A spectrum target generation method based on real-time spectrum monitoring data

Abstract

The application discloses a spectrum target generation method based on real-time spectrum monitoring data and belongs to the technical field of spectrum target generation. The method comprises the following steps: S1, receiving and analyzing real-time spectrum monitoring data, extracting characteristic parameters, and forming structured metadata; S2, classifying data according to signal systems, and starting corresponding target generation services; S3, dynamically associating and updating new data with existing candidate spectrum targets in a cache area, screening candidate targets, performing motion trend consistency verification on the screened candidate targets in combination with angle rotation information, and determining a final associated target or creating a new target according to a verification result by adopting a minimum azimuth deviation strategy; and S4, cyclically executing the above steps to realize continuous generation and updating of spectrum targets. By introducing motion trend consistency verification, the application effectively solves the target confusion problem of multiple azimuthally close radiation sources, and significantly improves the accuracy and reliability of spectrum target generation.

Description

Technical Field

[0001] This invention relates to the field of spectrum target generation technology, and in particular to a method for generating spectrum targets based on real-time spectrum monitoring data. Background Technology

[0002] Spectrum monitoring is a crucial means of understanding the electromagnetic environment and ensuring the effective utilization and security of spectrum resources. In typical spectrum monitoring scenarios, the system continuously receives real-time spectrum monitoring and reconnaissance data from various radiation sources. Accurately and efficiently extracting features from this massive, dynamic real-time data and generating stable, traceable "spectrum targets" to form a clear spectrum target situation is a key aspect of spectrum monitoring data processing and a foundation for providing effective decision support for operators.

[0003] Traditional methods for generating spectrum targets typically rely on setting simple association or matching rules for several basic characteristic parameters in real-time monitoring data. However, with the increasing complexity of monitoring environments, these traditional methods have gradually revealed their limitations. Especially in certain scenarios, such as when multiple radiation sources are located close to each other within the monitoring area and their movement trends remain relatively constant relative to the spectrum monitoring equipment, their instantaneous location information is highly similar, easily leading to erroneous associations. This can result in signal traces belonging to different radiation sources being incorrectly merged into the same "spectrum target," or the inability to effectively distinguish and establish different target trajectories. Figure 2 As shown, the target trajectories generated by traditional methods in this scenario may be confused, intersecting, or broken, which seriously affects the accuracy and reliability of spectrum target situation generation.

[0004] The essence of this problem lies in the fact that traditional methods lack in-depth exploration and utilization of the inherent motion patterns and trends of targets. When the instantaneous spatial position parameters (orientation) of multiple targets are difficult to distinguish, it is necessary to introduce higher-dimensional, discriminative features to assist in decision-making. Therefore, there is an urgent need for a spectrum target generation method that can more deeply mine the characteristics of real-time spectrum monitoring data, especially in challenging scenarios where targets are in similar orientations, and still achieve high accuracy and high reliability, in order to improve the overall efficiency of the spectrum monitoring system. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method for generating spectrum targets based on real-time spectrum monitoring data.

[0006] The objective of this invention is achieved through the following technical solution: a method for generating spectrum targets based on real-time spectrum monitoring data, comprising the following steps: S1. Data Processing and Feature Extraction Stage: Receive real-time spectrum monitoring and reconnaissance data, parse and extract the feature parameters of the real-time spectrum monitoring and reconnaissance data, and establish structured metadata; S2, Signal System Classification Processing Stage: Based on the signal system identification results in the feature parameters, the real-time spectrum monitoring data is classified, and different spectrum target generation services are provided for real-time spectrum monitoring data of different signal systems based on the classification results; S3. Dynamic spectrum target generation and update stage: Based on the candidate spectrum targets generated in the current cache, according to the preset spectrum target processing rules, the spectrum monitoring data to be processed is associated with an existing candidate spectrum target or a new candidate spectrum target is created based on the spectrum monitoring data to be processed, and the update result is stored in the cache. S4. Loop Processing Phase: Continuously acquire the latest real-time spectrum monitoring data and repeat steps S2 to S3 to generate spectrum targets until no new data is input and the generation ends.

[0007] Preferably, the characteristic parameters include at least: signal frequency, signal transmission time, signal transmission location, and signal system identification result.

[0008] Preferably, the S3 dynamic spectrum target generation and update stage further includes the following steps: S31. Candidate target screening step: Compare the spectrum monitoring data to be processed with the spectrum target feature data in the cache, and screen out the candidate spectrum targets in the cache that meet the preset spectrum target processing rules; S32. Motion trend consistency verification step: Analyze the motion trend of the spectrum target corresponding to the spectrum monitoring data to be processed and the selected candidate spectrum targets. If the motion trends are consistent, add the candidate spectrum target to the candidate target queue. S33. Target selection step: Based on the preset decision strategy, determine the final associated target from the verified candidate target queue or create a new candidate spectrum target according to the spectrum target generation rules.

[0009] Preferably, the preset spectrum target processing rule is as follows: the spectrum target feature data in the cache and the spectrum monitoring data to be processed must simultaneously meet the preset deviation threshold rule in three dimensions: signal outgoing time, signal frequency, and signal outgoing direction.

[0010] Preferably, the deviation threshold rules include the following: the time continuity deviation of the signal outgoing time is less than the time deviation threshold; the signal frequency stability deviation is less than the frequency deviation threshold; and the signal outgoing azimuth continuity deviation is less than the azimuth deviation threshold.

[0011] Preferably, in step S32, the motion trend consistency verification step, the motion trend is analyzed based on the angle rotation information to determine whether the motion trend is consistent.

[0012] Preferably, the preset decision-making strategy is: If the number of targets in the candidate target queue is greater than 1, the candidate spectrum target with the smallest azimuth deviation between the spectrum target corresponding to the spectrum monitoring data to be processed is selected as the final associated target. If the number of targets in the candidate target queue is equal to 1, then the only candidate spectrum target in the candidate target queue will be used as the final associated target. If the number of targets in the candidate target queue is 0, a new candidate spectrum target is created based on the spectrum monitoring data to be processed, according to the spectrum target generation rules.

[0013] Preferably, the preset spectrum target processing rules are set based on expert experience.

[0014] The beneficial effects of this invention are: 1) This invention introduces a motion trend consistency check, which extracts and analyzes angular rotation information to determine whether the motion trends are consistent. This allows for the differentiation of radiation sources that are spatially close but have different motion patterns from the perspective of dynamic behavioral characteristics. When multiple candidate targets exist, an optimization decision-making strategy of "minimizing azimuth deviation" is further adopted. This enables accurate and stable target trajectory association and construction even in complex monitoring scenarios where multiple target motion trajectories intersect or run parallel, overcoming the error-prone nature of traditional methods in such scenarios.

[0015] 2) This invention does not simply rely on a single or static threshold rule, but rather constructs a closed-loop processing flow from "feature extraction - classification - preliminary screening - trend verification - optimization decision-making," comprehensively utilizing the time-domain, frequency-domain, and spatial-domain features of the signal, as well as deep motion trend features, and performing correlation judgments through hierarchical and progressive rules. This structured processing method makes the generation process of spectral targets more rigorous and reliable, reducing the occurrence of false generation or target fragmentation caused by noise interference or instantaneous parameter fluctuations.

[0016] 3) This invention classifies signals based on their system identification results at the initial stage of data processing and initiates corresponding target generation services for signals of different systems. This design allows the method to flexibly adapt to the different characteristics of various communication or radar signals, employing more targeted association rules for processing. This improves the overall adaptability and parallel processing capability of the system, facilitating efficient analysis of complex heterogeneous electromagnetic environments. Attached Figure Description

[0017] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a schematic diagram illustrating the generation results of spectral targets with similar orientations using traditional methods. Figure 3 This is a schematic diagram illustrating the generation result of the spectral target with similar orientation in this invention. Detailed Implementation

[0018] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] See Figures 1-3 This invention provides a technical solution: a method for generating spectrum targets based on real-time spectrum monitoring data, comprising the following steps: S1. Data Processing and Feature Extraction Stage: Receive real-time spectrum monitoring and reconnaissance data, parse and extract the feature parameters of the real-time spectrum monitoring and reconnaissance data, and establish structured metadata; S2, Signal System Classification Processing Stage: Based on the signal system identification results in the feature parameters, the real-time spectrum monitoring data is classified, and different spectrum target generation services are provided for real-time spectrum monitoring data of different signal systems based on the classification results; S3. Dynamic spectrum target generation and update stage: Based on the candidate spectrum targets generated in the current cache, according to the preset spectrum target processing rules, the spectrum monitoring data to be processed is associated with an existing candidate spectrum target or a new candidate spectrum target is created based on the spectrum monitoring data to be processed, and the update result is stored in the cache. S4. Loop Processing Phase: Continuously acquire the latest real-time spectrum monitoring data and repeat steps S2 to S3 to generate spectrum targets until no new data is input and the generation ends.

[0020] In this embodiment, addressing the need for rapid, efficient, and accurate generation of spectrum targets in feature mining of real-time spectrum monitoring data, a spectrum target generation rule is formulated by comprehensively utilizing multi-dimensional feature parameters acquired from spectrum monitoring, such as signal frequency, signal outgoing time, signal outgoing location, and signal system identification results. This enables the processing of real-time spectrum monitoring data and the generation of spectrum targets. Based on the multi-dimensional feature parameters acquired from real-time spectrum monitoring, this invention further extracts spectrum target angle rotation information, effectively solving the problem of overlapping spectrum target motion trends, improving the accuracy of spectrum target generation, and providing fundamental support for spectrum target situational awareness generation.

[0021] In some embodiments, the characteristic parameters include at least: signal frequency, signal transmission time, signal transmission location, and signal system identification result.

[0022] In this embodiment, S1, the data processing and feature extraction stage: The spectrum monitoring equipment (such as a spectrum analyzer, monitoring station, etc.) continuously scans the airspace and receives the raw spectrum monitoring and reconnaissance data stream. S1 receives this real-time data stream and parses it. The purpose of parsing is to extract structured feature parameters for target association and generation from the raw I / Q data, spectrum diagram, or pulse descriptor word (PDW). These key feature parameters include at least: Signal frequency (f): The center frequency or carrier frequency of the current signal.

[0023] Signal transmission time (t): The absolute or relative timestamp at which the signal is intercepted by the monitoring equipment.

[0024] Signal outgoing azimuth (a): The azimuth angle of the signal source obtained by the direction finding equipment.

[0025] Signal type identification result (Type): The signal type identified through modulation identification, parameter matching and other techniques, such as FM, AM, BPSK, radar pulse train, etc.

[0026] These parameters are associated with the corresponding signal data to form "structured metadata" with time, frequency, orientation, and system type labels, laying the foundation for subsequent processing.

[0027] S2. Signal System Classification and Processing Stage: Based on the "Signal System Identification Result (Type)" extracted in step S1, the continuously input real-time spectrum monitoring data is classified. For example, all signal data identified as "FM communication" is assigned to the Class A processing queue, and all signal data identified as "XXX type radar pulse" is assigned to the Class B processing queue. The system pre-configures or dynamically configures a corresponding "Spectrum Target Generation Service" for each signal system type. These services may include exclusive association rules, filtering parameters, and motion models that match the signal characteristics of that system. For example, different association tolerances can be set for communication signals and radar signals in terms of frequency stability (Δf), pulse repetition interval, etc. This step realizes the diversion and specialization of data processing, improving the efficiency and accuracy of subsequent processing.

[0028] In some embodiments, the S3 dynamic spectrum target generation and update stage further includes the following steps: S31. Candidate target screening step: Compare the spectrum monitoring data to be processed with the spectrum target feature data in the cache, and screen out the candidate spectrum targets in the cache that meet the preset spectrum target processing rules; S32. Motion trend consistency verification step: Analyze the motion trend of the spectrum target corresponding to the spectrum monitoring data to be processed and the selected candidate spectrum targets. If the motion trends are consistent, add the candidate spectrum target to the candidate target queue. S33. Target selection step: Based on the preset decision strategy, determine the final associated target from the verified candidate target queue or create a new candidate spectrum target according to the spectrum target generation rules.

[0029] In this embodiment, S3, the dynamic spectrum target generation and update stage, is the core stage of the invention. Its purpose is to intelligently associate newly arriving single monitoring data (i.e., "spectrum monitoring data to be processed") with the existing and continuously maintained set of spectrum targets in the system (stored in the cache) to update the target trajectory or create new targets. This stage specifically includes the following sub-steps: S31. Candidate Target Screening Step: The current spectrum monitoring data to be processed (its feature parameter set is denoted as M_new = {t_new, f_new, a_new, Type_new}) is compared with the historical feature data of all generated "candidate spectrum targets" with the same signal system in the buffer. Each candidate spectrum target can be regarded as a sequence composed of historical data points. The comparison is based on a set of experience-based "preset spectrum target processing rules". In this embodiment, the rules require that the new data and the latest historical data of the target must simultaneously meet the continuity condition in three dimensions: Temporal continuity: |t_new - t_last| < Δt, where t_last is the time of the latest data point of the target, and Δt is the time deviation threshold (e.g., 30 seconds).

[0030] Frequency stability: |f_new - f_last| < Δf, where f_last is the statistical value of the target's historical frequency (such as the mean or the latest value), and Δf is the frequency deviation threshold (e.g., 25kHz).

[0031] Azimuth continuity: |a_new - a_last| < Δa, where a_last is the best estimate of the historical azimuth of the target (such as the latest value or filtered prediction value), and Δa is the azimuth deviation threshold (e.g., 2 degrees). Only cache targets that simultaneously meet all three conditions will be initially screened as "candidate spectrum targets" and enter the next round of analysis. This step completes the coarse correlation, filtering out obviously irrelevant targets.

[0032] S32. Motion Trend Consistency Verification Step: After the screening in step S31, there may still be multiple candidate targets (especially when they are close in azimuth). S32 aims to introduce the deep feature of motion trend for secondary filtering. For each candidate spectral target, combined with the current new data M_new, the "angle rotation information" Δw of the target after adding the new point is calculated. Δw reflects the rate of change or trend of the target's azimuth angle over time. By analyzing the changes of Δw within a time window, the motion trend of the target (such as uniform linear motion, turning, hovering, etc.) can be determined.

[0033] S32 requires that the instantaneous motion trend implied by the new data M_new must be consistent with the existing historical motion trend of the candidate spectrum target. For example, if the historical trajectory of a candidate spectrum target indicates that it is rotating clockwise at a constant angular velocity (Δw is a positive constant), but the instantaneous motion direction calculated by the new data M_new is counterclockwise (Δw is negative), then the candidate spectrum target is excluded. Only those candidate spectrum targets whose motion trend is consistent with the trend shown by the current new data can be retained in the final "candidate target queue". This step is the core of solving the problem of confusion between targets with similar azimuths, such as... Figure 3 As shown, this invention successfully distinguishes traditional methods through trend consistency verification. Figure 2 The confused trajectory shown in the figure.

[0034] S33. Target Selection Steps: Make a final decision on the "candidate target queue" after rigorous trend verification. If the number of targets in the queue is greater than 1, this indicates that there are still multiple candidate spectrum targets that are similar to the spectrum targets corresponding to the spectrum monitoring data to be processed in terms of both orientation and movement trend. In this case, the optimization decision strategy of "minimum orientation deviation" is adopted, and the candidate spectrum target with the smallest deviation is selected as the "final associated target". M_new is added to the historical sequence of candidate spectrum targets to complete the update of the target trajectory. If the number of targets in the queue is equal to 1: This is the ideal case, and M_new is directly associated with the unique candidate spectrum target.

[0035] If the number of targets in the queue is 0, it means that the new data cannot be associated with any existing targets, and it is likely to represent a new radiation source. In this case, the system will create a "new candidate spectrum target" starting from M_new, according to the spectrum target generation rules (such as initializing a trajectory sequence), and store it in the buffer.

[0036] Regardless of the decision outcome, the updated or newly created spectrum targets will be stored back in the cache for use in the next round of data processing.

[0037] S4. Loop Processing Phase: The system continuously repeats steps S2 and S3. Whenever new real-time spectrum monitoring data is input, a new round of classification, association, and update processes is immediately triggered. This process continues to loop, thereby achieving real-time, dynamic, and continuous tracking and situational awareness building of spectrum targets in the electromagnetic environment. When the monitoring task ends and no new data is input, the entire generation process concludes.

[0038] In summary, this embodiment significantly improves the accuracy and reliability of spectrum target generation in complex scenarios, especially when multiple targets are close in direction and their trajectories intersect, through a progressive processing chain of "feature extraction - classification - coarse association (multi-dimensional threshold filtering) - fine association (motion trend verification) - optimization decision-making," providing core technical support for generating high-quality spectrum situational awareness.

[0039] In some embodiments, the preset spectrum target processing rule is as follows: the spectrum target feature data in the cache and the spectrum monitoring data to be processed must simultaneously meet the preset deviation threshold rule in three dimensions: signal outgoing time, signal frequency, and signal outgoing direction.

[0040] In some embodiments, the deviation threshold rules include the following: the time continuity deviation of the signal outgoing time is less than the time deviation threshold; the signal frequency stability deviation is less than the frequency deviation threshold; and the signal outgoing azimuth continuity deviation is less than the azimuth deviation threshold.

[0041] In some embodiments, in step S32, the motion trend consistency verification step, the motion trend is analyzed based on the angle rotation information to determine whether the motion trend is consistent.

[0042] In some embodiments, the preset decision-making strategy is: If the number of targets in the candidate target queue is greater than 1, the candidate spectrum target with the smallest azimuth deviation between the spectrum target corresponding to the spectrum monitoring data to be processed is selected as the final associated target. If the number of targets in the candidate target queue is equal to 1, then the only candidate spectrum target in the candidate target queue will be used as the final associated target. If the number of targets in the candidate target queue is 0, a new candidate spectrum target is created based on the spectrum monitoring data to be processed, according to the spectrum target generation rules.

[0043] In some embodiments, the preset spectrum target processing rules are set based on expert experience.

[0044] The above description is merely a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the forms disclosed herein and should not be construed as excluding other embodiments. It can be used in various other combinations, modifications, and environments, and can be altered within the scope of the concept described herein through the above teachings or related technologies or knowledge. Modifications and variations made by those skilled in the art that do not depart from the spirit and scope of the present invention should be within the protection scope of the appended claims.

Claims

1. A method for generating spectrum targets based on real-time spectrum monitoring data, characterized in that: Includes the following steps: S1. Data Processing and Feature Extraction Stage: Receive real-time spectrum monitoring and reconnaissance data, parse and extract the feature parameters of the real-time spectrum monitoring and reconnaissance data, and establish structured metadata; S2, Signal System Classification Processing Stage: Based on the signal system identification results in the feature parameters, the real-time spectrum monitoring data is classified, and different spectrum target generation services are provided for real-time spectrum monitoring data of different signal systems based on the classification results; S3. Dynamic spectrum target generation and update stage: Based on the candidate spectrum targets generated in the current cache, according to the preset spectrum target processing rules, the spectrum monitoring data to be processed is associated with an existing candidate spectrum target or a new candidate spectrum target is created based on the spectrum monitoring data to be processed, and the update result is stored in the cache. S4. Loop Processing Phase: Continuously acquire the latest real-time spectrum monitoring data and repeat steps S2 to S3 to generate spectrum targets until no new data is input and the generation ends.

2. The spectrum target generation method based on real-time spectrum monitoring data according to claim 1, characterized in that: The characteristic parameters include at least: signal frequency, signal transmission time, signal transmission location, and signal system identification result.

3. The spectrum target generation method based on real-time spectrum monitoring data according to claim 1, characterized in that: The S3 dynamic spectrum target generation and update stage also includes the following steps: S31. Candidate target screening step: Compare the spectrum monitoring data to be processed with the spectrum target feature data in the cache, and screen out the candidate spectrum targets in the cache that meet the preset spectrum target processing rules; S32. Motion trend consistency verification step: Analyze the motion trend of the spectrum target corresponding to the spectrum monitoring data to be processed and the selected candidate spectrum targets. If the motion trends are consistent, add the candidate spectrum target to the candidate target queue. S33. Target selection step: Based on the preset decision strategy, determine the final associated target from the verified candidate target queue or create a new candidate spectrum target according to the spectrum target generation rules.

4. The spectrum target generation method based on real-time spectrum monitoring data according to claim 3, characterized in that: The preset spectrum target processing rule is as follows: the spectrum target feature data in the cache and the spectrum monitoring data to be processed must simultaneously meet the preset deviation threshold rules in three dimensions: signal outgoing time, signal frequency, and signal outgoing direction.

5. The spectrum target generation method based on real-time spectrum monitoring data according to claim 4, characterized in that: The aforementioned deviation threshold rules include the following: the time continuity deviation of the signal transmission time is less than the time deviation threshold; the signal frequency stability deviation is less than the frequency deviation threshold; and the signal transmission azimuth continuity deviation is less than the azimuth deviation threshold.

6. The spectrum target generation method based on real-time spectrum monitoring data according to claim 3, characterized in that: In step S32, the motion trend consistency verification step, the motion trend is analyzed based on the angle rotation information to determine whether the motion trend is consistent. The angle rotation information is angular velocity. When the positive / negative signs of the angular velocities are the same, the motion trend is considered to be consistent.

7. The spectrum target generation method based on real-time spectrum monitoring data according to claim 3, characterized in that: The preset decision-making strategy is as follows: If the number of targets in the candidate target queue is greater than 1, the candidate spectrum target with the smallest azimuth deviation between the spectrum target corresponding to the spectrum monitoring data to be processed is selected as the final associated target. If the number of targets in the candidate target queue is equal to 1, then the only candidate spectrum target in the candidate target queue will be used as the final associated target. If the number of targets in the candidate target queue is 0, a new candidate spectrum target is created based on the spectrum monitoring data to be processed, according to the spectrum target generation rules.

8. The spectrum target generation method based on real-time spectrum monitoring data according to claim 1, characterized in that: The preset spectrum target processing rules are set based on expert experience.

Images