received radio frequency signal to be analyzed is preprocessed based on an adaptive filter; Retrieve a pre-trained spectrum analysis model, input the pre-processed RF signal to be analyzed into the spectrum analysis model, and obtain the corresponding spectrum classification result; Retrieve frequency information and time series data corresponding to the radio frequency signal to be analyzed, obtain corresponding spectrum analysis data based on the spectrum classification result, the frequency information and the time series data, and display key parameters in the spectrum analysis data through a graphical interface, wherein the key parameters include frequency occupancy and signal strength distribution diagram.

2. The spectrum analysis method for a radio monitoring system according to claim 1, characterized in that: In the process of obtaining the corresponding spectrum classification result, the method further includes: If the spectrum analysis model identifies an abnormal signal that deviates from the normal pattern; Calculating deviation data between the abnormal signal and the normal mode signal; The abnormal signal is classified based on the deviation data, a corresponding target abnormal type is obtained, and a corresponding abnormal report is generated.

3. The spectrum analysis method for a radio monitoring system according to claim 2, characterized in that: After obtaining the corresponding target exception type, the method further includes: Detecting the target abnormal type to determine whether the target abnormal type is an illegal or unauthorized frequency band; If yes, then the interference source clearing operation is performed on the target abnormal type, and the interference source clearing operation includes using interference suppression technology, frequency band switching, and signal blocking; The corresponding actual processing effect is obtained according to the interference source clearing operation, and the spectrum clearing strategy is optimized based on the actual processing effect.

4. The spectrum analysis method for a radio monitoring system according to claim 1, characterized in that: The method also includes: Acquire current location information of the radio frequency signal to be analyzed, and determine a corresponding specific area based on the current location information; Collecting historical spectrum usage data within the specific area, wherein the historical spectrum usage data includes frequency occupancy rate, signal strength distribution, interference situation, and spectrum utilization rate; An optimal spectrum usage suggestion in a specific area is generated based on the current location information and historical spectrum usage data, wherein the optimal spectrum usage suggestion includes frequency band selection, power control, time slot allocation, and multi-band combination.

5. The spectrum analysis method for a radio monitoring system according to claim 1, characterized in that: After obtaining the radio frequency signal to be analyzed, the method further includes: Constructing a virtual spectrum simulation environment, wherein the spectrum simulation environment can simulate the use of radio spectrum in the real world; Generate a plurality of corresponding spectrum simulation model options for the spectrum simulation environment, and send all of the spectrum simulation model options to the user for selection, wherein the spectrum simulation environment allows the user to configure model parameters according to specific needs; Corresponding feedback data is obtained based on user options, a corresponding target spectrum simulation model is determined based on the feedback data, the target spectrum simulation model is controlled to run simulation in real time, and the simulation results are displayed through a graphical interface.

6. The spectrum analysis method for a radio monitoring system according to claim 1, characterized in that: The method also includes: Collecting different types of multimodal analysis data from pre-installed sensors, the analysis data including radio frequency signals, video signals, and audio signals; and pre-processing the collected multimodal analysis data; Extract corresponding key features from the preprocessed multimodal analysis data and fuse the key features of different modalities, where the key features include radio frequency features, video features, and audio features; The fused key features are input into the spectrum analysis model to obtain the corresponding spectrum analysis results.

7. The spectrum analysis method for a radio monitoring system according to claim 6, characterized in that: In the process of fusing the key features of different modalities, the method further includes: The fusion method includes early fusion, mid-term fusion and late fusion.

8. A spectrum analysis device for a radio monitoring system, characterized in that: include: The RF signal receiving module to be analyzed obtains the corresponding environmental noise level during the monitoring process, adjusts the sensitivity threshold corresponding to the signal detection based on the environmental noise level, is used to receive the corresponding RF signal to be analyzed, and pre-processes the received RF signal to be analyzed based on the adaptive filter; The spectrum classification result acquisition module calls the pre-trained spectrum analysis model and inputs the pre-processed RF signal to be analyzed into the spectrum analysis model to obtain the corresponding spectrum classification result; The spectrum analysis data acquisition module retrieves the frequency information and time series data corresponding to the radio frequency signal to be analyzed, and uses the spectrum classification result, the frequency information and the time series data to acquire the corresponding spectrum analysis data, and displays the key parameters in the spectrum analysis data through a graphical interface, wherein the key parameters include frequency occupancy and signal strength distribution diagram.

9. A spectrum analysis system for a radio monitoring system, characterized in that: The method comprises a processor in which a program of a spectrum analysis method for a radio monitoring system as claimed in any one of claims 1 to 7 is run.

10. A storage medium, characterized in that: A program for storing the spectrum analysis method for a radio monitoring system according to any one of claims 1 to 7 is stored.

Images