Vehicle-mounted wireless intelligent monitoring and real-time interference early warning method

Abstract

The invention provides a vehicle-mounted wireless intelligent monitoring and real-time interference early warning method, and belongs to the technical field of rail transit wireless communication, and the method comprises the steps: synchronously collecting multi-band wireless signals in a rail-mounted interval through a vehicle-mounted sensing terminal; analyzing the collected signals by using multiple parallel demodulation channels to obtain signal parameters; based on the analyzed signal parameters, constructing a signal quality thermodynamic diagram of the rail travel interval in real time, and performing state evaluation by using a dynamic interference early warning model; and when the state evaluation result meets a preset alarm threshold condition, interference early warning information is automatically generated and reported. According to the invention, comprehensive monitoring and accurate early warning of a rail transit wireless communication environment can be realized, the accuracy and real-time performance of interference identification are effectively improved, and stable operation of a CBTC system is guaranteed.

Description

A method for vehicle-mounted wireless intelligent monitoring and real-time interference early warning Technical Field

[0001] This invention relates to the field of wireless communication technology for rail transit, specifically to a method for vehicle-mounted wireless intelligent monitoring and real-time interference early warning. Background Technology

[0002] With the rapid development of rail transit, communication-based train control (CBTC) systems have become the core control system of modern rail transit. CBTC systems achieve real-time information exchange between trains and ground control centers through wireless communication, placing extremely high demands on the reliability and real-time performance of wireless communication. However, the wireless communication environment in rail transit is complex, with multiple systems coexisting, including public network signals, private network signals, and wireless local area network signals. This can easily lead to problems such as co-channel interference, adjacent channel interference, and signal shielding, seriously affecting the stable operation of the CBTC system.

[0003] Currently, traditional wireless communication monitoring methods mainly rely on fixed monitoring stations or manual inspections, which suffer from limited monitoring range, poor real-time performance, and inability to achieve precise positioning. Although some vehicle-mounted monitoring solutions exist, most can only perform simple signal strength measurements and lack the ability to simultaneously acquire and analyze multi-frequency signals, thus failing to achieve precise positioning and early warning of interference sources. Summary of the Invention

[0004] In view of the technical defects and drawbacks existing in the prior art, the present invention provides a vehicle-mounted wireless intelligent monitoring and real-time interference early warning method to overcome the above problems or at least partially solve the above problems, the specific solution of which is as follows;

[0005] As a first aspect of the present invention, a method for vehicle-mounted wireless intelligent monitoring and real-time interference early warning is provided, comprising the following steps:

[0006] S1. Through the full-band scanning module of the vehicle-mounted sensing terminal, the wireless signal of the first preset frequency band in the track section is collected synchronously.

[0007] S2. Using the multi-parallel demodulation channels integrated in the vehicle-mounted sensing terminal, the collected wireless signals are synchronously analyzed to obtain parameter information of public network signals, private network signals, wireless local area network signals, and the full-band spectrum.

[0008] S3. Based on the signal parameters obtained from the analysis, a signal quality heat map of the track section is constructed in real time, and a dynamic interference early warning model is used for state assessment.

[0009] S4. When the status assessment result meets the preset alarm threshold conditions, interference warning information is automatically generated and reported.

[0010] In some embodiments, the specific process of synchronous acquisition in step S1 includes:

[0011] Radio frequency signals are received through a wideband antenna, and the radio frequency module acquires the signals at a rate not lower than the preset scanning speed.

[0012] The acquired signal data is distributed in parallel to multiple independent demodulation channels to achieve synchronous processing of multi-band signals.

[0013] In some embodiments, the synchronous parsing in step S2 includes:

[0014] Through the first demodulation channel, the public network signal is demodulated securely without a card, and the base station parameters, including the mobile country code, mobile network code, physical cell identifier, reference signal received power, and signal-to-interference-plus-noise ratio, are obtained.

[0015] The private network signal is analyzed through the second demodulation channel, and private network parameters, including cell identifier and reference signal reception quality, are extracted in real time.

[0016] The wireless LAN signal is analyzed through the third demodulation channel to obtain network parameters, including media access control address, channel distribution and signal strength.

[0017] The fourth demodulation channel performs a full-band spectrum scan to obtain parameter information including frequency points, signal strength, and spectral characteristics, thereby generating a full-band spectrum waterfall plot and enabling visualization of the spectrum situation.

[0018] In some embodiments, the first demodulation channel performs cardless secure demodulation of public network signals, which is a physically isolated radio frequency channel that directly parses base station parameters.

[0019] In some embodiments, the step S3 of constructing a signal quality heatmap of the track section in real time based on the analyzed signal parameters includes:

[0020] The received power, received quality and signal-to-interference-plus-noise ratio of the reference signal obtained from the analysis are preprocessed to remove outliers and noise interference.

[0021] Based on the preprocessed signal parameters, a spatial interpolation algorithm is used to perform spatial interpolation on the track section to generate a continuous signal quality distribution map.

[0022] By integrating the signal quality distribution map with the geographical information of the track section, and converting the signal quality parameters into heat map color gradients through color mapping, a signal quality heat map is generated, enabling a visual display of signal quality.

[0023] In some embodiments, the state assessment using the dynamic interference early warning model in step S3 includes:

[0024] The signal quality heatmap, or the parameter matrix of reference signal received power, reference signal received quality, and signal-to-interference-plus-noise ratio extracted from the signal quality heatmap, is input into a pre-trained machine learning classification model.

[0025] The machine learning classification model outputs a state classification result of the current wireless communication environment based on the input features; the state classification result includes normal state, coverage blind spot, co-channel interference, adjacent channel interference, and signal shielding.

[0026] The criteria for determining the coverage blind zone is that the reference signal received power value is continuously lower than a preset coverage threshold; the criteria for determining co-channel interference and adjacent channel interference is that the signal-to-interference-plus-noise ratio is continuously lower than a preset interference threshold within the licensed frequency band, and that specific spectral characteristics exist; the criteria for determining the signal shielding is that the licensed signal strength drops sharply within a preset time and is accompanied by the appearance of an unlicensed broadband strong signal.

[0027] In some embodiments, step S4, which involves automatically generating and reporting interference warning information when the state evaluation result meets a preset alarm threshold condition, includes:

[0028] When the status assessment results identify a coverage blind spot, a coverage blind spot alarm is triggered;

[0029] When the status assessment results identify co-channel or adjacent-channel interference, a signal interference alarm is triggered.

[0030] When the status assessment results identify signal shielding, an interference source location alarm is triggered.

[0031] In some embodiments, the automatic generation and reporting of interference warning information in step S4 further includes:

[0032] The geographical location of the interference source is determined based on the interference source localization algorithm, and the interference intensity is evaluated. The interference source localization algorithm adopts the time difference of arrival localization method, which calculates the geographical location of the interference source based on the signal time difference of multiple vehicle-mounted sensing terminals.

[0033] Based on the interference intensity and the type of abnormal state, interference warning information is divided into three levels: emergency alarm, important alarm, and general alarm. When the interference intensity is greater than the first alarm threshold, it is classified as an emergency alarm; when the interference intensity is greater than the second alarm threshold but less than or equal to the first alarm threshold, it is classified as an important alarm; and when the interference intensity is less than or equal to the second alarm threshold, it is classified as a general alarm.

[0034] Interference warning information of different alarm levels is pushed through preset push channels. Emergency alarm level triggers audible and visual alarms and pushes them to the mobile terminals of operation and maintenance personnel. Important alarm level is pushed to the mobile terminals of operation and maintenance personnel. General alarm level is recorded in the historical alarm database and pushed to operation and maintenance personnel on a regular basis.

[0035] In some embodiments, the method further includes data storage and display, specifically including:

[0036] The analyzed signal parameters and generated interference warning information are uploaded to a ground server for distributed storage, and the storage time is no less than the preset storage period.

[0037] The monitoring data is visualized through the network management terminal. The visualization includes four major sections: public network monitoring, private network monitoring, WiFi monitoring, and spectrum monitoring, and supports large-screen push of real-time alarm information.

[0038] In some embodiments, the method further includes: using artificial intelligence algorithms to predict interference trends based on historical monitoring data, thereby transforming the operation and maintenance model from passive response to proactive prevention.

[0039] The present invention has the following beneficial effects:

[0040] This invention first uses the full-band scanning module of the onboard sensing terminal to synchronously collect signals within the track area, ensuring the comprehensiveness and timeliness of data acquisition and laying a reliable data foundation for subsequent analysis. Subsequently, it utilizes integrated multi-channel parallel demodulation to synchronously analyze public network, private network, WiFi, and full-spectrum signals, improving the efficiency of multi-dimensional signal parameter extraction and overcoming the latency issues of traditional serial processing methods. Based on this, by constructing a real-time signal quality heatmap and inputting it into a dynamic interference early warning model, it can accurately identify various abnormal states such as coverage blind spots, co-channel interference, adjacent channel interference, and signal shielding, achieving intelligent and refined assessment of the health of the communication environment. Finally, when the assessment results meet the alarm conditions, the system can automatically generate and report tiered early warning information, and even combine positioning algorithms to determine the interference source, thereby significantly shortening the fault response time and effectively ensuring the reliability and safety of critical services such as train control based on wireless communication, improving the operation and maintenance efficiency and intelligence level of the entire rail transit system. Attached Figure Description

[0041] Figure 1 is a flowchart illustrating a vehicle-mounted wireless intelligent monitoring and real-time interference early warning method provided by an embodiment of the present invention. Detailed Implementation

[0042] To enable those skilled in the art to better understand the technical solutions of the present invention, exemplary embodiments of the present invention are described below in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to aid understanding. These should be considered merely exemplary. Therefore, those skilled in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0043] Where there is no conflict, the various embodiments of the present invention and the features thereof may be combined with each other.

[0044] As used herein, the term “and / or” includes any and all combinations of one or more related enumerated entries.

[0045] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used herein, the singular forms “a” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will also be understood that when the terms “comprising” and / or “made of” are used in this specification, the presence of the stated feature, integral, step, operation, element, and / or component is specified, but the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or groups thereof is not excluded. Terms such as “connected” or “linked” are not limited to physical or mechanical connections but can include electrical connections, whether direct or indirect.

[0046] Unless otherwise specified, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art. It will also be understood that terms such as those defined in commonly used dictionaries should be interpreted as having the meaning consistent with their meaning in the context of the relevant art and the invention, and will not be interpreted as having an idealized or overly formal meaning unless expressly so defined herein.

[0047] In the technical solution of this invention, the collection, storage, use, processing, transmission, provision, and disclosure of user personal information all comply with relevant laws and regulations and do not violate public order and good morals. The use of user data in this technical solution follows relevant national laws and regulations (e.g., the "Information Security Technology - Personal Information Security Specification"). For example: appropriate measures are taken for personal information access control; restrictions are imposed on the display of personal information; the purpose of using personal information does not exceed the scope of direct or reasonable association; and explicit identity targeting is eliminated when using personal information to avoid precisely locating a specific individual.

[0048] To address at least one of the technical problems existing in the aforementioned related technologies, the present invention provides a method for vehicle-mounted wireless intelligent monitoring and real-time interference early warning. Figure 1 is a schematic flowchart of a method for vehicle-mounted wireless intelligent monitoring and real-time interference early warning provided by an embodiment of the present invention, including the following steps:

[0049] S1. Through the full-band scanning module of the vehicle-mounted sensing terminal, the wireless signal of the first preset frequency band in the track section is collected synchronously.

[0050] S2. Using the multi-channel parallel demodulation channel (channel) integrated in the vehicle-mounted sensing terminal, the collected wireless signals are synchronously analyzed to obtain parameter information of public network signals, private network signals, wireless local area network signals and full-band spectrum respectively.

[0051] S3. Based on the signal parameters obtained from the analysis, a signal quality heat map of the track section is constructed in real time, and a dynamic interference early warning model is used for state assessment.

[0052] S4. When the status assessment result meets the preset alarm threshold conditions, interference warning information is automatically generated and reported.

[0053] Optionally, in this embodiment, the first preset frequency band is specifically 9kHz to 6GHz.

[0054] This invention provides a vehicle-mounted wireless intelligent monitoring and real-time interference early warning method. By constructing an automated technical solution that integrates synchronous signal acquisition, multi-standard parallel analysis, intelligent status assessment and proactive early warning, it fundamentally solves the industry pain points of difficulty in comprehensive and real-time monitoring of multiple system signals in the complex wireless communication environment of rail transit, as well as the lag in the discovery and difficulty in locating interference events. It realizes the intelligent transformation of the operation and maintenance mode from passive response to proactive prevention.

[0055] Specifically, this invention first uses the full-band scanning module of the onboard sensing terminal to synchronously collect signals within the track area, ensuring the comprehensiveness and timeliness of data acquisition and laying a reliable data foundation for subsequent analysis. Subsequently, it utilizes integrated multi-channel parallel demodulation to synchronously analyze public network, private network, WiFi, and full-spectrum signals, improving the efficiency of multi-dimensional signal parameter extraction and overcoming the latency issues of traditional serial processing methods. Based on this, by constructing a real-time signal quality heatmap and inputting it into a dynamic interference early warning model, it can accurately identify various abnormal states such as coverage blind spots, co-channel interference, adjacent channel interference, and signal shielding, achieving intelligent and refined assessment of the health of the communication environment. Finally, when the assessment results meet the alarm conditions, the system can automatically generate and report tiered early warning information, and even combine positioning algorithms to determine the interference source, thereby significantly shortening the fault response time and effectively ensuring the reliability and safety of critical services such as train control based on wireless communication, improving the operation and maintenance efficiency and intelligence level of the entire rail transit system.

[0056] In some embodiments, the specific process of synchronous acquisition in step S1 includes:

[0057] Radio frequency signals are received through a wideband antenna, and the radio frequency module acquires the signals at a rate not lower than the preset scanning speed.

[0058] The acquired signal data is distributed in parallel to multiple independent demodulation channels to achieve synchronous processing of multi-band signals.

[0059] This embodiment receives radio frequency signals through a wideband antenna, and the radio frequency module performs signal acquisition at a rate no less than the preset scanning speed. Then, the acquired signal data is distributed in parallel to multiple independent demodulation channels, realizing synchronous processing of multi-band signals. This effectively improves the efficiency and real-time performance of signal acquisition, solves the problem of slow response speed in traditional serial processing methods, and provides a reliable data foundation for subsequent signal analysis and interference warning.

[0060] Optionally, in this embodiment, the preset scanning speed is 100 GHz / s, and the number of demodulation channels is 4. The radio frequency module receives radio frequency signals in the 9 kHz to 6 GHz band within the orbital range via a wideband antenna. It performs signal acquisition at a scanning speed of 100 GHz / s, enabling it to quickly capture all wireless signals within this band, including public 4G / 5G signals (700 MHz - 3.5 GHz), LTE-M private network signals (1.8 GHz), wireless LAN signals (2.4 GHz / 5 GHz), and other frequency band signals that may cause interference.

[0061] The acquired signal data is distributed in parallel to four independent demodulation channels, each dedicated to processing signals in a specific frequency band. The first demodulation channel processes public 4G / 5G signals, the second processes LTE-M private network signals, the third processes wireless LAN signals, and the fourth generates a full-band spectrum waterfall plot. This parallel processing method allows the signal analysis process to be completed within 300 milliseconds, improving the response speed by more than three times compared to traditional serial processing.

[0062] In some embodiments, the synchronous parsing in step S2 includes:

[0063] Through the first demodulation channel, the public network signal is demodulated securely without a card, and the base station parameters, including the mobile country code, mobile network code, physical cell identifier, reference signal received power, and signal-to-interference-plus-noise ratio, are obtained.

[0064] The private network signal is analyzed through the second demodulation channel, and private network parameters, including cell identifier and reference signal reception quality, are extracted in real time.

[0065] The wireless LAN signal is analyzed through the third demodulation channel to obtain network parameters, including media access control address, channel distribution and signal strength.

[0066] The fourth demodulation channel performs a full-band spectrum scan to obtain parameter information including frequency points, signal strength, and spectral characteristics, thereby generating a full-band spectrum waterfall plot and enabling visualization of the spectrum situation.

[0067] This embodiment sets up multiple independent demodulation channels to perform parallel analysis of public network signals, private network signals, and wireless LAN signals, and generates a full-band spectrum waterfall plot. This achieves synchronous analysis of signals from multiple systems and visualization of the spectrum situation, effectively improving the efficiency and real-time performance of signal analysis. It solves the problem of slow response speed in traditional serial analysis methods and provides a reliable data foundation for subsequent signal quality assessment and interference early warning.

[0068] Optionally, in this embodiment, the multi-parallel demodulation channel includes four independent demodulation channels, each using a dedicated signal processing chip, enabling true parallel processing.

[0069] The first demodulation channel employs SIM-free secure demodulation technology to analyze public 4G / 5G signals. Specifically, it directly analyzes base station parameters through a physically isolated radio frequency channel, avoiding access to the public network via a SIM card. The analyzed parameters include: Mobile Country Code (MCC), Mobile Network Code (MNC), Physical Cell Identifier (PCI), Reference Signal Received Power (RSRP), and Signal-to-Interference-plus-Noise Ratio (SINR). The RSRP measurement range is selectable from -140dBm to -44dBm, and the SINR measurement range is selectable from -20dB to 30dB.

[0070] The second demodulation channel is dedicated to processing LTE-M private network signals, extracting cell identifier (Cell ID) and reference signal reception quality (RSRQ) in real time. The Cell ID can be encoded in 16 bits to uniquely identify the cell; the RSRQ measurement range can be selected from -19.5dB to -3dB to evaluate the private network signal quality.

[0071] The third demodulation channel analyzes the wireless LAN signal to obtain network parameters such as the Media Access Control (MAC) address, channel distribution, and signal strength. The MAC address can be in 48-bit format and can uniquely identify the wireless device; the channel distribution covers channels 1-13 in the 2.4GHz band and channels 36-165 in the 5GHz band; the signal strength measurement range is -100dBm to -20dBm.

[0072] The fourth demodulation channel generates a full-band spectrum waterfall plot, ranging from 9kHz to 6GHz, with a scan rate of 100GHz / s, enabling visualization of the spectrum situation. The spectrum waterfall plot uses a three-dimensional display method of time-frequency-power, with a time resolution of 1 second and a selectable frequency resolution of 10kHz.

[0073] In some embodiments, the first demodulation channel performs cardless secure demodulation of public network signals by directly parsing base station parameters using a physically isolated radio frequency channel, avoiding access to the public network through a SIM card.

[0074] This embodiment achieves secure demodulation of public network signals without a SIM card by directly parsing base station parameters through a physically isolated radio frequency channel. This avoids the security risks associated with accessing the public network via a SIM card, while improving the efficiency and real-time performance of signal parsing. It also provides a more secure and reliable data foundation for subsequent signal quality assessment and interference warning.

[0075] Optionally, in this embodiment, the first demodulation channel uses a physically isolated radio frequency channel to achieve cardless secure demodulation. Specifically, a public network 4G / 5G signal is received through an independent radio frequency front-end module. This module is physically isolated from the main processor via an interface connection, preventing signal data from being directly exposed in the main processor's memory space.

[0076] The physically isolated radio frequency (RF) channel comprises an independent RF chip, baseband processing unit, and encryption module. The RF chip receives public network signals in the 700MHz-3.5GHz frequency band. The baseband processing unit demodulates and decodes the signals, extracting base station parameters such as MCC, MNC, PCI, RSRP, and SINR. The encryption module uses the AES-256 encryption algorithm to encrypt the parsed parameters before transmitting them to the main processor via a physically isolated SPI interface.

[0077] This physical isolation method prevents the main processor from directly accessing the raw signal data in the radio frequency channel, even if it is attacked by malware, effectively preventing the risk of signal data leakage. Furthermore, since it eliminates the need to access the public network via a SIM card, it avoids security vulnerabilities in the SIM card authentication process, thus improving the overall security of the system.

[0078] In some embodiments, the step S3 of constructing a signal quality heatmap of the track section in real time based on the analyzed signal parameters includes:

[0079] The received power, received quality and signal-to-interference-plus-noise ratio of the reference signal obtained from the analysis are preprocessed to remove outliers and noise interference.

[0080] Based on the preprocessed signal parameters, a spatial interpolation algorithm is used to perform spatial interpolation on the track section to generate a continuous signal quality distribution map.

[0081] By integrating the signal quality distribution map with the geographical information of the track section, and converting the signal quality parameters into heat map color gradients through color mapping, a signal quality heat map is generated, enabling a visual display of signal quality.

[0082] This embodiment removes outliers and noise interference through data preprocessing, generates a continuous signal quality distribution map using a spatial interpolation algorithm, and converts signal quality parameters into heatmap color gradients through color mapping. This enables comprehensive monitoring and visualization of signal quality in the rail transit wireless communication environment, effectively improving the accuracy and intuitiveness of signal quality assessment and providing reliable data support for subsequent interference warnings and operation and maintenance decisions.

[0083] In this embodiment, in step S3, based on the parsed signal parameters, a signal quality heat map of the track section is constructed in real time. Specifically, data preprocessing is performed on the parsed RSRP, RSRQ, and SINR. The 3σ criterion is used to remove outliers, and the median filtering algorithm is used to remove noise interference. The specific implementation of the 3σ criterion is as follows: calculate the mean and standard deviation of RSRP, RSRQ, and SINR, and regard the data points outside the range of the mean ± 3 times the standard deviation as outliers and剔除 them; the median filtering algorithm uses a 3×3 sliding window, sorts the data points within the window, and takes the median as the value of the current point, effectively removing noise interference.

[0084] Based on the preprocessed signal parameters, the Kriging interpolation algorithm is used to perform spatial interpolation on the track section to generate a continuous signal quality distribution map. The Kriging interpolation algorithm is a spatial interpolation method based on statistics that can consider spatial autocorrelation and generate a continuous signal quality distribution map. During the interpolation process, the track section is divided into 1m×1m grids, and interpolation calculations are performed on each grid point to obtain the continuous distributions of RSRP, RSRQ, and SINR. The specific parameter settings of the Kriging interpolation algorithm are as follows: the semivariogram uses a spherical model, the range is 50m, the nugget effect is 0.1, the sill value is 1.0, and the anisotropy ratio is 1.0.

[0085] The signal quality distribution map is fused with the geographical information of the track section, and the signal quality parameters are converted into a heat map color gradient through color mapping to generate a signal quality heat map, realizing the visual display of signal quality. Specifically, a red-yellow-green color gradient scheme is adopted. Red indicates poor signal quality (for example, RSRP ≤ -110dBm), yellow indicates average signal quality (for example, -110dBm < RSRP ≤ -95dBm), and green indicates good signal quality (for example, RSRP > -95dBm). The color mapping function uses linear interpolation to map the RSRP value to the color space of 0-255, realizing the visual display of signal quality.

[0086] In some embodiments, the use of the dynamic interference warning model for status evaluation in step S3 includes:

[0087] Input the signal quality heat map, or the parameter matrix of the reference signal received power, reference signal received quality, and signal-to-interference-plus-noise ratio extracted from the signal quality heat map, into a pre-trained machine learning classification model;

[0088] Based on the input features, the machine learning classification model outputs a status classification result for the current wireless communication environment; the status classification result includes normal status, coverage blind area, co-channel interference, adjacent-channel interference, and signal shielding;

[0089] The criteria for determining the coverage blind zone is that the reference signal received power value is continuously lower than a preset coverage threshold; the criteria for determining co-channel interference and adjacent channel interference is that the signal-to-interference-plus-noise ratio is continuously lower than a preset interference threshold within the licensed frequency band, and that specific spectral characteristics exist; the criteria for determining the signal shielding is that the licensed signal strength drops sharply within a preset time and is accompanied by the appearance of an unlicensed broadband strong signal.

[0090] This embodiment inputs signal parameters from a signal quality heatmap into a pre-trained machine learning model. Based on the extracted signal quality features, it performs state classification, accurately identifies abnormal states such as coverage blind spots, co-channel interference, adjacent channel interference, and signal shielding, and generates corresponding interference warning information. This achieves precise monitoring and intelligent early warning of the wireless communication environment of rail transit, effectively improving the accuracy and real-time performance of interference identification, and providing reliable data support for subsequent interference location and operation and maintenance decisions.

[0091] In this embodiment, step S3 utilizes a dynamic interference early warning model for state assessment. Specifically, the signal parameters from the signal quality heatmap are input into a pre-trained machine learning model. This machine learning model employs a convolutional neural network (CNN) architecture, including three convolutional layers, two pooling layers, and two fully connected layers. The model's input features are RSRP, RSRQ, and SINR from the signal quality heatmap, with an input size of 256×256×3, corresponding to the RSRP, RSRQ, and SINR channels, respectively.

[0092] The convolutional layers use 3×3 kernels and ReLU activation. The first convolutional layer outputs 64 feature maps, the second outputs 128 feature maps, and the third outputs 256 feature maps. The pooling layers use max pooling with a 2×2 window and a stride of 2. The fully connected layers have 512 neurons in the first layer and 5 neurons in the second layer, corresponding to five categories: normal state, coverage blind zone, co-channel interference, adjacent channel interference, and signal masking. The output layer uses the Softmax activation function to output the probability of each category.

[0093] Signal quality features are extracted using a machine learning model, and state classification is performed based on these features. When an abnormal state is detected, corresponding interference warning information is generated according to the type and severity of the abnormal state, and a confidence score for the interference event is calculated. The confidence score is calculated using the Softmax function, and an alarm is triggered when the score is greater than 0.8.

[0094] In some embodiments, step S4, which involves automatically generating and reporting interference warning information when the state evaluation result meets a preset alarm threshold condition, includes:

[0095] When the status assessment results identify a coverage blind spot, a coverage blind spot alarm is triggered;

[0096] When the status assessment results identify co-channel or adjacent-channel interference, a signal interference alarm is triggered.

[0097] When the status assessment results identify signal shielding, an interference source location alarm is triggered.

[0098] In this embodiment, the specific judgment rules can be selected as follows: by judging whether the received power value of the reference signal is not greater than the second preset threshold, if so, a coverage blind zone alarm is triggered; by judging whether the ratio of signal to interference plus noise of the private network signal is not greater than the first preset threshold, if so, a signal interference alarm is triggered; by judging whether an unlicensed strong signal is detected in the licensed frequency band, if so, an interference source location alarm is triggered.

[0099] This embodiment uses logical judgment based on the state assessment results and preset alarm threshold conditions to trigger a coverage blind zone alarm when a coverage blind zone is identified and the reference signal received power value is not greater than a second preset threshold. When co-channel interference or adjacent channel interference is identified and the signal-to-interference-plus-noise ratio is not greater than a first preset threshold, a signal interference alarm is triggered. When signal shielding is identified and an unlicensed strong signal is detected in the licensed frequency band, an interference source location alarm is triggered. This achieves accurate monitoring and intelligent early warning of the wireless communication environment of rail transit, effectively improving the accuracy and real-time performance of interference identification.

[0100] In this embodiment, in step S4, when the state assessment result meets the preset alarm threshold conditions, interference warning information is automatically generated and reported. Specifically, when the state assessment result identifies a coverage blind spot, it is determined whether the RSRP value is not greater than -110dBm (the second preset threshold). If it is, a coverage blind spot alarm is triggered. When the state assessment result identifies co-channel interference or adjacent-channel interference, it is determined whether the SINR value of the LTE-M private network is not greater than -10dB (the first preset threshold). If it is, a signal interference alarm is triggered. When the state assessment result identifies signal shielding, it is determined whether an unlicensed strong signal is detected in the licensed frequency band. If it is, an interference source location alarm is triggered.

[0101] In some embodiments, the automatic generation and reporting of interference warning information in step S4 further includes:

[0102] The geographical location of the interference source is determined based on the interference source localization algorithm, and the interference intensity is evaluated. The interference source localization algorithm adopts the time difference of arrival localization method, which calculates the geographical location of the interference source based on the signal time difference of multiple vehicle-mounted sensing terminals.

[0103] Based on the interference intensity and the type of abnormal state, interference warning information is divided into three levels: emergency alarm, important alarm, and general alarm. When the interference intensity is greater than the first alarm threshold, it is classified as an emergency alarm; when the interference intensity is greater than the second alarm threshold but less than or equal to the first alarm threshold, it is classified as an important alarm; and when the interference intensity is less than or equal to the second alarm threshold, it is classified as a general alarm.

[0104] Interference warning information of different alarm levels is pushed through preset push channels. Emergency alarm level triggers audible and visual alarms and pushes them to the mobile terminals of operation and maintenance personnel. Important alarm level is pushed to the mobile terminals of operation and maintenance personnel. General alarm level is recorded in the historical alarm database and pushed to operation and maintenance personnel on a regular basis.

[0105] In the above embodiments, the geographical location of the interference source is determined by using the time difference of arrival (TDOA) positioning method. The interference warning information is divided into three levels—emergency alarm, important alarm, and general alarm—based on the interference intensity and the type of abnormal state. The information is then pushed through a preset push channel, which realizes accurate positioning, intelligent classification, and timely push of interference events in the rail transit wireless communication environment. This effectively improves the accuracy and response speed of interference warnings and provides reliable data support for subsequent interference positioning and operation and maintenance decisions.

[0106] In this embodiment, step S4 determines the geographical location of the interference source based on the interference source localization algorithm and assesses the interference intensity. Specifically, the Time Difference of Arrival (TDOA) localization method is used to calculate the geographical location of the interference source based on the signal time difference of multiple vehicle-mounted sensing terminals. Assume there are three vehicle-mounted sensing terminals A, B, and C, located at coordinates... The arrival times of the signals emitted by the interference source at the three terminals are as follows: The location (x, y) of the interference source can be obtained by solving the following system of equations:

[0107] ; ; ;

[0108] Where c is the speed of light. The time when the interference source transmits its signal is indicated by the value of '_'. Solving the above equations using the least squares method yields the geographical location of the interference source. Interference intensity is assessed based on the change in the signal-to-interference-plus-noise ratio (SINR). A SINR drop exceeding 10 dB is considered strong interference; a drop of 5-10 dB is considered moderate interference; and a drop less than 5 dB is considered weak interference.

[0109] Based on the interference intensity and the type of abnormal state, interference warning information is divided into three levels: emergency alarm, important alarm, and general alarm. Specifically, when the interference intensity is greater than -5dB (the first alarm threshold), it is classified as an emergency alarm; when the interference intensity is greater than -10dB and less than or equal to -5dB (the second alarm threshold), it is classified as an important alarm; and when the interference intensity is less than or equal to -10dB, it is classified as a general alarm.

[0110] Interference warnings at different alarm levels are pushed through preset channels. Emergency alarms trigger audible and visual alarms and are pushed to the mobile terminals of maintenance personnel; important alarms are pushed to the mobile terminals of maintenance personnel; and general alarms are recorded in the historical alarm database and pushed to maintenance personnel periodically. Push channels include SMS, WeChat, email, and other methods to ensure that maintenance personnel receive alarm information promptly.

[0111] In some embodiments, the entire process of synchronous parsing described in step S2 is completed within a preset time threshold.

[0112] In this embodiment, the preset time threshold is specifically 300 milliseconds. The collected wireless signals are synchronously analyzed through the multi-parallel demodulation channels integrated in the vehicle-mounted sensing terminal, and the entire analysis process is completed within 300 milliseconds.

[0113] In some embodiments, the method further includes data storage and display, specifically including:

[0114] The analyzed signal parameters and generated interference warning information are uploaded to a ground server for distributed storage, and the storage time is no less than the preset storage period.

[0115] The monitoring data is visualized through the network management terminal. The visualization includes four major sections: public network monitoring, private network monitoring, WiFi monitoring, and spectrum monitoring, and supports large-screen push of real-time alarm information.

[0116] This embodiment, by setting up data storage and display steps, uploads the parsed signal parameters and generated interference warning information to a ground server for distributed storage, and visualizes the monitoring data through a network management terminal. This achieves long-term storage and multi-dimensional visualization analysis of monitoring data, solving the problems of incomplete data storage and unintuitive display in traditional methods, and providing comprehensive and reliable data support for subsequent data analysis and operation and maintenance decisions.

[0117] In this embodiment, the preset storage period is 365 days. The parsed signal parameters and generated interference warning information are uploaded to a ground server for distributed storage, with a storage time of no less than 365 days.

[0118] The specific implementation method includes: the onboard sensing terminal uploads the collected signal parameters (including RSRP, RSRQ, SINR, Cell ID, MAC address, etc.) and the generated interference warning information (including coverage blind spot alarm, signal interference alarm, interference source location alarm, etc.) to the ground server in real time; the ground server adopts a distributed storage architecture, storing data on multiple nodes to ensure data security and reliability; the storage time is no less than 365 days, and it supports the query and analysis of historical data.

[0119] The monitoring data is visualized through the network management terminal, which includes four main sections: public network monitoring, private network monitoring, WiFi monitoring, and spectrum monitoring. The public network monitoring section displays the coverage quality and interference of public 4G / 5G signals; the private network monitoring section displays the coverage quality and interference of LTE-M private network signals; the WiFi monitoring section displays the coverage quality and interference of wireless LAN signals; and the spectrum monitoring section displays a spectrum waterfall chart and spectrum status across all frequency bands. Simultaneously, the network management terminal supports real-time alarm information push to a large screen. When interference events are detected, alarm information is automatically pushed to the large screen for display, reminding maintenance personnel to handle the situation promptly.

[0120] In some embodiments, the method is applicable to rail transit communication environments and is used to manage and support the operation and maintenance of spectrum in scenarios where multiple systems such as public network signals, private network signals, and wireless local area network signals coexist.

[0121] In this embodiment, the method is specifically applicable to a rail transit wireless communication environment that supports a CBTC (Communication-Based Train Control) system. A CBTC system is an advanced train control system that enables real-time information exchange between the train and the ground control center via wireless communication, placing extremely high demands on the reliability and real-time performance of wireless communication.

[0122] In practical applications of CBTC systems, the wireless communication environment typically includes:

[0123] Public network signal: 4G / 5G mobile communication network, used for mobile communication services for train passengers.

[0124] Private network signal: LTE-M (Long Term Evolution - Metro) private network, used for train control, dispatching and passenger information systems in the CBTC system.

[0125] Wireless LAN signal: WiFi network used for wireless coverage services in stations and trains.

[0126] In some embodiments, the method further includes: using artificial intelligence algorithms to predict interference trends based on historical monitoring data, thereby transforming the operation and maintenance model from passive response to proactive prevention.

[0127] This embodiment uses artificial intelligence algorithms to predict interference trends based on historical monitoring data, realizing a transformation from a passive response to a proactive prevention operation and maintenance model. This effectively improves the accuracy and timeliness of interference early warning, solves the problem that traditional methods can only passively respond to interference events, and provides more accurate data support for subsequent interference prevention and operation and maintenance decisions.

[0128] In this embodiment, artificial intelligence algorithms are used to predict interference trends based on historical monitoring data. Specifically, the historical monitoring data includes signal parameters (RSRP, RSRQ, SINR, Cell ID, MAC address, etc.) and interference warning information (coverage blind zone alarm, signal interference alarm, interference source location alarm, etc.) from the past 365 days.

[0129] The artificial intelligence algorithm employs a Long Short-Term Memory (LSTM) network model, which is capable of capturing long-term dependencies in time-series data. The model input consists of historical monitoring data from the past 30 days, and the output is a prediction of the disturbance trend for the next 7 days. During model training, historical data from the past 365 days is used as the training set, and the model parameters are optimized through backpropagation to minimize the prediction error.

[0130] Example embodiments have been disclosed herein, and while specific terminology has been used, it is for illustrative purposes only and should be construed as such, and is not intended to be limiting. In some instances, it will be apparent to those skilled in the art that features, characteristics, and / or elements described in conjunction with particular embodiments may be used alone, or in combination with features, characteristics, and / or elements described in conjunction with other embodiments, unless otherwise expressly indicated. Therefore, those skilled in the art will understand that various changes in form and detail may be made without departing from the scope of the invention as set forth in the appended claims.

Claims

1. A method for vehicle-mounted wireless intelligent monitoring and real-time interference early warning, characterized in that, Includes the following steps: S1. Simultaneously collect wireless signals in the first preset frequency band within the track section using the full-band scanning module of the vehicle-mounted sensing terminal; S2. Utilize the multi-channel parallel demodulation channel integrated in the vehicle-mounted sensing terminal to synchronously analyze the collected wireless signals, obtaining parameter information for public network signals, private network signals, wireless local area network signals, and the full-band spectrum; S3. Based on the analyzed signal parameters, construct a signal quality heat map of the track section in real time, and use a dynamic interference early warning model for status assessment; S4. When the status assessment result meets the preset alarm threshold conditions, interference warning information is automatically generated and reported.

2. The method according to claim 1, characterized in that, The specific process of synchronous acquisition in step S1 includes: receiving radio frequency signals through a wideband antenna, and acquiring signals by the radio frequency module at a rate not lower than the preset scanning speed; distributing the acquired signal data in parallel to multiple independent demodulation channels to achieve synchronous processing of multi-band signals.

3. The method according to claim 1, characterized in that, The synchronous analysis in step S2 includes: performing cardless secure demodulation of public network signals through the first demodulation channel to obtain base station parameters including mobile country code, mobile network code, physical cell identifier, reference signal received power, and signal-to-interference-plus-noise ratio; analyzing private network signals through the second demodulation channel to extract private network parameters including cell identifier and reference signal received quality in real time; analyzing wireless local area network signals through the third demodulation channel to obtain network parameters including media access control address, channel distribution, and signal strength; and performing full-band spectrum scanning through the fourth demodulation channel to obtain parameter information including frequency points, signal strength, and spectral characteristics to generate a full-band spectrum waterfall plot and visualize the spectrum situation.

4. The method according to claim 3, characterized in that, The first demodulation channel performs cardless secure demodulation of public network signals, and directly parses base station parameters using a physically isolated radio frequency channel.

5. The method according to claim 3, characterized in that, Step S3, which involves constructing a real-time signal quality heatmap of the track section based on the analyzed signal parameters, includes: preprocessing the analyzed reference signal received power, reference signal received quality, and signal-to-interference-plus-noise ratio to remove outliers and noise interference; using a spatial interpolation algorithm to spatially interpolate the track section based on the preprocessed signal parameters to generate a continuous signal quality distribution map; and fusing the signal quality distribution map with the geographical information of the track section, converting the signal quality parameters into heatmap color gradients through color mapping to generate a signal quality heatmap, thereby achieving a visual display of signal quality.

6. The method according to claim 5, characterized in that, The state assessment using the dynamic interference early warning model in step S3 includes: inputting the signal quality heatmap, or a parameter matrix of reference signal received power, reference signal received quality, and signal-to-interference-plus-noise ratio extracted from the signal quality heatmap, into a pre-trained machine learning classification model; the machine learning classification model outputs a state classification result for the current wireless communication environment based on the input features; the state classification result includes normal state, coverage blind zone, co-channel interference, adjacent channel interference, and signal shielding; wherein, the criteria for determining the coverage blind zone is that the reference signal received power value is continuously lower than a preset coverage threshold; the criteria for determining co-channel interference and adjacent channel interference are that the signal-to-interference-plus-noise ratio is continuously lower than a preset interference threshold within the licensed frequency band, and that specific spectral characteristics exist; the criteria for determining signal shielding is that the licensed signal strength drops sharply within a preset time and is accompanied by the appearance of an unlicensed broadband strong signal.

7. The method according to claim 6, characterized in that, When the status assessment result meets the preset alarm threshold conditions, the interference warning information is automatically generated and reported in step S4, including: when the status assessment result identifies a coverage blind zone, a coverage blind zone alarm is triggered; when the status assessment result identifies co-channel interference or adjacent channel interference, a signal interference alarm is triggered; when the status assessment result identifies signal shielding, an interference source location alarm is triggered.

8. The method according to claim 1, characterized in that, The automatic generation and reporting of interference warning information in step S4 further includes: determining the geographical location of the interference source based on an interference source localization algorithm and assessing the interference intensity. The interference source localization algorithm uses a time difference of arrival (TDOA) localization method to calculate the geographical location of the interference source based on the signal TDOA of multiple vehicle-mounted sensing terminals. According to the interference intensity and the type of abnormal state, the interference warning information is divided into three levels: emergency alarm, important alarm, and general alarm. An interference intensity greater than a first alarm threshold is classified as an emergency alarm; an interference intensity greater than a second alarm threshold but less than or equal to the first alarm threshold is classified as an important alarm; and an interference intensity less than or equal to the second alarm threshold is classified as a general alarm. The interference warning information of different alarm levels is pushed through a preset push channel. Emergency alarms trigger audible and visual alarms and are pushed to the mobile terminals of maintenance personnel; important alarms are pushed to the mobile terminals of maintenance personnel; and general alarms are recorded in the historical alarm database and periodically pushed to maintenance personnel.

9. The method according to claim 1, characterized in that, The method also includes data storage and display, specifically including: uploading the parsed signal parameters and generated interference warning information to a ground server for distributed storage, with a storage time of not less than a preset storage period; and visually displaying the monitoring data through a network management terminal, which includes four major sections: public network monitoring, private network monitoring, WiFi monitoring, and spectrum monitoring, and supports large-screen push of real-time alarm information.

10. The method according to claim 1, characterized in that, The method also includes: using artificial intelligence algorithms to predict interference trends based on historical monitoring data.

Images