Railway dedicated communication intelligent guarantee method

Abstract

The invention provides a special communication intelligent guarantee method for a railway, which comprises the following steps of: determining a train running line and acquiring geographical data along the line of the train running line; dividing a train operation line based on the geographic data along the line to obtain a plurality of communication interference areas; when the train travels to the vicinity of the communication interference area, dynamically adjusting the boundary of the communication interference area by combining the environmental data of the communication interference area; predicting entering and exiting moments when the train passes through the boundary of the communication interference area, and determining a communication switching moment by combining the entering and exiting moments; selecting a corresponding communication mode for switching at a communication switching moment in combination with different types of communication interference areas; the boundary of the communication interference area is dynamically adjusted by collecting the environmental data of the train near the communication interference area, so that too early or too late switching of the communication modes can be avoided, normal communication of the train inside and outside the communication interference area is ensured, and when a special communication mode is selected for switching based on different types of the communication interference area, the communication mode switching efficiency is improved. And normal operation of train communication can be further ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of communication guarantee, and in particular to an intelligent guarantee method for railway-specific communication. Background Art

[0002] With the rapid development of the railway industry, train speeds are constantly increasing, and train density is continuously increasing. The stability and reliability of railway communication systems are becoming increasingly critical. Whether it is the transmission of dispatching instructions within the train control system, the interaction between onboard equipment and the ground, or the smooth flow of emergency communications, all rely on stable communication links. During operation, trains pass through a variety of terrains and infrastructure. Different altitudes, slopes, and different infrastructure can affect communication during railway operation. Traditional railway communication assurance methods often rely on static threshold trigger mechanisms. For example, a fixed distance threshold is preset 500 meters before the tunnel entrance. When the train's GPS positioning reaches this distance, it is forced to switch to tunnel-specific communication mode. In addition, some lines still use post-repair strategies. This means that when a communication interruption is detected, manual scheduling is used to restart equipment or switch links. The recovery period often exceeds 30 seconds. While these methods can meet basic communication needs, they are difficult to adapt to dynamically changing scenarios. Fixed thresholds cannot cope with the increased signal attenuation caused by the environment. In high-speed, high-density operation scenarios, post-repair methods are prone to scheduling delays due to communication interruptions, even endangering driving safety. Summary of the Invention

[0003] In view of this, the present invention proposes an intelligent guarantee method for railway-specific communications, which can dynamically adjust the boundaries of communication interference zones based on environmental changes and adopt targeted communication modes to ensure communications.

[0004] The technical solution of the present invention is achieved as follows: A railway dedicated communication intelligent guarantee method comprises the following steps: Step S1: Determine the train route and obtain geographical data along the train route; Step S2: Divide the train route based on geographical data along the route to obtain several communication interference areas; Step S3: When the train travels near the communication interference area, dynamically adjust the communication interference area boundary based on the environmental data of the communication interference area; Step S4: predicting the entry and exit times of the train crossing the communication interference zone boundary, and determining the communication switching time based on the entry and exit times; Step S5: Based on the different types of communication interference areas, select the corresponding communication mode for switching at the communication switching moment.

[0005] Preferably, the specific steps of step S1 include: Step S11: Obtain a train dispatch table and determine the train route based on the train number, starting point, departure time, and arrival time; Step S12: obtaining topographic data along the train route using satellite remote sensing technology, wherein the topographic data includes altitude, slope, and land cover type; Step S13: Obtain infrastructure distribution data along the train route, including specific location information of bridges, tunnels, stations along the route, and signal towers.

[0006] Preferably, the step S1 further includes: Step S14: standardize the format of the topographic data and infrastructure distribution data, and perform coordinate system 1, eliminate outliers and supplement missing data to form geographic data along the route.

[0007] Preferably, the specific steps of step S2 include: Step S21: Acquire historical communication abnormality data of the train during operation, wherein the communication abnormality data includes the location of communication interruption and signal attenuation during operation, the corresponding terrain altitude, vegetation density, signal strength, and bit error rate; Step S22: performing unsupervised classification on the communication anomaly data using a K-means clustering algorithm, dividing multiple types of communication interference areas according to the clustering results, and setting a standard area template containing typical geographical feature parameters and signal feature parameters for each communication interference area; Step S23: input the collected geographic data along the line and the standard area template into a support vector machine, determine the matching degree by calculating the cosine similarity, and determine the specific location and type of the communication interference area on the train running line based on the matching degree; Step S24: input the type of communication interference area and the corresponding geographical data along the line into the trained convolutional neural network, and the convolutional neural network outputs the boundary of the communication interference area.

[0008] Preferably, the types of the communication interference areas include terrain obstruction type, multipath interference type and electromagnetic interference type.

[0009] Preferably, the specific steps of step S3 include: Step S31: Set a buffer warning zone outside the boundary of the entrance side of the communication interference zone, and collect real-time environmental data when the train travels into the buffer warning zone. The real-time environmental data includes rainfall, electromagnetic intensity, wind speed, and vegetation shielding degree; Step S32: training the mapping relationship between historical environmental data and boundary offsets through a BP neural network, and establishing a dynamic adjustment benchmark model; Step S33: input the real-time environmental data into the dynamic adjustment benchmark model, output the predicted offset of the boundary adjustment, and use the Kalman filter algorithm to filter the noise of the predicted offset to obtain a smoothed actual offset; Step S34: superimpose the actual offset onto the boundary of the communication interference area.

[0010] Preferably, after collecting the real-time environmental data, step S31 performs data cleaning on the real-time environmental data, and uses principal component analysis to perform dimensionality reduction and principal component extraction, and the extracted principal components serve as input to the dynamic adjustment benchmark model.

[0011] Preferably, the specific steps of step S4 include: Step S41: collecting the speed of the train near the entrance of the communication interference area, and predicting the time when the train reaches the boundary of the entrance of the communication interference area by using a linear regression algorithm; Step S42: Calculate the time when the train leaves the exit boundary of the communication interference area based on the length of the communication interference area and the current speed of the train; Step S43: Determine a safety time threshold based on the type of the communication interference area, and calculate the communication switching time based on the safety time threshold, the entry time, and the exit time.

[0012] Preferably, the specific steps of step S43 are: subtracting the safety time threshold from the entry time, and adding the safety time threshold to the exit time to obtain the communication switching time.

[0013] Preferably, step S5 includes the following steps: Step S51: When the type of the communication interference area is terrain shielding type, select the low-frequency band communication mode or the leaky cable transmission mode; Step S52: When the communication interference area is of multipath interference type, select a multi-antenna MIMO communication mode or beamforming technology; Step S53: When the communication interference area is of electromagnetic interference type, select a frequency hopping communication mode or a spread spectrum communication technology.

[0014] Compared with the prior art, the present invention has the following beneficial effects: The present invention provides a railway-specific intelligent communication guarantee method. On a train running route, geographical data along the route is collected. Then, the train running route can be divided according to the geographical data along the route. Areas in the running route that may affect train communication are divided and recorded as communication interference areas. When a train travels into a communication interference area, it is necessary to adopt a different communication mode. To ensure timely switching, after calculating the entry and exit times of the train crossing the communication interference area boundary, the communication switching time can be calculated. At the communication switching time, the train communication mode is switched to ensure normal communication. For each communication interference zone, the external environment will cause the scope of the communication interference zone to expand. Therefore, when the train travels near the communication interference zone, the environmental data of the corresponding area can be collected, and the boundary of the communication interference zone can be dynamically adjusted based on the environmental data to accurately obtain the communication switching time, thereby accurately switching the communication mode and avoiding delayed switching or premature switching. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only preferred embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0016] Figure 1 This is a flow chart of a railway dedicated communication intelligent guarantee method of the present invention; Figure 2 This is a flow chart of step S1 of a railway dedicated communication intelligent guarantee method of the present invention; Figure 3 This is a flow chart of step S2 of a railway dedicated communication intelligent guarantee method of the present invention; Figure 4 This is a flow chart of step S3 of a railway dedicated communication intelligent guarantee method of the present invention; Figure 5 This is a flow chart of step S4 of a railway dedicated communication intelligent guarantee method of the present invention; Figure 6 This is a flow chart of step S5 of a railway dedicated communication intelligent guarantee method of the present invention; DETAILED DESCRIPTION In order to better understand the technical content of the present invention, a specific embodiment is provided below, and the present invention is further described in conjunction with the accompanying drawings.

[0017] See also Figures 1 to 6 The present invention provides a railway dedicated communication intelligent guarantee method, comprising the following steps: Step S1: Determine the train route and obtain geographical data along the train route; Step S2: Divide the train route based on geographical data along the route to obtain several communication interference areas; Step S3: When the train travels near the communication interference area, dynamically adjust the communication interference area boundary based on the environmental data of the communication interference area; Step S4: predicting the entry and exit times of the train crossing the communication interference zone boundary, and determining the communication switching time based on the entry and exit times; Step S5: Based on the different types of communication interference areas, select the corresponding communication mode for switching at the communication switching moment.

[0018] The present invention provides a railway-specific intelligent communication guarantee method for timely switching of communication modes when a train passes through an interference area. First, the train that needs communication guarantee is determined, and the train's running route is obtained. Based on the known train running route, the geographical data along the running route is collected. Then, the areas on the running route that may cause interference to communication can be known through the geographical data along the line. After the train running route is divided, several communication interference areas can be obtained. When the train travels to the communication interference area, the normal communication will be affected by the external environment, and the change of the external environment will cause the boundary of the communication interference area to change. If the predefined communication interference area is used, the communication between the train and the surrounding area will be affected. When the communication switching time is selected based on the boundary of the communication interference zone, a switching delay will occur, causing the train to be interfered with before switching the communication mode. Therefore, the present invention can dynamically adjust the boundary of the communication interference zone. When the train travels near the communication interference zone, real-time environmental data of the communication interference zone can be collected. Based on the real-time environmental data, the boundary of the communication interference zone can be dynamically adjusted. After determining the boundary of the communication interference zone, the entry and exit times of the train entering and leaving the communication interference zone can be predicted, and then the communication switching time is calculated based on the entry and exit times. At the communication switching time, different communication modes are selected for switching based on the different types of communication interference zones to ensure normal communication.

[0019] Preferably, the specific steps of step S1 include: Step S11: Obtain a train dispatch table and determine the train route based on the train number, starting point, departure time, and arrival time; Step S12: obtaining topographic data along the train route using satellite remote sensing technology, wherein the topographic data includes altitude, slope, and land cover type; Step S13: Obtaining infrastructure distribution data along the train route, including specific location information of bridges, tunnels, stations along the route, and signal towers; Step S14: standardize the format of the topographic data and infrastructure distribution data, and perform coordinate system 1, eliminate outliers and supplement missing data to form geographic data along the route.

[0020] After obtaining the train dispatch table, the relevant information of each train can be queried from the train dispatch table. The basic information such as the schedule and starting point of the train that needs communication guarantee can be queried from the train dispatch table to the corresponding train route. By combining the train route with satellite remote sensing technology, the topographic data along the train route can be collected, including the altitude, slope and surface cover type of each location passed. At the same time, the infrastructure distribution data along the line can be obtained through urban planning, such as bridges, tunnels, stations along the line and signal towers. For topographic data, the difference in altitude will have a greater impact on the transmission of communication signals. If the train is located between two mountains, the mountain will also affect the signal. It will cause obstruction, and the type and coverage of surface vegetation will also affect communication. For infrastructure distribution data, when the train enters the tunnel, the original communication mode cannot be carried out due to the obstruction of the tunnel itself and needs to be changed. Similarly, facilities such as stations and signal towers along the line will also affect communication. When the train line passes through these areas, obvious communication anomalies will occur. Therefore, it is necessary to accurately collect topographic data and infrastructure distribution data. After obtaining the topographic data and infrastructure distribution data, data preprocessing can be performed, including format standardization, coordinate system one, elimination of outliers, supplementation of missing data, etc. After ensuring the accuracy of the data, geographic data along the line can be formed.

[0021] Preferably, the specific steps of step S2 include: Step S21: Acquire historical communication abnormality data of the train during operation, wherein the communication abnormality data includes the location of communication interruption and signal attenuation during operation, the corresponding terrain altitude, vegetation density, signal strength, and bit error rate; Step S22: performing unsupervised classification on the communication anomaly data using a K-means clustering algorithm, dividing multiple types of communication interference areas based on the clustering results, and setting a standard area template containing typical geographical feature parameters and signal feature parameters for each communication interference area. The types of communication interference areas include terrain obstruction type, multipath interference type, and electromagnetic interference type; Step S23: input the collected geographic data along the line and the standard area template into a support vector machine, determine the matching degree by calculating the cosine similarity, and determine the specific location and type of the communication interference area on the train running line based on the matching degree; Step S24: input the type of communication interference area and the corresponding geographical data along the line into the trained convolutional neural network, and the convolutional neural network outputs the boundary of the communication interference area.

[0022] Before dividing the communication interference area, it is necessary to obtain the communication anomaly data of the vertical vehicle during historical operation, and then process the communication anomaly data through the K-means clustering algorithm to analyze the terrain altitude, vegetation density, etc. of the location where the communication anomaly or interruption occurs. Through clustering, it can be divided into multiple types of communication interference areas, among which the types of communication interference areas include terrain obstruction type, multipath interference type and electromagnetic interference type. Terrain obstruction type includes mountainous areas and tunnel groups, multipath interference type includes high-rise dense areas and canyon areas, and electromagnetic interference type includes areas near high-voltage power grids and around industrial areas. At the same time, each communication interference area will form a corresponding standard area template. In the standard area template Typical geographic feature parameters and signal feature parameters are set in the model, and then a support vector machine model is constructed. The collected geographic data along the line and the standard area template are input into the support vector machine. The support vector machine compares the feature parameters and calculates the cosine similarity between the feature vectors. The matching degree can be obtained according to the cosine similarity, so that the specific location and type of the communication interference area on the train running line can be determined. Different types of communication interference areas have different boundary ranges due to differences in their own geographic data. After the type of communication interference area and the corresponding geographic data along the line are input into the trained convolutional neural network, the boundary of the corresponding communication interference area can be automatically identified.

[0023] Preferably, the specific steps of step S3 include: Step S31: Set a buffer warning zone outside the boundary of the entrance side of the communication interference zone, and collect real-time environmental data when the train travels into the buffer warning zone. The real-time environmental data includes rainfall, electromagnetic intensity, wind speed, and vegetation shielding degree; Step S32: training the mapping relationship between historical environmental data and boundary offsets through a BP neural network, and establishing a dynamic adjustment benchmark model; Step S33: input the real-time environmental data into the dynamic adjustment benchmark model, output the predicted offset of the boundary adjustment, and use the Kalman filter algorithm to filter the noise of the predicted offset to obtain a smoothed actual offset; Step S34: superimpose the actual offset onto the boundary of the communication interference area.

[0024] The boundary of the communication interference zone will change with environmental differences. The adjustment of the boundary of the communication interference zone needs to be carried out when the train is about to enter the communication interference zone. Therefore, a buffer warning zone is set outside the boundary on the entrance side of the communication interference zone. For example, the buffer warning zone is within 1 kilometer of the boundary. When the train enters the buffer warning zone, a dynamic adjustment instruction can be triggered. At this time, it is necessary to collect real-time environmental data near the communication interference zone. Different real-time environmental data have different impacts on the boundary of the communication interference zone. The present invention introduces a BP neural network. After training historical environmental data, a mapping relationship with the boundary offset can be obtained, and a dynamic adjustment benchmark model is constructed at the same time. After the real-time environmental data near the communication interference zone is input into the dynamic adjustment benchmark model, the predicted offset of the boundary adjustment can be output. Then, the predicted offset is noise-filtered by the Kalman filter algorithm to eliminate the error caused by environmental data fluctuations and obtain a smooth actual offset. After the actual offset is moved to the boundary of the communication interference zone, dynamic adjustment of the boundary can be achieved.

[0025] Preferably, after collecting the real-time environmental data, step S31 performs data cleaning on the real-time environmental data, and uses principal component analysis to perform dimensionality reduction and principal component extraction, and the extracted principal components serve as input to the dynamic adjustment benchmark model.

[0026] The collected real-time environmental data needs to be preprocessed and feature extracted. Outliers can be eliminated during data cleaning. After the principal components are extracted using principal component analysis, the extracted principal components can be input as feature vectors into the dynamic adjustment benchmark model to ensure that the dynamic adjustment benchmark model can accurately process and obtain the predicted offset.

[0027] Preferably, the specific steps of step S4 include: Step S41: collecting the speed of the train near the entrance of the communication interference area, and predicting the time when the train reaches the boundary of the entrance of the communication interference area by using a linear regression algorithm; Step S42: Calculate the time when the train leaves the exit boundary of the communication interference area based on the length of the communication interference area and the current speed of the train; Step S43: Determine the safety time threshold based on the type of communication interference zone, and calculate the communication switching time based on the safety time threshold, the entry time, and the exit time. The specific steps are: subtract the safety time threshold from the entry time, and add the safety time threshold to the exit time to obtain the communication switching time.

[0028] By collecting the train's positioning information, it is possible to determine whether the train is approaching a communication interference zone. When the train reaches the buffer warning zone, the speed of the train can be collected, and then the time when the train reaches the boundary of the communication interference zone can be predicted, which is recorded as the entry time. After the train enters the communication interference zone, the length of the communication interference zone and the current speed of the train in the communication interference zone can be combined to simultaneously calculate the train's exit time from the communication interference zone. For different types of communication interference zones, the switching time will also vary due to different communication modes. Based on the type of communication interference zone, a safety time threshold can be determined. Then, a communication switching time is set before the entry and exit times. That is, the safety time threshold is subtracted from the entry and exit times and added to the safety time threshold, respectively, to obtain the communication switching time. At the communication switching time before the entry time, the dedicated communication mode is switched to ensure normal communication within the communication interference zone. At the communication switching time after the exit time, the train switches back to the normal communication mode to ensure normal communication outside the communication interference zone.

[0029] Preferably, step S5 includes the following steps: Step S51: When the type of the communication interference area is terrain shielding type, select the low-frequency band communication mode or the leaky cable transmission mode; Step S52: When the communication interference area is of multipath interference type, select a multi-antenna MIMO communication mode or beamforming technology; Step S53: When the communication interference area is of electromagnetic interference type, select a frequency hopping communication mode or a spread spectrum communication technology.

[0030] Different communication modes need to be switched for different types of communication interference areas. When the communication interference area is blocked by terrain, you can choose a low-frequency band (such as 700MHz-900MHz) communication mode or a leaky cable transmission mode. This is because low-frequency electromagnetic waves have strong diffraction capabilities and low attenuation when penetrating obstacles. Leaky cables can form continuous signal coverage along the track, avoiding signal interruption caused by terrain blockage. When the communication interference area is multipath interference, such as areas with dense high-rise buildings or bridges, you can choose multi-antenna MIMO communication mode or beamforming technology. By using multiple antennas to simultaneously transmit and receive signals, the signal superposition interference caused by multipath reflection can be offset. Beamforming can focus the signal energy in the direction of the train, reducing the impact of scattered signals.

[0031] When the communication interference area is electromagnetic interference, such as near high-voltage power grids or industrial plants, select frequency hopping communication mode or spread spectrum communication technology (such as direct sequence spread spectrum). The frequency hopping mode avoids electromagnetic interference in specific frequency bands by quickly switching communication frequencies. Spread spectrum technology spreads the signal to a wide frequency band, reducing the impact of narrowband electromagnetic interference on signal analysis and ensuring the integrity of communication data.

[0032] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A railway dedicated communication intelligent guarantee method, characterized in that: The following steps are involved: Step S1: Determine the train route and obtain geographical data along the train route; Step S2: Divide the train route based on geographical data along the route to obtain several communication interference areas; Step S3: When the train travels near the communication interference area, dynamically adjust the communication interference area boundary based on the environmental data of the communication interference area; Step S4: predicting the entry and exit times of the train crossing the communication interference zone boundary, and determining the communication switching time based on the entry and exit times; Step S5: Based on the different types of communication interference areas, select the corresponding communication mode for switching at the communication switching moment.

2. A railway dedicated communication intelligent guarantee method according to claim 1, characterized in that: The specific steps of step S1 include: Step S11: Obtain a train dispatch table and determine the train route based on the train number, starting point, departure time, and arrival time; Step S12: obtaining topographic data along the train route using satellite remote sensing technology, wherein the topographic data includes altitude, slope, and land cover type; Step S13: Obtain infrastructure distribution data along the train route, including specific location information of bridges, tunnels, stations along the route, and signal towers.

3. The method for intelligent guarantee of railway-specific communication according to claim 2, characterized in that: The step S1 further includes: Step S14: standardize the format of the topographic data and infrastructure distribution data, and perform coordinate system 1, eliminate outliers and supplement missing data to form geographic data along the route.

4. The method for intelligent guarantee of railway-specific communication according to claim 1, characterized in that: The specific steps of step S2 include: Step S21: Acquire historical communication abnormality data of the train during operation, wherein the communication abnormality data includes the location of communication interruption and signal attenuation during operation, the corresponding terrain altitude, vegetation density, signal strength, and bit error rate; Step S22: performing unsupervised classification on the communication anomaly data using a K-means clustering algorithm, dividing multiple types of communication interference areas according to the clustering results, and setting a standard area template containing typical geographical feature parameters and signal feature parameters for each communication interference area; Step S23: input the collected geographic data along the line and the standard area template into a support vector machine, determine the matching degree by calculating the cosine similarity, and determine the specific location and type of the communication interference area on the train running line based on the matching degree; Step S24: input the type of communication interference area and the corresponding geographical data along the line into the trained convolutional neural network, and the convolutional neural network outputs the boundary of the communication interference area.

5. A railway dedicated communication intelligent guarantee method according to claim 4, characterized in that: The types of communication interference areas include terrain obstruction type, multipath interference type and electromagnetic interference type.

6. The method for intelligent guarantee of railway-specific communication according to claim 1, characterized in that: The specific steps of step S3 include: Step S31: Set a buffer warning zone outside the boundary of the entrance side of the communication interference zone, and collect real-time environmental data when the train travels into the buffer warning zone. The real-time environmental data includes rainfall, electromagnetic intensity, wind speed, and vegetation shielding degree; Step S32: training the mapping relationship between historical environmental data and boundary offsets through a BP neural network, and establishing a dynamic adjustment benchmark model; Step S33: input the real-time environmental data into the dynamic adjustment benchmark model, output the predicted offset of the boundary adjustment, and use the Kalman filter algorithm to filter the noise of the predicted offset to obtain a smoothed actual offset; Step S34: superimpose the actual offset onto the boundary of the communication interference area.

7. A railway dedicated communication intelligent guarantee method according to claim 6, characterized in that: After collecting the real-time environmental data, the step S31 performs data cleaning on the real-time environmental data, and uses the principal component analysis method to perform dimensionality reduction and principal component extraction. The extracted principal components are used as inputs of the dynamic adjustment benchmark model.

8. The method for intelligent guarantee of railway-specific communication according to claim 1, characterized in that: The specific steps of step S4 include: Step S41: collecting the speed of the train near the entrance of the communication interference area, and predicting the time when the train reaches the boundary of the entrance of the communication interference area by using a linear regression algorithm; Step S42: Calculate the time when the train leaves the exit boundary of the communication interference area based on the length of the communication interference area and the current speed of the train; Step S43: Determine a safety time threshold based on the type of the communication interference area, and calculate the communication switching time based on the safety time threshold, the entry time, and the exit time.

9. The method for intelligent guarantee of railway-specific communication according to claim 8, characterized in that: The specific steps of step S43 are: subtracting the safety time threshold from the entry time, and adding the safety time threshold to the exit time to obtain the communication switching time.

10. The method for intelligent guarantee of railway-specific communication according to claim 5, characterized in that: The step S5 comprises the following steps: Step S51: When the type of the communication interference area is terrain shielding type, select the low-frequency band communication mode or the leaky cable transmission mode; Step S52: When the communication interference area is of multipath interference type, select a multi-antenna MIMO communication mode or beamforming technology; Step S53: When the communication interference area is of electromagnetic interference type, select a frequency hopping communication mode or a spread spectrum communication technology.

Images