Feature identification method for wireless communication digital signal

Abstract

The invention discloses a feature recognition method for a wireless communication digital signal, and relates to the field of wireless communication, and the method comprises the steps: further calculating an extracted feature vector F (t) through constructing a comprehensive quality algorithm formula, outputting a quality score Qs of the wireless communication signal, and carrying out the preliminary comparison and evaluation of the quality score Qs and a preset signal quality threshold value Q. And meanwhile, dynamic characteristics are introduced, secondary comprehensive calculation is carried out on the signal quality, and a second wireless communication signal quality score Qs2 is output. According to the optimization process, the dynamic change characteristics of the signals can be more comprehensively analyzed, the signals are accurately classified through the classifier, the signal quality is divided, an adjustment mechanism can be automatically triggered according to the signal quality grade, and the transmission efficiency is improved by means of adjusting the communication channel frequency, optimizing redundant coding, dynamically adjusting the transmission power and the like. The communication reliability and the anti-interference capability of the system in a complex environment are improved, and the safe operation of the rail transit system is effectively guaranteed.

Description

Technical Field

[0001] The present invention relates to the field of wireless communication technology, and in particular to a feature recognition method for wireless communication digital signals. Background Art

[0002] Wireless communication technology is the foundation of modern information transmission. It is widely used in many fields such as mobile communications, the Internet of Things, and intelligent transportation. In these application scenarios, the quality of wireless communication signals directly affects the stability and reliability of the system, especially in rapidly changing environments, such as intelligent rail transit systems. Signal transmission quality is crucial to the normal operation of the system. Specifically, in the Internet of Things devices in the intelligent transportation system, the wireless communication signals that transmit data in real time exhibit complex dynamic characteristics due to factors such as multipath effects and interference, which makes it difficult for traditional signal processing methods to cope with the nonlinear behavior of wireless signals. Therefore, how to accurately identify and evaluate the characteristics of these wireless communication digital signals has become a key issue in improving the quality of wireless communications.

[0003] At present, in existing wireless communication systems, although a variety of signal processing technologies are used to improve the transmission stability of signals, traditional signal processing methods are often unable to accurately capture the nonlinear characteristics of signals in complex environments such as multipath propagation, delay jitter, and channel interference. This results in the system being unable to promptly identify changes or anomalies in signal quality in practical applications, especially in high-speed changing environments such as rail transit, where instantaneous changes in signals cannot be effectively monitored and classified. Existing processing methods have poor adaptability in complex environments, have a high error rate during signal transmission, and are unable to dynamically adjust signal transmission parameters to cope with complex environmental changes; This situation is mainly due to the limitations of traditional methods in dealing with dynamic and nonlinear signal characteristics. With the increasing complexity of wireless communication signals and the demand for data transmission, existing methods are unable to maintain high-precision signal recognition and classification under the influence of multipath effects and phase noise, which leads to a decrease in signal transmission efficiency and an increase in data loss rate. When the system cannot accurately evaluate the signal quality, it will cause frequent interference during signal transmission, increase the bit error rate, and even cause communication interruption, which seriously affects the operation safety and efficiency of the intelligent rail transit system. Therefore, there is an urgent need for an innovative method that can accurately extract signal features and optimize signal recognition algorithms to effectively improve the robustness and adaptability of signal transmission. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a method for identifying features of wireless communication digital signals, which solves the problems mentioned in the background technology.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: A method for identifying characteristics of a wireless communication digital signal, comprising the following steps: S1. Collect wireless communication signal parameters through the Internet of Things devices in the intelligent rail transit system, pre-process the signal parameters, obtain a standard wireless communication signal parameter group, and perform a composite analysis on the standard wireless communication signal parameter group to obtain multi-dimensional time series data S(t); S2. Use the time delay embedding algorithm to map the multidimensional time series data S(t) to the multidimensional phase space, construct the multidimensional vector Y(t), and track each multidimensional vector Y(t) in the phase space to form the trajectory of the signal; S3, extracting digital signal features from the trajectory in the phase space, and then combining the acquired standard wireless communication signal parameters with the digital signal features to generate a feature vector F(t); S4. Construct a comprehensive quality algorithm formula, input the feature vector F(t) into the comprehensive quality algorithm formula, calculate and output the wireless communication signal quality Qs, and preset the signal quality threshold Q and the obtained wireless communication signal quality Qs, perform preliminary comparative evaluation, and analyze the current rail transit wireless communication digital signal quality status; S5. Based on the preliminary comparative evaluation results, dynamic features are introduced, and a second comprehensive calculation is performed to output a second wireless communication signal quality score Qs2, the recognition accuracy of the digital signal is optimized, and a second comparative evaluation is performed with the signal quality threshold Q to analyze the quality status of the optimized wireless communication digital signal; S6. Based on the extracted feature vector F(t), a classifier is constructed using the support vector machine SVM classification algorithm. The wireless communication digital signal is then classified according to the wireless communication signal quality Qs. The classification results are further refined into corresponding quality levels, and the trigger conditions of the adjustment mechanism are generated based on the quality levels to optimize and adjust the abnormal signals of the wireless communication digital signal.

[0006] Preferably, S1 includes S11, S12 and S13; S11. Real-time collection of wireless communication signal parameters through IoT devices in the intelligent rail transit system; IoT devices include ultra-wideband receivers, spectrum analyzers, RTT measurement tools, and wireless network controllers; Wireless communication signal parameters include multipath delay spread Dm, instantaneous spectrum change rate Ss, signal delay jitter Js and signal packet loss rate Pl; S12, preprocessing is performed based on the collected wireless communication signal parameters, the preprocessing normalizes the wireless communication signal parameters to ensure that parameters of different dimensions can be processed uniformly, and a filter is used to remove high-frequency noise, retain key dynamic and nonlinear characteristics in the signal, and summarize after preprocessing to obtain a standard wireless communication signal parameter group; S13, performing multi-dimensional data compounding on the collected standard wireless communication signal parameter group, and representing the wireless communication signal parameters as multi-dimensional time series data S(t); The multidimensional time series data S(t) is obtained by compounding the following multidimensional data; ; Where Dm(t), Ss(t), Js(t) and Pl(t) represent the multipath delay spread, instantaneous spectrum change rate, signal delay jitter and signal packet loss rate at time t, respectively.

[0007] Preferably, S2 includes S21 and S22; S21, mapping the multidimensional time series data S(t) into a multidimensional vector Y(t) through time delay embedding algorithm; The multidimensional vector Y(t) is calculated as output by the following time-delay embedding algorithm; ; In the formula, represents the time delay, which is a key parameter used to construct a multidimensional vector and represents the time difference between two signal sampling points. m represents the embedding dimension, which represents the dimension of the vector and is usually selected through optimization to ensure that the dynamic characteristics of the signal can be fully expressed. The multidimensional time series data S(t) are respectively By adding the values ​​at multiple moments, a new multidimensional vector Y(t) is obtained. The multidimensional vector Y(t) contains the information of the signal at different time points, thus forming a high-dimensional space that can reflect the dynamic changes of the signal. S22. Using MATLAB computing tools and the nonlinear dynamics analysis software TISEAN toolkit, the multidimensional vector Y(t) at each time point t is tracked, and drawn and tracked in phase space to form a complete trajectory.

[0008] Preferably, S3 includes S31 and S32; S31, extracting digital signal features of the signal from the trajectory in the phase space, the digital signal features including attractor dimension Da, fractal dimension Fd and Lyapunov exponent upper limit value Lmax; The attractor dimension Da is calculated by using nonlinear analysis tools to calculate the distance between two points in the phase space, counting the number of point pairs that meet a certain distance threshold r, and then calculating the attractor dimension Da according to the power law relationship; the attractor dimension Da is a parameter used to describe the complexity of the signal trajectory in the phase space, reflecting the "spatial volume" occupied by the signal trajectory in the multidimensional space. The higher the dimension, the greater the complexity of the signal, and the more degrees of freedom the trajectory exhibits; The fractal dimension Fd is calculated by dividing the phase space into multiple small boxes, counting the number of trajectory points in these small boxes, and calculating the fractal dimension Fd based on the change of the number of small boxes occupied by the trajectory as the boxes change. The fractal dimension Fd is used to describe the self-similarity of the trajectory in the phase space, that is, the trajectory has similar geometric features at different scales. The fractal dimension Fd can reflect the complex structure of the signal and is particularly suitable for analyzing chaotic signals. The upper limit value Lmax of the Lyapunov exponent is obtained by selecting two points with a very small initial distance in the phase space and calculating the evolution of the two points over time. The upper limit value Lmax of the Lyapunov exponent is an important indicator for measuring the chaotic characteristics of the signal system. It represents the separation rate of adjacent trajectory points in the phase space and can reflect the sensitivity of the signal system to the initial conditions.

[0009] Preferably, S32, after feature extraction, the extracted digital signal features and the standard wireless communication signal parameter group are combined and summarized to obtain a feature vector F(t); The expression of the eigenvector F(t) is as follows; .

[0010] Preferably, S4 includes S41 and S42; S41, constructing a comprehensive quality algorithm formula, extracting the feature vector F(t) and inputting it into the comprehensive quality algorithm formula, calculating and outputting the wireless communication signal quality Qs, and analyzing the quality status of the current wireless communication digital signal; The wireless communication signal quality Qs is calculated and output by the following comprehensive quality algorithm formula; ; Wherein, a1, a2, a3, a4, a5, a6 and a7 represent the preset weights of the attractor dimension Da, the fractal dimension Fd, the upper limit value of the Lyapunov exponent Lmax, the multipath delay spread Dm, the instantaneous spectrum change rate Ss, the signal delay jitter Js and the signal packet loss rate Pl, respectively, and a1+a2+a3+a4+a5+a6+a7=1, and the specific values ​​are set by the user.

[0011] Preferably, S42, based on the standard quality of historical wireless communication digital signal data, a preset signal quality threshold Q is then compared with the obtained wireless communication signal quality Qs to perform a preliminary comparative evaluation and analysis of the quality status of the wireless communication digital signal in the intelligent rail transit, and the specific evaluation content is as follows; When the wireless communication signal quality Qs ≥ the signal quality threshold Q, it means that the signal quality is normal and the wireless communication digital signal meets the expected requirements; When the wireless communication signal quality Qs is less than the signal quality threshold Q, it means that the signal quality is abnormal and the wireless communication data is interfered with abnormally. At this time, the signal quality status is deeply analyzed.

[0012] Preferably, S5 includes S51, S52 and S53; S51, when the signal quality is abnormal, the signal quality and channel performance of the wireless communication digital signal during the transmission process are analyzed to obtain dynamic features, the dynamic features including the phase difference fluctuation rate Xw and the channel mutual information Ic; The phase difference fluctuation rate Xw decomposes the received signal into the in-phase component E and the orthogonal component Y through the IQ demodulator, calculates the instantaneous phase Sx of the signal through these components, and then calculates the phase difference fluctuation rate Xw through the phase difference at adjacent time points. The instantaneous phase Sx and the phase difference fluctuation rate Xw are calculated and outputted respectively through the following algorithm formulas; ; ; In the formula, arctan represents the tangent function, Sx(t) represents the instantaneous phase at time t, E(t) represents the in-phase component at time t, Y(t) represents the orthogonal component at time t, dSx(t) represents the slight change in the instantaneous phase, and dt represents the slight change in time t; The channel mutual information Ic is calculated by the Shannon formula. The specific algorithm formula is: , where log represents the logarithmic function, S represents the signal power, N represents the noise power, and the channel mutual information Ic represents the amount of useful information that can be extracted from the signal transmitted by the transmitter at the receiving end.

[0013] Preferably, S52, based on the acquired dynamic features and in combination with the wireless communication signal quality Qs, a secondary comprehensive calculation is performed to output a second wireless communication signal quality score Qs2, and the quality status of the wireless communication digital signal is deeply analyzed by introducing the dynamic features; The second wireless communication signal quality score Qs2 is calculated and outputted by the following algorithm formula; ; Wherein, b1, b2 and b3 represent the preset weight values ​​of the wireless communication signal quality Qs, the phase difference fluctuation rate Xw and the channel mutual information Ic respectively, and b1+b2+b3=1, and the specific values ​​are set by the user; S53, performing a secondary comparative evaluation based on the obtained second wireless communication signal quality score Qs2 and the signal quality threshold Q, deeply analyzing the quality of the wireless communication digital signal, and generating a corresponding condition mechanism based on the secondary comparative evaluation result, the specific evaluation content is as follows; When the second wireless communication signal quality score Qs2 ≥ the signal quality threshold Q, it means that after comprehensive analysis combined with dynamic characteristics, the quality of the wireless communication digital signal is normal, and monitoring is continued at this time; When the second wireless communication signal quality score Qs2 is less than the signal quality threshold Q, it means that after comprehensive analysis combined with the dynamic characteristics, the quality of the wireless communication digital signal is abnormal. At this time, the abnormal signal is classified and adjusted.

[0014] Preferably, S6 includes S61 and S62; S61. Based on the secondary evaluation result, a classifier is used to classify the quality of the wireless communication signal into a normal signal, a primary interference abnormal signal, and a secondary interference abnormal signal. The specific classification method is as follows; When the second wireless communication signal quality score Qs2 ≥ the signal quality threshold Q, it is classified as a normal signal; When the second wireless communication signal quality score Qs2 is less than the signal quality threshold Q, it is classified as a first-level interference abnormal signal; When the second wireless communication signal quality score Qs2 is less than twice the signal quality threshold Q, it is classified as a secondary interference abnormal signal; Level 1 interference abnormal signal < Level 2 interference abnormal signal; S62. When the primary interference abnormal signal and the secondary interference abnormal signal are identified, the adjustment mechanism is automatically triggered. The adjustment mechanism includes adjusting the frequency of the communication channel to avoid the interference signal frequency band, optimizing the signal redundant coding and dynamically adjusting the signal transmission power.

[0015] Beneficial Effects The present invention provides a method for identifying characteristics of wireless communication digital signals. It has the following beneficial effects: (1) This method collects a variety of wireless communication signal parameters through IoT devices. After preprocessing and normalization, these signal parameters are used to obtain a standard wireless communication signal parameter group, which is further compounded into multidimensional time series data S(t). Through this composite analysis method, complex wireless signal parameters can be converted into multidimensional data with time series characteristics, ensuring that the dynamic relationship between different signal parameters is effectively expressed, and improving the system's ability to process and analyze complex signals. In scenarios with large dynamic changes such as rail transit, this method can more accurately capture and process nonlinear changes in signals and ensure the accuracy of signal recognition.

[0016] (2) This method maps multidimensional time series data S(t) to multidimensional phase space by introducing a time delay embedding algorithm, constructs a multidimensional vector Y(t), and tracks the trajectory of the signal in the phase space. By extracting key nonlinear dynamic signal features from the phase space trajectory, such as the attractor dimension Da, the fractal dimension Fd, and the upper limit of the Lyapunov exponent Lmax, this method can deeply capture the nonlinear dynamic behavior of the signal, especially the complexity and chaotic characteristics of the signal. Compared with traditional linear feature extraction methods, this method can more effectively analyze the intrinsic characteristics of complex signals, especially when the multipath effect is significant. It can significantly improve the accuracy and stability of signal classification, providing strong support for signal processing in complex environments.

[0017] (3) This method constructs a comprehensive quality algorithm formula. The present invention further calculates the extracted feature vector F(t), outputs the quality score Qs of the wireless communication signal, and makes a preliminary comparative evaluation with the preset signal quality threshold Q. On this basis, the phase difference volatility and channel mutual information Ic of the dynamic characteristics are introduced to perform a secondary comprehensive calculation of the signal quality, and output the second wireless communication signal quality score Qs2. This optimization process can more comprehensively analyze the dynamic change characteristics of the signal, and accurately classify the signal by constructing a classifier through the support vector machine SVM classification algorithm, dividing the signal into normal signals, primary interference abnormal signals and secondary interference abnormal signals. The system can automatically trigger the adjustment mechanism according to the signal quality level, and improve the communication reliability and anti-interference ability of the system in complex environments by adjusting the communication channel frequency, optimizing redundant coding and dynamically adjusting the transmission power, so as to effectively ensure the safe operation of the rail transit system. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 The present invention is a schematic diagram of the steps of a feature recognition method for a wireless communication digital signal. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] Example 1 The present invention provides a method for identifying characteristics of wireless communication digital signals. Figure 1 , including the following steps: S1. Collect wireless communication signal parameters through the Internet of Things devices in the intelligent rail transit system, pre-process the signal parameters, obtain a standard wireless communication signal parameter group, and perform a composite analysis on the standard wireless communication signal parameter group to obtain multi-dimensional time series data S(t); S2. Use the time delay embedding algorithm to map the multidimensional time series data S(t) to the multidimensional phase space, construct the multidimensional vector Y(t), and track each multidimensional vector Y(t) in the phase space to form the trajectory of the signal; S3, extracting digital signal features from the trajectory in the phase space, and then combining the acquired standard wireless communication signal parameters with the digital signal features to generate a feature vector F(t); S4. Construct a comprehensive quality algorithm formula, input the feature vector F(t) into the comprehensive quality algorithm formula, calculate and output the wireless communication signal quality Qs, and preset the signal quality threshold Q and the obtained wireless communication signal quality Qs, perform preliminary comparative evaluation, and analyze the current rail transit wireless communication digital signal quality status; S5. Based on the preliminary comparative evaluation results, dynamic features are introduced, and a second comprehensive calculation is performed to output a second wireless communication signal quality score Qs2, the recognition accuracy of the digital signal is optimized, and a second comparative evaluation is performed with the signal quality threshold Q to analyze the quality status of the optimized wireless communication digital signal; S6. Based on the extracted feature vector F(t), a classifier is constructed using the support vector machine SVM classification algorithm. The wireless communication digital signal is then classified according to the wireless communication signal quality Qs. The classification results are further refined into corresponding quality levels, and the trigger conditions of the adjustment mechanism are generated based on the quality levels to optimize and adjust the abnormal signals of the wireless communication digital signal.

[0021] In this embodiment, the method collects wireless communication signal parameters through an Internet of Things device, preprocesses them, and generates a standard wireless communication signal parameter group. Through multidimensional composite analysis, different parameters are converted into multidimensional time series data S (t), ensuring the unified processing and expression of multiple signal features. Using the time delay embedding algorithm, the multidimensional time series data is mapped to the multidimensional phase space, the multidimensional vector Y (t) is constructed, and the signal trajectory is tracked in the phase space, thereby capturing the nonlinear characteristics of the signal in dynamic changes. Then, the digital signal features are extracted from the trajectory in the phase space, and the comprehensive feature vector F (t) is generated in combination with the standard wireless communication signal parameter group, further improving the capture accuracy of the signal dynamic behavior. The comprehensive quality algorithm formula is reconstructed to calculate the quality score Qs of the wireless communication signal, and a preliminary comparison and evaluation is performed with the preset signal quality threshold Q to determine the quality state of the current signal. Then, by introducing dynamic features, a secondary comprehensive calculation is performed, and the second wireless communication signal quality score Qs2 is output, and a secondary comparison is performed with the signal quality threshold Q. This method has achieved important improvements in many aspects. First, by introducing nonlinear features, the system's ability to process wireless communication signals in complex environments is significantly enhanced, especially under conditions such as multipath propagation, delay jitter and interference, the recognition accuracy is greatly improved. Secondly, the addition of the secondary quality assessment mechanism ensures the stability and robustness of signal processing, enabling the system to dynamically adapt to environmental changes. Finally, the classification-based adjustment mechanism further optimizes the system's processing of abnormal signals, ensuring the continuity and reliability of signal transmission. In short, this method effectively improves the recognition accuracy, anti-interference ability and adaptive adjustment ability of wireless communication signals, and is suitable for complex wireless communication environments such as rail transit.

[0022] Example 2 This embodiment is explained in Example 1, please refer to Figure 1 , specifically: S1 includes S11, S12 and S13; S11. Real-time collection of wireless communication signal parameters through IoT devices in the intelligent rail transit system; IoT devices include ultra-wideband receivers, spectrum analyzers, RTT measurement tools, and wireless network controllers; Wireless communication signal parameters include multipath delay spread Dm, instantaneous spectrum change rate Ss, signal delay jitter Js and signal packet loss rate Pl; S12, preprocessing is performed based on the collected wireless communication signal parameters, the preprocessing normalizes the wireless communication signal parameters to ensure that parameters of different dimensions can be processed uniformly, and a filter is used to remove high-frequency noise, retain key dynamic and nonlinear characteristics in the signal, and summarize after preprocessing to obtain a standard wireless communication signal parameter group; S13, performing multi-dimensional data compounding on the collected standard wireless communication signal parameter group, and representing the wireless communication signal parameters as multi-dimensional time series data S(t); The multidimensional time series data S(t) is obtained by compounding the following multidimensional data; ; Where Dm(t), Ss(t), Js(t) and Pl(t) represent the multipath delay spread, instantaneous spectrum change rate, signal delay jitter and signal packet loss rate at time t, respectively.

[0023] In this embodiment, the method uses the Internet of Things devices in the intelligent rail transit system, and the system can collect wireless communication signal parameters in real time. The real-time acquisition of wireless communication signal parameters provides a high-quality data basis for subsequent processing. Then, the collected wireless communication signal parameters are normalized and filtered for noise reduction, which effectively eliminates the influence of dimensional differences and high-frequency noise, retains the key dynamic and nonlinear characteristics in the signal, thereby ensuring the consistency and high reliability of the data, and obtaining a standard wireless communication signal parameter group. Finally, the processed standard signal parameter group is subjected to multi-dimensional data compounding to generate multi-dimensional time series data S(t). This step constructs a multi-dimensional representation that reflects the complex characteristics of the signal by combining the relationship between the dynamic changes of multiple signal parameters, thereby improving the system's ability to parse and process complex signals.

[0024] Example 3 This embodiment is explained in Example 2. Please refer to Figure 1 , specifically: S2 includes S21 and S22; S21, mapping the multidimensional time series data S(t) into a multidimensional vector Y(t) through time delay embedding algorithm; The multidimensional vector Y(t) is calculated as output by the following time-delay embedding algorithm; ; In the formula, represents the time delay, which is a key parameter used to construct a multidimensional vector and represents the time difference between two signal sampling points. m represents the embedding dimension, which represents the dimension of the vector and is usually selected through optimization to ensure that the dynamic characteristics of the signal can be fully expressed. The multidimensional time series data S(t) are respectively By adding the values ​​at multiple moments, a new multidimensional vector Y(t) is obtained. The multidimensional vector Y(t) contains the information of the signal at different time points, thus forming a high-dimensional space that can reflect the dynamic changes of the signal. S22. Using MATLAB computing tools and the nonlinear dynamics analysis software TISEAN toolkit, the multidimensional vector Y(t) at each time point t is tracked, and drawn and tracked in phase space to form a complete trajectory.

[0025] In this embodiment, this method realizes multi-dimensional dynamic analysis of wireless communication signals, greatly improving the ability to capture and identify the features of complex signals. In S21, a time delay embedding algorithm is used to map the multi-dimensional time series data S(t) into a multi-dimensional vector Y(t), effectively integrating the change information of the signal at different time points and constructing a high-dimensional vector space that can fully express the dynamic characteristics of the signal. By accurately selecting the time delay and embedding dimension m, ensuring the comprehensive expression of signal features in multidimensional space, thereby enhancing the system's analytical ability in dealing with signal complexity and nonlinear characteristics. In S22, with the help of MATLAB computing tools and TISEAN nonlinear dynamics analysis tools, the multidimensional vector Y(t) at each time point is tracked and the signal trajectory is drawn in the phase space. Compared with existing signal processing methods, this method has achieved important improvements in many aspects. First, through the time delay embedding algorithm, the signal is mapped to the multidimensional phase space, breaking through the limitations of traditional one-dimensional signal processing and being able to capture richer dynamic information and potential features. It helps to reveal the complex nonlinear relationships and chaotic characteristics in the signal. Ultimately, this method significantly improves the accuracy and depth of signal feature extraction, especially in scenarios with complex multipath propagation and interference. The system can analyze the changing trend of the signal more efficiently and accurately, enhance the accuracy and robustness of signal recognition, and provide a more reliable solution for wireless communications in complex environments.

[0026] Example 4 This embodiment is explained in Example 3, please refer to Figure 1 , specifically: S3 includes S31 and S32; S31, extracting digital signal features of the signal from the trajectory in the phase space, the digital signal features including attractor dimension Da, fractal dimension Fd and Lyapunov exponent upper limit value Lmax; The attractor dimension Da is calculated by using nonlinear analysis tools to calculate the distance between two points in the phase space, counting the number of point pairs that meet a certain distance threshold r, and then calculating the attractor dimension Da according to the power law relationship; the attractor dimension Da is a parameter used to describe the complexity of the signal trajectory in the phase space, reflecting the "spatial volume" occupied by the signal trajectory in the multidimensional space. The higher the dimension, the greater the complexity of the signal, and the more degrees of freedom the trajectory exhibits; The fractal dimension Fd is calculated by dividing the phase space into multiple small boxes, counting the number of trajectory points in these small boxes, and calculating the fractal dimension Fd based on the change of the number of small boxes occupied by the trajectory as the boxes change. The fractal dimension Fd is used to describe the self-similarity of the trajectory in the phase space, that is, the trajectory has similar geometric features at different scales. The fractal dimension Fd can reflect the complex structure of the signal and is particularly suitable for analyzing chaotic signals. The upper limit value Lmax of the Lyapunov exponent is obtained by selecting two points with a very small initial distance in the phase space and calculating the evolution of the two points over time. The upper limit value Lmax of the Lyapunov exponent is an important indicator for measuring the chaotic characteristics of the signal system. It represents the separation rate of adjacent trajectory points in the phase space and can reflect the sensitivity of the signal system to the initial conditions.

[0027] S32, after feature extraction, the extracted digital signal features and the standard wireless communication signal parameter group are combined and summarized to obtain a feature vector F(t); The expression of the eigenvector F(t) is as follows; .

[0028] In this embodiment, the method in S31, by extracting key digital signal features from the trajectory in the phase space, the system can fully capture the complex dynamic behavior of the signal. In S32, by combining these extracted digital signal features with the standard wireless communication signal parameter group, the system generates a feature vector F (t) that fully describes the signal characteristics and parameter information. Through the construction of the feature vector F (t), the system has laid a solid foundation for subsequent signal classification and evaluation, and this method has achieved breakthrough improvements in many aspects. First, through the digital signal features, the system's ability to analyze signal complexity and chaotic behavior is significantly enhanced, solving the limitations of traditional signal processing methods in processing complex and nonlinear signals. Secondly, by combining these extracted digital signal features with the standard parameter group, the generated feature vector F (t) greatly improves the accuracy of signal classification and quality assessment. Finally, this method enables the system to have stronger robustness and adaptability when dealing with multipath effects, delay jitter and interference in complex communication environments, providing a more effective solution for wireless signal analysis in complex scenarios, and significantly improving the system's signal processing capabilities and classification accuracy.

[0029] Example 5 This embodiment is explained in Example 1, please refer to Figure 1 , specifically: S4 includes S41 and S42; S41, constructing a comprehensive quality algorithm formula, extracting the feature vector F(t) and inputting it into the comprehensive quality algorithm formula, calculating and outputting the wireless communication signal quality Qs, and analyzing the quality status of the current wireless communication digital signal; The wireless communication signal quality Qs is calculated and output by the following comprehensive quality algorithm formula; ; Wherein, a1, a2, a3, a4, a5, a6 and a7 represent the preset weights of the attractor dimension Da, the fractal dimension Fd, the upper limit value of the Lyapunov exponent Lmax, the multipath delay spread Dm, the instantaneous spectrum change rate Ss, the signal delay jitter Js and the signal packet loss rate Pl, respectively, and a1+a2+a3+a4+a5+a6+a7=1, and the specific values ​​are set by the user.

[0030] S42, based on the standard quality of historical wireless communication digital signal data, a preset signal quality threshold Q is then compared with the obtained wireless communication signal quality Qs to perform a preliminary evaluation and analysis of the quality status of the wireless communication digital signal in the intelligent rail transit, and the specific evaluation contents are as follows; When the wireless communication signal quality Qs ≥ the signal quality threshold Q, it means that the signal quality is normal and the wireless communication digital signal meets the expected requirements; When the wireless communication signal quality Qs is less than the signal quality threshold Q, it means that the signal quality is abnormal and the wireless communication data is interfered with abnormally. At this time, the signal quality status is deeply analyzed.

[0031] In this embodiment, in S41 of the method, the system constructs a comprehensive quality algorithm formula based on the extracted feature vector F(t), and obtains the wireless communication signal quality score Qs after weighted calculation. This step provides a quantitative evaluation of the signal state, ensuring that the multi-dimensional characteristics of the signal can be fully reflected in the quality evaluation. The score not only reflects the overall quality of the signal, but also accurately reveals the problems caused by nonlinear interference and multipath effects that may exist in the signal. In S42, the system sets a signal quality threshold Q based on historical wireless communication signal data, and makes a preliminary comparison between the real-time calculated signal quality score Qs and the signal quality threshold Q. This preliminary evaluation step provides an effective real-time monitoring mechanism for the system, which can quickly identify signal anomalies and ensure the stability of communication quality. Compared with the prior art, the present method constructs a comprehensive signal quality evaluation algorithm by introducing the feature vector F(t), which significantly improves the accuracy and sensitivity of signal quality evaluation. Traditional technologies often rely on a single signal feature and tend to ignore the nonlinear characteristics in complex channels. The present method can not only capture the multi-dimensional features in the signal, but also dynamically judge the signal state through the preset signal quality threshold Q, thereby optimizing the anomaly detection capability.

[0032] Example 6 This embodiment is explained in Example 5. Please refer to Figure 1 , specifically: S5 includes S51, S52 and S53; S51, when the signal quality is abnormal, the signal quality and channel performance of the wireless communication digital signal during the transmission process are analyzed to obtain dynamic features, the dynamic features including the phase difference fluctuation rate Xw and the channel mutual information Ic; The phase difference fluctuation rate Xw decomposes the received signal into the in-phase component E and the orthogonal component Y through the IQ demodulator, calculates the instantaneous phase Sx of the signal through these components, and then calculates the phase difference fluctuation rate Xw through the phase difference at adjacent time points. The instantaneous phase Sx and the phase difference fluctuation rate Xw are calculated and outputted respectively through the following algorithm formulas; ; ; In the formula, arctan represents the tangent function, Sx(t) represents the instantaneous phase at time t, E(t) represents the in-phase component at time t, Y(t) represents the orthogonal component at time t, dSx(t) represents the slight change in the instantaneous phase, and dt represents the slight change in time t; The channel mutual information Ic is calculated by the Shannon formula. The specific algorithm formula is: , where log represents the logarithmic function, S represents the signal power, N represents the noise power, and the channel mutual information Ic represents the amount of useful information that can be extracted from the signal transmitted by the transmitter at the receiving end.

[0033] S52, based on the acquired dynamic features and in combination with the wireless communication signal quality Qs, a secondary comprehensive calculation is performed to output a second wireless communication signal quality score Qs2, and the quality status of the wireless communication digital signal is deeply analyzed by introducing dynamic features; The second wireless communication signal quality score Qs2 is calculated and outputted by the following algorithm formula; ; Wherein, b1, b2 and b3 represent the preset weight values ​​of the wireless communication signal quality Qs, the phase difference fluctuation rate Xw and the channel mutual information Ic respectively, and b1+b2+b3=1, and the specific values ​​are set by the user; S53, performing a secondary comparative evaluation based on the obtained second wireless communication signal quality score Qs2 and the signal quality threshold Q, deeply analyzing the quality of the wireless communication digital signal, and generating a corresponding condition mechanism based on the secondary comparative evaluation result, the specific evaluation content is as follows; When the second wireless communication signal quality score Qs2 ≥ the signal quality threshold Q, it means that after comprehensive analysis combined with dynamic characteristics, the quality of the wireless communication digital signal is normal, and monitoring is continued at this time; When the second wireless communication signal quality score Qs2 is less than the signal quality threshold Q, it means that after comprehensive analysis combined with the dynamic characteristics, the quality of the wireless communication digital signal is abnormal. At this time, the abnormal signal is classified and adjusted.

[0034] In this embodiment, in S51, when the signal quality is detected to be abnormal, the system extracts key dynamic features by analyzing the quality and performance of the wireless communication signal during the transmission process. By introducing dynamic features, the system can more deeply understand the transmission quality and change trend of the signal. In S52, based on the dynamic features, combined with the previously calculated signal quality score Qs, the system performs a secondary calculation and outputs a second wireless communication signal quality score Qs2, further improving the accuracy of the signal quality assessment. This secondary assessment enables the system to have higher sensitivity by comprehensively analyzing dynamic features and static signal features, and can accurately identify complex interference and abnormal conditions in the signal. In S53, the system compares Qs2 obtained by the secondary calculation with the preset signal quality threshold Q. If Qs2≥Q, it means that the signal quality has returned to normal and the system will continue to monitor; if Qs2<Q, it means that the signal quality is abnormal, and the system will trigger an adjustment mechanism, such as adjusting the communication channel frequency, optimizing redundant coding, or dynamically adjusting the transmission power, thereby improving the stability of the signal. Compared with the prior art, this method has achieved key improvements in many aspects. First, by introducing dynamic features, the system can not only analyze static signal features, but also deeply analyze the dynamic changes of signals during transmission, greatly improving the ability to identify complex and nonlinear signals. Secondly, the introduction of secondary comprehensive calculation and comparative evaluation mechanism makes the system more sensitive and accurate in signal quality evaluation and processing, especially in multipath propagation and complex interference environments. The system can detect and respond to signal quality issues more quickly. Finally, by triggering the adjustment mechanism, this method ensures that the system can automatically optimize communication parameters when faced with abnormal signals, enhances the reliability of signal transmission and the adaptability of the system, and provides more stable wireless communication support for complex application scenarios such as intelligent rail transit.

[0035] Example 7 This embodiment is explained in Example 6. Please refer to Figure 1 , specifically: S6 includes S61 and S62; S61. Based on the secondary evaluation result, a classifier is used to classify the quality of the wireless communication signal into a normal signal, a primary interference abnormal signal, and a secondary interference abnormal signal. The specific classification method is as follows; When the second wireless communication signal quality score Qs2 ≥ the signal quality threshold Q, it is classified as a normal signal; When the second wireless communication signal quality score Qs2 is less than the signal quality threshold Q, it is classified as a first-level interference abnormal signal; When the second wireless communication signal quality score Qs2 is less than twice the signal quality threshold Q, it is classified as a secondary interference abnormal signal; Level 1 interference abnormal signal < Level 2 interference abnormal signal; S62. When the primary interference abnormal signal and the secondary interference abnormal signal are identified, the adjustment mechanism is automatically triggered. The adjustment mechanism includes adjusting the frequency of the communication channel to avoid the interference signal frequency band, optimizing the signal redundant coding and dynamically adjusting the signal transmission power.

[0036] In this embodiment, in S61, the system uses a classifier to classify the signal quality into normal signals, primary interference abnormal signals and secondary interference abnormal signals based on the results of the secondary evaluation. The classification standard is based on the secondary comparison of the signal quality score Qs2 obtained by the second calculation with the signal quality threshold Q to ensure that the classification process has a high degree of accuracy. When Qs2≥Q, the signal is classified as a normal state; when Qs2 is lower than Q, it is classified as a primary interference abnormal signal, and when Qs2 is lower than 2 times of Q, the signal is identified as a secondary interference abnormal signal. Through this classification, the system can distinguish different degrees of signal abnormality, providing a basis for subsequent processing and optimization. In S62, the system automatically triggers the corresponding adjustment mechanism for the identified primary and secondary interference abnormal signals. The adjustment mechanism includes adjusting the communication channel frequency to avoid the interference signal frequency band, optimizing redundant coding to improve the signal's anti-interference ability, and dynamically adjusting the transmission power to enhance the signal's stability. This mechanism ensures that the system can automatically adjust the transmission parameters according to the actual situation, quickly respond to signal interference, and improve the reliability of the communication link. Compared with traditional technologies, this method achieves significant improvements. First, through the precise signal classification mechanism, the system can effectively divide the signal according to different conditions of signal quality, so as to deal with abnormal signals of different levels in a more targeted manner. Secondly, the automatically triggered adjustment mechanism can respond to interference signals immediately, and by adjusting the frequency, optimizing the coding and dynamic power adjustment, the system has adaptive interference response capabilities. These improvements have greatly improved the system's anti-interference performance and signal stability in complex communication environments, and provided a strong guarantee for efficient communication in application scenarios such as intelligent rail transit. In the end, the system's response speed, automatic adjustment capability and anti-interference performance have been greatly improved, significantly enhancing the stability and reliability of the entire wireless communication system.

[0037] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and alterations may be made to the embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for identifying characteristics of wireless communication digital signals, characterized in that: The following steps are involved: S1. Collect wireless communication signal parameters through the Internet of Things devices in the intelligent rail transit system, pre-process the signal parameters, obtain a standard wireless communication signal parameter group, and perform a composite analysis on the standard wireless communication signal parameter group to obtain multi-dimensional time series data S(t); S2. Use the time delay embedding algorithm to map the multidimensional time series data S(t) to the multidimensional phase space, construct the multidimensional vector Y(t), and track each multidimensional vector Y(t) in the phase space to form the trajectory of the signal; S3, extracting digital signal features from the trajectory in the phase space, and then combining the acquired standard wireless communication signal parameters with the digital signal features to generate a feature vector F(t); S4. Construct a comprehensive quality algorithm formula, input the feature vector F(t) into the comprehensive quality algorithm formula, calculate and output the wireless communication signal quality Qs, and preset the signal quality threshold Q and the obtained wireless communication signal quality Qs, perform preliminary comparative evaluation, and analyze the current rail transit wireless communication digital signal quality status; S5. Based on the preliminary comparative evaluation results, dynamic features are introduced, and a second comprehensive calculation is performed to output a second wireless communication signal quality score Qs2, the recognition accuracy of the digital signal is optimized, and a second comparative evaluation is performed with the signal quality threshold Q to analyze the quality status of the optimized wireless communication digital signal; S6. Based on the extracted feature vector F(t), a classifier is constructed using the support vector machine SVM classification algorithm. The wireless communication digital signal is then classified according to the wireless communication signal quality Qs. The classification results are further refined into corresponding quality levels, and the trigger conditions of the adjustment mechanism are generated based on the quality levels to optimize and adjust the abnormal signals of the wireless communication digital signal.

2. The method for identifying characteristics of a wireless communication digital signal according to claim 1, characterized in that: Said S1 includes S11, S12 and S13; S11. Real-time collection of wireless communication signal parameters through IoT devices in the intelligent rail transit system; The IoT device includes an ultra-wideband receiver, a spectrum analyzer, an RTT measurement tool, and a wireless network controller; The wireless communication signal parameters include multipath delay spread Dm, instantaneous spectrum change rate Ss, signal delay jitter Js and signal packet loss rate Pl; S12, preprocessing is performed based on the collected wireless communication signal parameters, wherein the preprocessing is performed by normalizing the wireless communication signal parameters, and then using a filter to remove high-frequency noise, retaining key dynamic and nonlinear characteristics in the signal, and summarizing after preprocessing to obtain a standard wireless communication signal parameter group; S13, performing multi-dimensional data compounding on the collected standard wireless communication signal parameter group, and representing the wireless communication signal parameters as multi-dimensional time series data S(t); The multidimensional time series data S(t) is obtained by compounding the following multidimensional data; ； Where Dm(t), Ss(t), Js(t) and Pl(t) represent the multipath delay spread, instantaneous spectrum change rate, signal delay jitter and signal packet loss rate at time t, respectively.

3. The method for identifying characteristics of a wireless communication digital signal according to claim 2, characterized in that: The S2 includes S21 and S22; S21, mapping the multidimensional time series data S(t) into a multidimensional vector Y(t) through time delay embedding algorithm; The multidimensional vector Y(t) is calculated and outputted by the following time delay embedding algorithm; ； In the formula, represents the time delay, m represents the embedding dimension; S22. Using MATLAB computing tools and the nonlinear dynamics analysis software TISEAN toolkit, the multidimensional vector Y(t) at each time point t is tracked, and drawn and tracked in phase space to form a complete trajectory.

4. The method for identifying characteristics of a wireless communication digital signal according to claim 3, characterized in that: The S3 includes S31 and S32; S31, extracting digital signal features of the signal from the trajectory in the phase space, wherein the digital signal features include attractor dimension Da, fractal dimension Fd and Lyapunov exponent upper limit value Lmax; The attractor dimension Da is calculated by using a nonlinear analysis tool to calculate the distance between two points in the phase space, counting the number of point pairs that meet a certain distance threshold r, and then calculating the attractor dimension Da according to a power law relationship; The fractal dimension Fd is calculated by dividing the phase space into a plurality of small boxes, counting the number of trajectory points contained in these small boxes, and calculating the fractal dimension Fd based on the change law of the number of small boxes occupied by the trajectory as the boxes change; The upper limit value Lmax of the Lyapunov exponent is obtained by selecting two initial points in the phase space and calculating the evolution of the two points over time. The upper limit value Lmax of the Lyapunov exponent is an important indicator for measuring the chaotic characteristics of the signal system. It represents the separation rate of adjacent trajectory points in the phase space and can reflect the sensitivity of the signal system to the initial conditions.

5. The method for identifying characteristics of a wireless communication digital signal according to claim 4, characterized in that: S32, after feature extraction, the extracted digital signal features and the standard wireless communication signal parameter group are combined and summarized to obtain a feature vector F(t); The expression of the feature vector F(t) is as follows: 。 6. The method for identifying characteristics of a wireless communication digital signal according to claim 4, characterized in that: The S4 includes S41 and S42; S41, constructing a comprehensive quality algorithm formula, extracting the feature vector F(t) and inputting it into the comprehensive quality algorithm formula, calculating and outputting the wireless communication signal quality Qs, and analyzing the quality status of the current wireless communication digital signal; The wireless communication signal quality Qs is calculated and output by the following comprehensive quality algorithm formula; ； Wherein, a1, a2, a3, a4, a5, a6 and a7 represent the preset weights of the attractor dimension Da, the fractal dimension Fd, the upper limit value of the Lyapunov exponent Lmax, the multipath delay spread Dm, the instantaneous spectrum change rate Ss, the signal delay jitter Js and the signal packet loss rate Pl, respectively, and a1+a2+a3+a4+a5+a6+a7=1, and the specific values ​​are set by the user.

7. The method for identifying characteristics of a wireless communication digital signal according to claim 6, characterized in that: S42, based on the standard quality of historical wireless communication digital signal data, a preset signal quality threshold Q is then compared with the obtained wireless communication signal quality Qs to perform a preliminary evaluation and analysis of the quality status of the wireless communication digital signal in the intelligent rail transit, and the specific evaluation contents are as follows; When the wireless communication signal quality Qs ≥ the signal quality threshold Q, it means that the signal quality is normal and the wireless communication digital signal meets the expected requirements; When the wireless communication signal quality Qs is less than the signal quality threshold Q, it means that the signal quality is abnormal and the wireless communication data is interfered with abnormally. At this time, the signal quality status is deeply analyzed.

8. The method for identifying characteristics of a wireless communication digital signal according to claim 7, characterized in that: The S5 includes S51, S52 and S53; S51, when the signal quality is abnormal, the signal quality and channel performance of the wireless communication digital signal during the transmission process are analyzed to obtain dynamic features, wherein the dynamic features include the phase difference fluctuation rate Xw and the channel mutual information Ic; The phase difference fluctuation rate Xw decomposes the received signal into an in-phase component E and an orthogonal component Y through an IQ demodulator, calculates the instantaneous phase Sx of the signal through these components, and then calculates the phase difference fluctuation rate Xw through the phase difference at adjacent time points. The instantaneous phase Sx and the phase difference fluctuation rate Xw are respectively calculated and output by the following algorithm formulas; ; ; In the formula, arctan represents the tangent function, Sx(t) represents the instantaneous phase at time t, E(t) represents the in-phase component at time t, Y(t) represents the orthogonal component at time t, dSx(t) represents the slight change in the instantaneous phase, and dt represents the slight change in time t; The channel mutual information Ic is calculated by the Shannon formula. The specific algorithm formula is: , where log represents the logarithmic function, S represents the signal power, and N represents the noise power.

9. The method for identifying characteristics of a wireless communication digital signal according to claim 8, characterized in that: S52, based on the acquired dynamic features and in combination with the wireless communication signal quality Qs, a secondary comprehensive calculation is performed to output a second wireless communication signal quality score Qs2, and the quality status of the wireless communication digital signal is deeply analyzed by introducing dynamic features; The second wireless communication signal quality score Qs2 is calculated and output by the following algorithm formula; ； Wherein, b1, b2 and b3 represent the preset weight values ​​of the wireless communication signal quality Qs, the phase difference fluctuation rate Xw and the channel mutual information Ic respectively, and b1+b2+b3=1, and the specific values ​​are set by the user; S53, performing a secondary comparative evaluation based on the obtained second wireless communication signal quality score Qs2 and the signal quality threshold Q, deeply analyzing the quality of the wireless communication digital signal, and generating a corresponding condition mechanism based on the secondary comparative evaluation result, the specific evaluation content is as follows; When the second wireless communication signal quality score Qs2 ≥ the signal quality threshold Q, it means that after comprehensive analysis combined with dynamic characteristics, the quality of the wireless communication digital signal is normal, and monitoring is continued at this time; When the second wireless communication signal quality score Qs2 is less than the signal quality threshold Q, it means that after comprehensive analysis combined with the dynamic characteristics, the quality of the wireless communication digital signal is abnormal. At this time, the abnormal signal is classified and adjusted.

10. The method for identifying characteristics of a wireless communication digital signal according to claim 9, characterized in that: The S6 includes S61 and S62; S61. Based on the secondary evaluation result, a classifier is used to classify the quality of the wireless communication signal into a normal signal, a primary interference abnormal signal, and a secondary interference abnormal signal. The specific classification method is as follows; When the second wireless communication signal quality score Qs2 ≥ the signal quality threshold Q, it is classified as a normal signal; When the second wireless communication signal quality score Qs2 is less than the signal quality threshold Q, it is classified as a first-level interference abnormal signal; When the second wireless communication signal quality score Qs2 is less than twice the signal quality threshold Q, it is classified as a secondary interference abnormal signal; The first-level interference abnormal signal is less than the second-level interference abnormal signal; S62. When the primary interference abnormal signal and the secondary interference abnormal signal are identified, an adjustment mechanism is automatically triggered, wherein the adjustment mechanism includes adjusting the frequency of the communication channel to avoid the interference signal frequency band, optimizing the signal redundancy coding, and dynamically adjusting the signal transmission power.

Images