On-line monitoring and analyzing system and method for digital radio station

Abstract

The invention belongs to the technical field of wireless communication, and discloses a digital radio station on-line monitoring analysis system and method, and the system comprises a signal collection and processing module which synchronously collects electromagnetic signal data of a digital radio station in different frequency bands and different carrier systems, carries out the preprocessing of the collected electromagnetic signal data, and transmits the preprocessed electromagnetic signal data to a wireless communication module; a standardized time-frequency signal data set is generated through time-frequency transformation and feature mapping; the logic sequence generation module is used for discretizing the time-frequency signal data set, discretizing a continuous signal change process into symbol sequence units, forming a logic sequence through a preset constraint rule, and describing a logic path of a digital radio station state; the logic conflict detection module is used for receiving the logic sequence, calculating a mutual constraint relationship among different logic paths and judging whether a logic conflict exists or not; and the precision, the real-time performance and the reliability of the operation state monitoring of the digital radio station are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wireless communication, more particularly, the present application relates to a digital radio online monitoring and analyzing system and method. BACKGROUND

[0002] The existing digital radio online monitoring and analyzing system and method mainly have the following problems: With the rapid development of wireless communication technology, the demand for signal transmission and management of digital radio under multi-frequency and multi-carrier mode is increasing. In order to ensure the safety and reliability of digital radio communication, it is necessary to monitor and analyze electromagnetic signals in real time to identify potential interference sources, abnormal signals and communication abnormal states. Although the existing digital radio online monitoring and analyzing system has the ability of multi-frequency signal acquisition and feature extraction, it still has many shortcomings in complex signal environment.

[0003] The existing technology lacks global consistency analysis of signal state logical path in the process of monitoring electromagnetic signals under multi-frequency and multi-carrier mode. The system often only identifies features for a single signal channel, and it is difficult to reveal the logical relationship and potential conflicts between different signal paths. When there are superimposed signals or asynchronous modulation forms in multiple frequency bands, the traditional analysis method is difficult to accurately distinguish the constraint relationship between the main logical path and the branch path, which may lead to misjudgment of signal source identification or inconsistent event traceability results, thereby affecting the reliability of the overall analysis of the system.

[0004] Traditional methods often detect conflicts through symbol matching or path consistency comparison, which can only identify the node position where the conflict occurs, but cannot analyze the evolution process and the correlation state before and after the conflict in the time dimension. Because the formation reason and continuous influence of the conflict cannot be judged, the system often adopts the way of directly discarding or simply replacing the rules when dealing with abnormal states, which leads to the loss of logical information, the mutation of system state, and affects the continuity and credibility of the monitoring sequence. In the aspect of conflict correction, the existing scheme relies on artificial experience or preset rules, which cannot realize robust correction of abnormal states while ensuring data integrity.

[0005] In the process of abnormal detection of traditional digital radio online monitoring system, static threshold or single feature index is often used to determine signal abnormality, ignoring the continuous change characteristics of electromagnetic signals in time sequence dimension. Because the cooperative effect between signal state evolution rate and path divergence degree cannot be described, the system is difficult to identify complex phenomena such as parallel evolution, sudden branch or cross interference, which leads to the lag of response to potential abnormal state, and cannot realize predictive response in the early stage of abnormal formation.

[0006] In view of this, the present application proposes a digital radio online monitoring and analyzing system and method to solve the above problems. SUMMARY

[0007] In order to overcome the above-mentioned defects of the prior art, in order to achieve the above-mentioned purpose, the present application provides the following technical scheme: a digital radio online monitoring and analyzing system, comprising: A signal acquisition and processing module synchronously acquires electromagnetic signal data of the digital radio under different frequency bands and different carrier modes, pre-processes the acquired electromagnetic signal data, and generates a standardized time-frequency signal data set through time-frequency conversion and feature mapping; A logic sequence generation module discretizes the time-frequency signal data set, discretizes the continuous signal change process into a symbol sequence unit, forms a logic sequence through a preset constraint rule, and describes the logic path of the digital radio state; A logic conflict detection module receives the logic sequence, calculates the mutual constraint relationship between different logic paths, and judges whether there is a logic conflict; when detecting a logic conflict, an abnormal state fusion mechanism is called to correct the conflicting logic path and the context logic neighborhood, and generate an abnormal state confidence interval; An abnormal evolution monitoring module constructs a behavior chain of the digital radio operation according to the abnormal state confidence interval, calculates abnormal evolution potential through the extension rate and bifurcation intensity of the behavior chain, and automatically triggers an abnormal response when the abnormal evolution potential exceeds a preset abnormal evolution potential threshold.

[0008] Specifically, the method for synchronously acquiring electromagnetic signals of the digital radio under different frequency bands and different carrier modes comprises: A preset monitoring frequency spectrum range of the digital radio signal is segmented and scanned, and a dynamic bandwidth is divided, so as to divide the entire monitoring frequency spectrum range into different narrow-band subintervals; a tunable radio frequency front end is used to realize switching and synchronous listening of each frequency band; The modulation mode, carrier type and symbol rate of the electromagnetic signal in each frequency band are identified, the local reference clock and sampling rate are dynamically adjusted according to the identification result, and the time synchronization and amplitude-frequency response matching of the electromagnetic signal of different carrier modes are performed; The electromagnetic signals of different frequency bands are sampled in parallel through multiple radio frequency sampling channels, a unified clock distribution mechanism is used to keep the sampling time sequence of each channel consistent, so as to obtain electromagnetic signal data of different frequency bands with time synchronization characteristics.

[0009] Specifically, the method for generating a standardized time-frequency signal data set comprises: The acquired electromagnetic signal data is subjected to noise suppression processing, including time domain filtering, frequency domain filtering or adaptive filtering, the amplitude of the electromagnetic signal data of each channel is normalized, the pre-processed electromagnetic signal data is converted into a time-frequency representation through a short-time Fourier transform method, and signal features are extracted from the time-frequency representation; The extracted signal features are subjected to mean normalization processing, realizing scale unification of signal features under different sampling times, frequency bands and carrier systems, to obtain a standardized signal feature vector; the standardized feature vector is organized in time sequence to obtain a standardized time-frequency signal data set.

[0010] Specifically, the acquisition method of the symbol sequence unit comprises: The standardized signal feature vectors arranged in time sequence in the standardized time-frequency signal data set are mapped into discrete symbol units by a threshold segmentation method, wherein the threshold segmentation method is to divide the feature value interval into different discrete levels according to the amplitude of the feature value of the standardized signal feature vector, and each discrete level corresponds to a unique discrete symbol; each discrete symbol is arranged in time sequence to form a symbol sequence unit describing signal changes.

[0011] Specifically, the acquisition method of the logic sequence comprises: The symbol sequence unit obtained by discretization is used to generate a logic sequence according to a preset constraint rule, and the preset constraint rule comprises a legal transition relationship of symbol state nodes in the symbol sequence unit, a transition probability threshold, a time duration condition and a context dependency relationship of symbol combinations. In the logic sequence generation process, the state transition legality of the symbol sequence unit is verified in sequence, a logic transition edge is established for a symbol state node satisfying the preset constraint rule, and the symbol state node not satisfying the rule is marked as an abnormal or uncertain state; finally, the symbol state nodes and legal transition relationships satisfying the constraint rule are connected in time sequence to form a logic sequence describing the state evolution of the digital radio station, which depicts the logic path of the digital radio station state.

[0012] Specifically, the method for calculating the mutual constraint relationship between different logic paths comprises: The logic sequence is received, all possible logic paths in the logic sequence are traversed to obtain a logic path set, and the local logic consistency index of any two logic paths in the logic path set is calculated; when the local logic consistency index is less than a preset local logic consistency index threshold and the preset constraint rule of the logic paths is contradictory, a logic conflict is determined to occur.

[0013] Specifically, the method for generating an abnormal state trusted interval comprises: When a logic conflict is detected, the neighborhood logic state of each conflicting logic path in the time sequence is identified as a context logic neighborhood; and the time neighborhood difference term is calculated for the time sequence of the logic path neighborhood of each conflicting logic path. The conflicting logic path and the context logic neighborhood are weighted and fused and overlapped and corrected by an abnormal state fusion function to obtain a fused logic path, and the fused logic path is defined as an abnormal state trusted interval.

[0014] Specifically, the method for constructing the behavior chain of the digital radio operation comprises: mapping the symbolic state nodes in the fused logical path to the behavior chain nodes of the digital radio operation, each node representing the operation behavior of the digital radio in a preset time slice; establishing directed edges between the behavior chain nodes according to the time sequence and legal transition relationship of the nodes in the fused logical path, the direction of the edge representing the causal order of the operation behaviors of the digital radio; and forming the behavior chain of the digital radio operation by merging the nodes and the directed edges for different fused logical paths.

[0015] Specifically, the method for automatically triggering the abnormal response comprises: calculating the state variables of adjacent nodes before and after each node arranged in time sequence in the behavior chain, the state variables including symbolic states or logical states; and dividing the state variables by the time interval to obtain the behavior extension amplitude in unit time, i.e. the extension rate of the behavior chain; For the nodes with branches in the behavior chain, identifying all feasible logical transition directions corresponding to the nodes in the fused logical path to form a branch logical path set; calculating the logical state difference value of each branch logical path; and the logical state difference value being the difference amplitude of the logical states of the branch logical path and the main logical path in the same time slice; statistically counting the number of transitions of the nodes to each branch direction, dividing the transition number of any branch logical path by the total sum of the transition numbers of all branches of the node to obtain the occurrence probability of the branch path; and summing the logical state difference values of the branch logical paths by weighting according to the occurrence probability of the branch path to obtain the bifurcation intensity of the node; calculating the abnormal evolution potential energy by the extension rate and the bifurcation intensity of the behavior chain, and automatically triggering the abnormal response when the abnormal evolution potential energy exceeds a preset abnormal evolution potential energy threshold.

[0016] An online monitoring and analysis method for a digital radio comprises: S1, synchronously collecting electromagnetic signals of the digital radio under different frequency bands and different carrier systems, pre-processing the collected electromagnetic signals, and generating a standardized time-frequency signal dataset through time-frequency transformation and feature mapping; S2, performing feature extraction and discretization processing on the time-frequency signal dataset, discretizing the continuous signal change process into symbolic sequence units, forming logical sequences through preset constraint rules, and depicting logical paths of the digital radio state; S3, receiving the logical sequences, calculating the mutual constraint relationship between different logical paths; when detecting a logical conflict, calling an abnormal state fusion mechanism, performing overlapping correction on the conflicting logical path and the context logical neighborhood, and generating an abnormal state credible interval; S4, according to the abnormal state confidence interval, the behavior chain of the digital radio operation is constructed, the abnormal evolution potential is calculated through the extension rate and bifurcation intensity of the behavior chain, and when the abnormal evolution potential exceeds the preset abnormal evolution potential threshold, an abnormal response is automatically triggered.

[0017] Compared with the prior art, the present application has the following beneficial effects: Through the quantitative calculation of the consistency index, the system can quantify the overlapping relationship and difference degree between different logical paths, and realize the accurate measurement of the multi-path logic consistency. The detection of the logic conflict is based on the set relationship of the symbolic node, and does not depend on a single signal feature, which can effectively reduce the false judgment caused by noise interference. The running state of the digital radio is abstracted as a logic sequence and the set operation relationship is introduced, so that the system can still maintain a unified evaluation standard for the logic consistency in a multi-band and different system environment. When the logic conflict occurs, the system can call the abnormal state fusion mechanism in real time for correction, realize the self-healing dynamic adjustment of the logic layer, and thus improve the intelligence and reliability of the online monitoring of the digital radio.

[0018] By setting the time window before and after the conflict node, the symbolic state node and its transition relationship, transition probability and duration information are extracted, and the context logic neighborhood is constructed, so that the system has a time association awareness when processing the logic conflict, and can identify the causes and evolution trend of the conflict. By calculating the state difference of the conflict path and the neighborhood path in the time sequence, the time neighborhood difference term is formed, the smoothing correction of the logic path in the time dimension is realized, the abnormal fluctuation is corrected through the neighborhood difference, the self-healing recovery of the logic layer anomaly is realized, and the continuity and reliability of the evolution sequence of the digital radio running state are greatly improved.

[0019] By calculating the rate of change of the state variable of each node in the behavior chain with time, the behavior extension amplitude per unit time is obtained, so that the system can quantitatively describe the speed and trend of the digital radio state evolution, thereby reflecting the dynamic characteristics of the energy flow of the system. The extension rate and bifurcation intensity of the behavior chain are coupled and fused, and when the comprehensive potential exceeds the set threshold, the system can automatically trigger an abnormal response, realize early warning of potential anomalies, and improve the safety and response efficiency of the digital radio operation. BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1 It is a structure schematic diagram of a digital radio online monitoring and analysis system of the present application; Figure 2 It is a flowchart of a digital radio online monitoring and analysis method of the present application. DETAILED DESCRIPTION

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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. Example

[0022] Please see Figure 1 As shown, this embodiment provides an online monitoring and analysis system for digital radio stations, specifically including the following steps: The signal acquisition and processing module synchronously acquires electromagnetic signal data from digital radio stations under different frequency bands and different carrier systems, preprocesses the acquired electromagnetic signal data, and generates a standardized time-frequency signal dataset through time-frequency transformation and feature mapping. The logic sequence generation module discretizes the time-frequency signal dataset, discretizing the continuous signal change process into symbol sequence units, and forming a logic sequence through preset constraint rules to characterize the logical path of the digital radio station state. The logic conflict detection module receives a logic sequence, calculates the mutual constraints between different logic paths, and determines whether a logic conflict exists. When a logic conflict is detected, it calls the heterogeneous fusion mechanism to overlap and correct the conflicting logic path with the context logic neighborhood, generating a heterogeneous reliable interval. The abnormal evolution monitoring module constructs a behavior chain for the operation of the digital radio station based on the anomaly confidence interval. It calculates the abnormal evolution potential energy through the extension rate and bifurcation strength of the behavior chain. When the abnormal evolution potential energy exceeds the preset abnormal evolution potential energy threshold, it automatically triggers an abnormal response.

[0023] Methods for synchronously acquiring electromagnetic signals from digital radio stations under different frequency bands and carrier standards include: The monitoring spectrum range of the digital radio signal is preset, and the monitoring spectrum range is segmented and dynamically divided into bandwidths, dividing the entire monitoring spectrum range into different narrow band sub-intervals; the switching and synchronous monitoring of each frequency band is realized by using a tunable radio frequency front end. The modulation method, carrier type and symbol rate of electromagnetic signals are identified in each frequency band. The local reference clock and sampling rate are dynamically adjusted according to the identification results. Time synchronization and amplitude-frequency response matching are performed for electromagnetic signals of different carrier systems. Furthermore, electromagnetic signals of different frequency bands are sampled in parallel through multiple radio frequency sampling channels, and a unified clock allocation mechanism is used to keep the sampling timing of each channel consistent, thereby obtaining electromagnetic signal data of different frequency bands with time synchronization characteristics.

[0024] It should be noted that the process of maintaining consistent sampling timing across channels using a unified clock distribution mechanism involves setting a highly stable master clock source (e.g., a temperature-compensated crystal oscillator, OCXO) in the acquisition system as the time reference for the entire system. The master clock signal is then distributed to the analog-to-digital converters of multiple RF sampling channels via a low-phase-noise clock distributor or phase-locked loop module, ensuring that each channel receives a reference clock signal with the same frequency and phase.

[0025] Methods for generating standardized time-frequency signal datasets include: The acquired electromagnetic signal data is subjected to noise suppression processing, including time-domain filtering, frequency-domain filtering, or adaptive filtering, to remove environmental noise, interference signals, and high-frequency jitter introduced by radio frequency devices; the electromagnetic signal data of each channel is normalized to eliminate the influence of amplitude differences between different channels or different frequency bands. The preprocessed electromagnetic signal data is converted into a time-frequency representation using the short-time Fourier transform method to characterize the instantaneous frequency changes and power distribution of the electromagnetic signal data; signal features are extracted from the time-frequency representation, including instantaneous amplitude, instantaneous phase, energy spectral density, frequency center, and bandwidth; The extracted signal features are normalized by mean to achieve scale uniformity of signal features under different sampling times, frequency bands and carrier systems, resulting in standardized signal feature vectors. The standardized feature vectors are then organized in chronological order to obtain a standardized time-frequency signal dataset.

[0026] It should be noted that extracting signal features from the time-frequency representation includes: calculating the complex modulus as the instantaneous amplitude and the complex phase angle as the instantaneous phase for the complex signal data in the time-frequency representation; summing or averaging the squares of the signal amplitude to obtain the energy distribution of each time slice or frequency slice, thereby obtaining the energy spectral density; calculating the center frequency of the signal in each time slice based on power weighting, as the frequency center; and calculating the bandwidth of the signal based on the variance of the power distribution.

[0027] The mapped signal features are normalized by mean to achieve scale uniformity of signal features under different sampling times, frequency bands and carrier systems, resulting in standardized signal feature vectors. The standardized feature vectors are then organized in chronological order to obtain a standardized time-frequency signal dataset.

[0028] Methods for obtaining symbol sequence units include: The threshold segmentation method maps the standardized signal feature vectors arranged in time order in the standardized time-frequency signal dataset into discrete symbol units. The threshold segmentation method divides the feature value interval into different discrete levels according to the amplitude of the feature values ​​of the standardized signal feature vectors, and each discrete level corresponds to a unique discrete symbol. Each discrete symbol is arranged in time order to form a symbol sequence unit describing the signal changes.

[0029] It should be noted that the threshold segmentation method is as follows: statistical analysis is performed on the amplitude distribution of the standardized signal feature vector to determine the full amplitude range of the feature values; the segmentation threshold is determined according to the preset number of discrete levels, and the feature value interval is divided into different discrete levels using an equidistant division method.

[0030] Methods for obtaining logical sequences include: For the discretized symbol sequence units, a logical sequence is generated according to preset constraint rules. The preset constraint rules include the legal transition relationship of symbol state nodes in the symbol sequence unit (a symbol can only transition to a specific symbol (determined according to digital radio protocol or historical behavior learning)), the transition probability threshold (a symbol whose transition probability is lower than the preset transition probability threshold is judged as untrustworthy), the time duration condition and the contextual dependency relationship of symbol combination (the order or combination of certain symbols must satisfy the context rules (e.g., a carrier establishment signal must appear before the synchronization signal)). During the generation of the logical sequence, the legality of the state transitions of the symbol sequence units is verified in turn. Logical transition edges are established for symbol state nodes that meet the preset constraint rules, and symbol state nodes that do not meet the rules are marked as abnormal or uncertain states. Finally, the symbol state nodes that meet the constraint rules and the legal transition relationships are connected in chronological order to form a logical sequence describing the state evolution of the digital radio station and characterizing the logical path of the digital radio station state.

[0031] Methods for calculating the mutual constraints between different logical paths include: Receive a logical sequence, traverse all possible logical paths in the logical sequence to obtain a set of logical paths; for any two logical paths in the set of logical paths, calculate the local logical consistency index. The local logical consistency metric is: ;in, This represents a local logical consistency index for any two logical paths, with a value ranging from 0 to 1; if This indicates that the two logical paths are completely identical. This indicates that the two logical paths do not overlap at all; Represents the first in the logical path set One logical path; Represents the first in the logical path set One logical path; Indicates two logical paths and The number of common symbol nodes (intersection size) measures the number of nodes that are exactly the same on the symbol sequence of two logical paths, that is, the consistent part of the logical paths; Indicates two logical paths and The number of symbolic nodes in the union of two logical paths measures the total number of all distinct symbolic nodes contained in the two logical paths. When the local logic consistency index is less than the preset local logic consistency index threshold, and the preset constraint rules of the logic path are contradictory (such as the same state node having different transition behaviors on two paths), a logic conflict is determined to have occurred.

[0032] This solution addresses the following technical problems in existing technologies: Existing technologies, in the monitoring of multi-band, multi-carrier electromagnetic signals, lack global consistency analysis of signal state logic paths, failing to identify potential logical contradictions or conflicts between different signal paths. This leads to misjudgments or logical breaks during signal feature analysis and pattern recognition. Traditional signal logic analysis methods often employ single-path or linear reasoning. When multiple superimposed signals or asynchronous modulation forms exist within the same frequency band, the system struggles to distinguish the constraint relationships between the main logic path and branch paths, affecting the accuracy of signal source identification and the reliability of event tracing. Furthermore, the lack of quantitative standards for path conflict detection and correction makes it difficult to determine the degree of overlap and consistency of different logic paths through unified indicators, hindering the system's ability to automatically identify conflict sources during abnormal state identification.

[0033] Compared to existing technologies, the beneficial effects are as follows: By quantitatively calculating consistency indicators, the system can quantify the overlap and differences between different logical paths, achieving accurate measurement of multi-path logical consistency. Logical conflict detection is based on the set relationship of symbol nodes, not relying on single signal features, effectively reducing misjudgments caused by noise interference. The system abstracts the operating state of a digital radio into a logical sequence and introduces set operation relationships, enabling the system to maintain a unified evaluation standard for logical consistency even in multi-band and heterogeneous environments. When logical conflicts occur, the system can call the heterogeneous fusion mechanism in real time for correction, achieving self-healing dynamic adjustment of the logical layer, thereby improving the intelligence and reliability of online monitoring of digital radios.

[0034] Methods for generating heterogeneous confidence intervals include: When a logical conflict is detected, the neighborhood logical state of each conflicting logical path in the time series is identified as the context logical neighborhood; The specific operation is as follows: For each conflict node in the conflict logic path, set the preceding and following time windows along the time axis, extract all symbol state nodes and their corresponding state transition relationships, transition probabilities and time duration information within the window, and use these nodes and their interrelationships as the neighborhood logical states of the conflict node, thereby constructing the context logical neighborhood of the conflict node.

[0035] For each conflicting logical path, calculate the time neighborhood difference term in the time series of the logical path's neighborhood. It should be noted that the time neighborhood difference term here is not simply a difference for a single path, but rather a calculation of the difference in changes on the time axis for the corresponding nodes of the conflicting logical path and its neighborhood logical state in the time series. A time window is set before and after the conflict node. The symbolic state nodes and their logical transition relationships of the conflict path within the window are extracted to obtain the neighborhood logical state sequence. The states of the conflict path node and the neighboring nodes at the same time point are compared to obtain the state difference or rate of change. This difference value is the time neighborhood difference term.

[0036] By using a heteromorphic fusion function, conflicting logical paths are weighted and fused with the contextual logical neighborhood, and overlap correction is performed to obtain the fused logical path. The fused logical path represents the corrected and robust digital radio station state evolution sequence, and the fused logical path is defined as the abnormal confidence interval.

[0037] The heteromorphic fusion function is: ;in, Indicates the logical path after merging; This represents the credibility weight coefficient, used to balance the influence of the two logical paths in the fusion result. When it is large, Dominant in the final fusion result, when When smaller, Dominant; This represents the temporal neighborhood weight coefficient, used to control the influence of the temporal neighborhood difference term in the fusion process; Represents the temporal neighborhood difference term, used to smooth out temporal inconsistencies in conflicting logical paths; Represents a local time variable; This approach addresses the following technical problems of existing technologies: Traditional conflict detection often relies on symbol matching or path consistency comparison, only identifying the nodes where conflicts occur but failing to analyze the temporal context of the conflict. This prevents the system from accurately determining the causes and ongoing impact of the conflict. Once a logical conflict is detected, traditional methods typically discard the abnormal path or perform simple rule substitution, resulting in the loss of logical information or abrupt changes in system state, making it difficult to guarantee the continuity and reliability of the monitoring sequence. Existing methods often rely on empirical parameters or manual judgment to handle conflict paths, lacking a unified mathematical model to balance the influence between the original path and the neighboring state, making it difficult to achieve robust anomaly correction while ensuring data integrity.

[0038] Compared to existing technologies, the advantages are as follows: By setting time windows before and after conflict nodes, extracting symbolic state nodes and their transition relationships, transition probabilities, and duration information, and constructing a contextual logical neighborhood, the system gains a time-related awareness when handling logical conflicts, enabling it to identify the causes, consequences, and evolution trends of conflicts. By calculating the state differences between the conflict path and the neighborhood path in the time series, forming a time neighborhood difference term, the logical path is smoothed and corrected in the time dimension. Abnormal fluctuations are corrected through neighborhood difference, achieving self-healing recovery of logical layer anomalies, and significantly improving the continuity and reliability of the digital radio station's operational state evolution sequence.

[0039] Methods for constructing the behavior chain of digital radio operation include: The symbolic state nodes in the merged logical path are mapped to the behavior chain nodes of the digital radio station. Each node represents the operation behavior of the digital radio station in a preset time slice. Based on the time order and legal transition relationship of the nodes in the merged logical path, directed edges are established between the behavior chain nodes. The direction of the edge represents the causal order of the digital radio station's operation behavior. For different merged logical paths, the behavior chain of the digital radio station is formed by merging nodes and directed edges.

[0040] Methods for automatically triggering exception responses include: For each node in the behavior chain arranged in chronological order, calculate the state variables of the nodes before and after that node. The state variables include symbolic states or logical states. Divide the state variables by the time interval to obtain the behavior extension amplitude per unit time, i.e., the extension rate of the behavior chain. It should be noted that if the state variable is a symbolic state, the magnitude of change can be calculated through encoding (such as mapping the symbol to a vector or probability distribution); if the state variable is a logical state, the magnitude of change can be calculated through logical differences (such as changes in logical expressions, the number of constraint contradictions, etc.).

[0041] For nodes with branches in the behavior chain, identify all feasible logical transition directions corresponding to the node in the fused logical path to form a set of branch logical paths; for each branch logical path, calculate the logical state difference value of the branch logical path; the logical state difference value is the difference magnitude between the logical state of the branch logical path and the main logical path in the same time slice. The number of times a node moves to each branch direction is counted. The probability of a branch path occurring is obtained by dividing the number of moves of any branch logical path by the total number of moves of all branches of that node. The logical state difference values ​​of each branch logical path are weighted and summed according to the probability of occurrence of that branch path to obtain the bifurcation strength of the node. The abnormal evolution potential energy is calculated by the extension rate and bifurcation strength of the behavioral chain. When the abnormal evolution potential energy exceeds the preset abnormal evolution potential energy threshold, an abnormal response is automatically triggered.

[0042] The anomalous evolution potential energy is: ;in, Represents a behavior chain node At the point of time The abnormal evolutionary potential energy; Indicates the behavioral chain at a given time point. Nodes; Indicates the behavior chain node in time The rate of energy change, i.e. the rate of extension of the behavioral chain; Indicates the behavior chain node in time The bifurcation strength; This represents the behavior extension weight coefficient, used to adjust the contribution of the rate of change of the state of the behavior chain nodes (behavior extension amplitude) to the potential energy of abnormal evolution. This represents the bifurcation strength weighting coefficient, used to adjust the contribution of the bifurcation strength of behavioral chain nodes to the anomalous evolution potential energy; Represented as an integral variable, it represents the time from time 0 to time point 1. Historical time series; This represents the upper limit of the integral, and this represents the current time point in the evolution of the behavior chain.

[0043] This solution addresses the following technical problems of existing technologies: Traditional online monitoring of digital radio stations relies heavily on static thresholds or single characteristic indicators to determine signal anomalies, failing to characterize the continuous variation trend of electromagnetic signals under multi-band and multi-carrier conditions from a time-series perspective, resulting in a delayed response to potential anomalies. During digital radio monitoring, different signal paths may exhibit parallel evolution, sudden branching, or cross-interference. Existing solutions struggle to identify the differences in these branch logics and their impact, leading to the system's inability to accurately distinguish between the main signal and interference signals. Traditional anomaly detection focuses only on abrupt changes in characteristic values, neglecting the synergistic effect between the signal state evolution rate and the degree of path divergence, thus failing to provide predictive responses in the early stages of signal anomaly formation.

[0044] The advantages over existing technologies are as follows: By calculating the rate of change of the state variables of each node in the behavior chain over time, the behavioral extension amplitude per unit time is obtained, enabling the system to quantitatively describe the speed and trend of digital radio state evolution, thereby reflecting the dynamic characteristics of system energy flow. By coupling and integrating the extension rate of the behavior chain with the bifurcation strength, the system can automatically trigger an abnormal response when the combined potential energy exceeds a set threshold, achieving early warning of potential anomalies and improving the safety and response efficiency of digital radio operation.

[0045] The preset abnormal evolution potential energy threshold is set by staff. By collecting different abnormal evolution potential energies, the average value of multiple abnormal evolution potential energies is taken as the preset abnormal evolution potential energy threshold. Similarly, the preset local logic consistency index threshold is set.

[0046] In this embodiment, by quantifying consistency indicators, the system can quantify the overlap and differences between different logical paths, achieving accurate measurement of multi-path logical consistency. Logical conflict detection is based on the set relationship of symbol nodes, not relying on a single signal feature, effectively reducing misjudgments caused by noise interference. The operating state of the digital radio is abstracted into a logical sequence and set operation relationships are introduced, enabling the system to maintain a unified evaluation standard for logical consistency even in multi-band and heterogeneous environments. When a logical conflict occurs, the system can call the heterogeneous fusion mechanism in real time for correction, achieving self-healing dynamic adjustment of the logical layer, thereby improving the intelligence and reliability of online monitoring of digital radios.

[0047] By setting time windows before and after conflict nodes, symbolic state nodes and their transition relationships, transition probabilities, and durations are extracted to construct a contextual logical neighborhood. This enables the system to have temporal awareness when handling logical conflicts, allowing it to identify the causes, consequences, and evolution trends of conflicts. By calculating the state differences between the conflict path and the neighborhood path in the time series, a time neighborhood difference term is formed, achieving smoothing correction of the logical path in the time dimension. Abnormal fluctuations are corrected through neighborhood difference, enabling self-healing recovery of logical layer anomalies and significantly improving the continuity and reliability of the digital radio station's operational state evolution sequence.

[0048] By calculating the rate of change of the state variables of each node in the behavior chain over time, the behavioral extension amplitude per unit time is obtained, enabling the system to quantitatively describe the speed and trend of digital radio state evolution, thereby reflecting the dynamic characteristics of system energy flow. By coupling and integrating the extension rate of the behavior chain with the bifurcation strength, the system can automatically trigger an abnormal response when the combined potential energy exceeds a set threshold, achieving early warning of potential anomalies and improving the safety and response efficiency of digital radio operation. Example

[0049] Please see Figure 2 As shown, parts not described in detail in this embodiment are described in Embodiment 1. A method for online monitoring and analysis of digital radio stations is provided, including: S1. Synchronously collect electromagnetic signals from digital radio stations under different frequency bands and different carrier systems, preprocess the collected electromagnetic signals, and generate a standardized time-frequency signal dataset through time-frequency transformation and feature mapping. S2. Perform feature extraction and discretization on the time-frequency signal dataset, discretize the continuous signal change process into symbol sequence units, form a logical sequence through preset constraint rules, and characterize the logical path of the digital radio station state. S3. Receive the logical sequence and calculate the mutual constraints between different logical paths. When a logical conflict is detected, call the heterogeneous fusion mechanism to overlap and correct the conflicting logical path with the context logical neighborhood to generate a heterogeneous reliable interval. S4. Based on the anomaly confidence interval, construct the behavior chain of the digital radio operation, calculate the abnormal evolution potential energy through the extension rate and bifurcation strength of the behavior chain, and automatically trigger the abnormal response when the abnormal evolution potential energy exceeds the preset abnormal evolution potential energy threshold.

[0050] Since the electronic device described in this embodiment is the electronic device used to implement the digital radio online monitoring and analysis system and method described in this application embodiment, those skilled in the art can understand the specific implementation method and various variations of the electronic device in this embodiment based on the digital radio online monitoring and analysis system and method described in this application embodiment. Therefore, how the electronic device implements the method in this application embodiment will not be described in detail here. Any electronic device used by those skilled in the art to implement the digital radio online monitoring and analysis system and method described in this application embodiment falls within the scope of protection of this application.

[0051] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0052] The above description is merely a preferred embodiment of the present invention, and the scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for users of ordinary technical skills, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A digital radio station online monitoring and analysis system, characterized in that, include: The signal acquisition and processing module synchronously acquires electromagnetic signal data from digital radio stations under different frequency bands and different carrier systems, preprocesses the acquired electromagnetic signal data, and generates a standardized time-frequency signal dataset through time-frequency transformation and feature mapping. The logic sequence generation module discretizes the time-frequency signal dataset, discretizing the continuous signal change process into symbol sequence units, and forming a logic sequence through preset constraint rules to characterize the logical path of the digital radio station state. The logic conflict detection module receives a logic sequence, calculates the mutual constraints between different logic paths, and determines whether there is a logic conflict. When a logical conflict is detected, the heterogeneous fusion mechanism is invoked to overlap and correct the conflicting logical path with the context logical neighborhood, generating a heterogeneous reliable interval. The abnormal evolution monitoring module constructs a behavior chain for the operation of the digital radio station based on the anomaly confidence interval. It calculates the abnormal evolution potential energy through the extension rate and bifurcation strength of the behavior chain. When the abnormal evolution potential energy exceeds the preset abnormal evolution potential energy threshold, it automatically triggers an abnormal response.

2. The digital radio station online monitoring and analysis system according to claim 1, characterized in that, The method for synchronously acquiring electromagnetic signals from digital radio stations under different frequency bands and different carrier systems includes: The monitoring spectrum range of the digital radio signal is preset, and the monitoring spectrum range is segmented and dynamically divided into bandwidths, dividing the entire monitoring spectrum range into different narrow band sub-intervals; the switching and synchronous monitoring of each frequency band is realized by using a tunable radio frequency front end. The modulation method, carrier type and symbol rate of electromagnetic signals are identified in each frequency band. The local reference clock and sampling rate are dynamically adjusted according to the identification results. Time synchronization and amplitude-frequency response matching are performed for electromagnetic signals of different carrier systems. Furthermore, electromagnetic signals of different frequency bands are sampled in parallel through multiple radio frequency sampling channels, and a unified clock allocation mechanism is used to keep the sampling timing of each channel consistent, thereby obtaining electromagnetic signal data of different frequency bands with time synchronization characteristics.

3. The digital radio station online monitoring and analysis system according to claim 2, characterized in that, The method for generating standardized time-frequency signal datasets includes: The acquired electromagnetic signal data is subjected to noise suppression processing, including time-domain filtering, frequency-domain filtering or adaptive filtering. The amplitude of the electromagnetic signal data of each channel is normalized. The preprocessed electromagnetic signal data is converted into time-frequency representation through short-time Fourier transform method, and signal features are extracted from the time-frequency representation. The extracted signal features are normalized by mean to achieve scale uniformity of signal features under different sampling times, frequency bands and carrier systems, resulting in standardized signal feature vectors. The standardized feature vectors are then organized in chronological order to obtain a standardized time-frequency signal dataset.

4. The digital radio station online monitoring and analysis system according to claim 3, characterized in that, The method for obtaining the symbol sequence unit includes: The threshold segmentation method maps the standardized signal feature vectors arranged in time order in the standardized time-frequency signal dataset into discrete symbol units. The threshold segmentation method divides the feature value interval into different discrete levels according to the amplitude of the feature values ​​of the standardized signal feature vectors, and each discrete level corresponds to a unique discrete symbol. Each discrete symbol is arranged in time order to form a symbol sequence unit describing the signal changes.

5. The digital radio station online monitoring and analysis system according to claim 4, characterized in that, The method for obtaining the logical sequence includes: For the discretized symbol sequence units, a logical sequence is generated according to preset constraint rules. The preset constraint rules include the legal transition relationship of the symbol state nodes in the symbol sequence unit, the transition probability threshold, the time duration condition, and the contextual dependency relationship of the symbol combination. During the generation of the logical sequence, the legality of the state transitions of the symbol sequence units is verified in turn. Logical transition edges are established for symbol state nodes that meet the preset constraint rules, and symbol state nodes that do not meet the rules are marked as abnormal or uncertain states. Finally, the symbol state nodes that meet the constraint rules and the legal transition relationships are connected in chronological order to form a logical sequence describing the state evolution of the digital radio station and characterizing the logical path of the digital radio station state.

6. The digital radio station online monitoring and analysis system according to claim 5, characterized in that, The method for calculating the mutual constraints between different logical paths includes: Receive a logical sequence, traverse all possible logical paths in the logical sequence to obtain a set of logical paths; for any two logical paths in the set of logical paths, calculate a local logical consistency index; when the local logical consistency index is less than a preset local logical consistency index threshold and the preset constraint rules of the logical paths contradict each other, a logical conflict is determined to have occurred.

7. The digital radio station online monitoring and analysis system according to claim 6, characterized in that, The method for generating the heterogeneous confidence interval includes: When a logical conflict is detected, the neighborhood logical state of each conflicting logical path in the time series is identified as the context logical neighborhood; for each conflicting logical path in the time series of the logical path neighborhood, the time neighborhood difference term is calculated. The conflicting logical paths are weighted and overlap-corrected with the context logical neighborhood by using a heterogeneous fusion function to obtain the fused logical path, which is then defined as the abnormal confidence interval.

8. The digital radio station online monitoring and analysis system according to claim 7, characterized in that, The method for constructing the behavior chain of digital radio operation includes: The symbolic state nodes in the merged logical path are mapped to the behavior chain nodes of the digital radio station. Each node represents the operation behavior of the digital radio station in a preset time slice. Based on the time order and legal transition relationship of the nodes in the merged logical path, directed edges are established between the behavior chain nodes. The direction of the edge represents the causal order of the digital radio station's operation behavior. For different merged logical paths, the behavior chain of the digital radio station is formed by merging nodes and directed edges.

9. The digital radio station online monitoring and analysis system according to claim 8, characterized in that, The method for automatically triggering an exception response includes: For each node in the behavior chain arranged in chronological order, calculate the state variables of the nodes before and after that node. The state variables include symbolic states or logical states. Divide the state variables by the time interval to obtain the behavior extension amplitude per unit time, i.e., the extension rate of the behavior chain. For nodes with branches in the behavior chain, identify all feasible logical transition directions corresponding to the node in the fused logical path to form a set of branch logical paths; for each branch logical path, calculate the logical state difference value of the branch logical path; the logical state difference value is the difference magnitude between the logical state of the branch logical path and the main logical path in the same time slice. The number of times a node moves to each branch direction is counted. The probability of a branch path occurring is obtained by dividing the number of moves of any branch logical path by the total number of moves of all branches of that node. The logical state difference values ​​of each branch logical path are weighted and summed according to the probability of occurrence of that branch path to obtain the bifurcation strength of the node. The abnormal evolution potential energy is calculated by the extension rate and bifurcation strength of the behavioral chain. When the abnormal evolution potential energy exceeds the preset abnormal evolution potential energy threshold, an abnormal response is automatically triggered.

10. A method for online monitoring and analysis of digital radio stations, implemented using the online monitoring and analysis system for digital radio stations as described in any one of claims 1 to 9, characterized in that, include: S1. Synchronously collect electromagnetic signals from digital radio stations under different frequency bands and different carrier systems, preprocess the collected electromagnetic signals, and generate a standardized time-frequency signal dataset through time-frequency transformation and feature mapping. S2. Perform feature extraction and discretization on the time-frequency signal dataset, discretize the continuous signal change process into symbol sequence units, form a logical sequence through preset constraint rules, and characterize the logical path of the digital radio station state. S3. Receive the logical sequence and calculate the mutual constraints between different logical paths; When a logical conflict is detected, the heterogeneous fusion mechanism is invoked to overlap and correct the conflicting logical path with the context logical neighborhood, generating a heterogeneous reliable interval. S4. Based on the anomaly confidence interval, construct the behavior chain of the digital radio operation, calculate the abnormal evolution potential energy through the extension rate and bifurcation strength of the behavior chain, and automatically trigger the abnormal response when the abnormal evolution potential energy exceeds the preset abnormal evolution potential energy threshold.

Images