Networking millimeter wave radar railway foreign matter monitoring system and monitoring method thereof

The networkable millimeter wave radar system realizes three-dimensional positioning and angle information calculation of foreign objects on railway tracks. Combined with imaging systems and intelligent analysis algorithms, the high cost and insufficient real-time performance of traditional detection methods are solved, and the safety of railway transportation systems is improved. Sex and reliability.

CN118981016APending Publication Date: 2024-11-19CHONGQING UNIV OF POSTS & TELECOMM

Patent Information

Application Number
CN202411206784.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-30
Publication Date
2024-11-19

AI Technical Summary

Technical Problem

Traditional railway track foreign object detection methods have high costs, limited detection distance, large environmental impact, and insufficient real-time performance, which is difficult to meet the needs of modern railway transportation systems for efficient, safe and real-time monitoring.

Method used

The networkable millimeter-wave radar railway foreign object monitoring system is adopted to achieve the three-dimensional positioning and angle information calculation of the target through a single multi-send and multi-receive millimeter-wave radar. Combined with the imaging system and intelligent analysis algorithm, it provides real-time and accurate foreign object detection and early warning functions. .

Benefits of technology

It improves detection accuracy and real-time performance, can work effectively in harsh environments, reduces human resource consumption, reduces maintenance costs, and enhances the safety and reliability of railway transportation systems.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118981016A_ABST
    Figure CN118981016A_ABST
Patent Text Reader

Abstract

The invention relates to a networking millimeter wave radar railway foreign matter monitoring system and a monitoring method thereof. The networking millimeter wave radar railway foreign matter monitoring system comprises a host part installed near a rail and a server installed in a railway control center. The host part comprises a millimeter wave radar SoC module, a host FPGA processing module and a host wireless communication module. The server comprises a server wireless communication module, a server FPGA (Field Programmable Gate Array) processing module and display equipment; according to the invention, the FPGA processing module and the deep learning algorithm are combined, so that the millimeter wave radar system can realize high automation and intelligence. The FPGA processing module receives and analyzes radar data, and uses an advanced detection algorithm to identify and classify foreign matters on a track. The deep learning technology further improves the recognition accuracy and efficiency. According to the method, manual intervention can be remarkably reduced, the timeliness and accuracy of detection are improved, and stable detection performance can be kept in a complex environment.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of millimeter wave radars, and in particular relates to a railway foreign object monitoring system and a monitoring method thereof. Background Art

[0002] Railway transportation plays an important role in modern society, and ensuring the safety and reliability of railway tracks is crucial to ensuring the safety of passengers and goods. However, there may be various foreign objects on the railway tracks, such as stones, wood, animal carcasses, garbage, tools, broken tracks, etc. If these foreign objects are not discovered and cleaned up in time, they may cause train accidents, equipment damage, and traffic delays, which seriously threaten the safety and efficiency of the railway transportation system. Therefore, it is of great significance to conduct timely and accurate detection of foreign objects on railway tracks. Timely detection and removal of foreign objects can ensure the smooth operation of railway lines, reduce the probability of train accidents, improve the safety and reliability of train operation, and reduce traffic delays and maintenance costs.

[0003] In terms of ensuring the safety and reliability of railway tracks, traditional foreign body detection methods have the following shortcomings: Traditional manual inspection methods rely on inspectors to conduct regular inspections of railway tracks. Although this method is direct, it has obvious disadvantages. Manual inspection consumes a lot of human resources and is costly. In addition, the work efficiency of inspectors is limited, and it is impossible to achieve real-time monitoring of railway tracks. In addition, manual inspection is easily restricted by bad weather and complex terrain, and there is a risk of missed detection and false detection; ultrasonic radar technology uses the propagation characteristics of sound waves in the medium for detection, but the attenuation of ultrasonic signals in the air is large, and it is easily affected by environmental noise and weather conditions, which reduces the reliability and accuracy of detection. In addition, ultrasonic radar has weak detection capabilities for small-sized or low-reflectivity foreign bodies, which is easy to cause missed detection; laser radar technology detects objects by emitting lasers and receiving reflected signals, and has the characteristics of high resolution and high precision. However, laser radar equipment is expensive and complex to maintain, and it is difficult to deploy on a large scale for railway track detection; visual cameras detect foreign bodies through image acquisition and processing technology, which has the advantages of being intuitive and easy to understand. The effective detection distance of the camera is limited, making it difficult to achieve long-distance warning in a high-speed railway environment; secondly, the visual system is easily affected by changes in lighting conditions, such as strong light, shadows, and night time, which will reduce detection accuracy; thirdly, the image processing algorithm has high requirements for computing resources and poor real-time performance, making it difficult to cope with high-speed moving train environments.

[0004] In summary, traditional foreign body detection methods have their own limitations in railway track monitoring, including high cost, limited detection distance, great environmental impact, and lack of real-time performance. These shortcomings make it difficult for these traditional methods to meet the needs of modern railway transportation systems for efficient, safe, and real-time monitoring. Therefore, millimeter wave radar technology, as an emerging detection method, has become the preferred solution for railway track foreign body detection with its reliability in harsh environments, long-distance detection capabilities, and high precision.

[0005] After searching, the application publication number CN115755025A, a high-speed railway obstacle monitoring and early warning method based on millimeter wave radar, includes the following steps: 1) deploying two sets of millimeter wave radar modules in a group near the high-speed railway track monitoring point; 2) monitoring the foreign body intrusion into the track by the millimeter wave radar; 3) judging the authenticity of the foreign body intrusion into the track. The two millimeter wave radar modules in the same group are spaced apart; the two sets of millimeter wave radar modules alternately send continuous frequency modulation waves in adjacent time slots, and the emission direction is the possible intrusion direction of foreign bodies; when the first millimeter wave radar module sends a signal, the first and second millimeter wave radar modules simultaneously monitor the echo signal; when the second millimeter wave radar module sends a signal, the second and first millimeter wave radar modules simultaneously monitor the echo signal; the first millimeter wave radar module sends a signal in an odd time slot; the second millimeter wave radar module sends a signal in an even time slot; in an odd time slot, the foreign body distance solution is performed by the second millimeter wave radar module, and the abnormality monitoring is performed by the first millimeter wave radar module.

[0006] This patent is a railway radar foreign body monitoring based on millimeter-wave radar, which uses two millimeter-wave radars to transmit and receive at intervals to achieve target detection and lateral length estimation, but cannot accurately locate the target. In contrast, this patent uses a single multi-transmitter multi-receiver millimeter-wave radar. In contrast, this patent proposes a new method using a single multi-transmitter multi-receiver millimeter-wave radar, which calculates the angle information of the target by measuring the phase difference between the arrays, thereby achieving three-dimensional positioning of the target, which can not only locate the three-dimensional position of the object, but also determine the object's volume and other information.

[0007] In addition, since most foreign objects on railways pass through horizontally and the millimeter-wave emission direction is the direction in which foreign objects may invade, the two millimeter-wave radars of this patent are arranged facing each other, and the scanning aperture of the millimeter-wave radar is limited, which results in the detection range of a set of equipment in this patent being much smaller than the detection range of this patent.

[0008] In addition, the patent is entitled "A networkable millimeter-wave radar railway foreign body monitoring system and its monitoring method". A networking part consists of multiple hosts and a server. The host receives information and sends it to the server for simultaneous processing. The detection range of a host is 100 meters, and a network can cover a range of one kilometer or more. The host and the server can be freely paired to achieve a wide range of railway foreign body monitoring.

[0009] At the same time, the millimeter wave radar of the present invention is equipped with an imaging system, which can image suspicious foreign objects in real time, making monitoring more effective. Summary of the invention

[0010] The present invention aims to solve the above problems of the prior art. A networkable millimeter wave radar railway foreign body monitoring system and a monitoring method thereof are proposed. The technical solution of the present invention is as follows:

[0011] A networkable millimeter wave radar railway foreign body monitoring system, comprising: a host part installed near the rails and a server installed in the railway control center; the host part comprises: a millimeter wave radar SoC module, a host FPGA processing module, and a host wireless communication module; the server comprises: a server wireless communication module, a server FPGA processing module, and a display device; wherein,

[0012] The millimeter wave radar SoC module is used to transmit millimeter wave signals and receive returned reflected waves; the host FPGA processing module is used to receive and process the return signal data from the millimeter wave radar SoC module, including demodulating, filtering and denoising the signal; the host wireless communication module is used to transmit the data processed by the host FPGA processing module to the wireless communication module installed at the railway center service end over a long distance, and receive control and feedback signals from the service end;

[0013] The server-side wireless communication module is used to receive radar signals processed by the FPGA processing module of the host part, and transmit these data to the FPGA processing module of the server for further processing and analysis, and send the control and feedback signals of the server to the host part; the server-side FPGA processing module is used to receive and analyze the data transmitted from the host part, summarize and integrate the radar signals from multiple hosts, and pass the synthesized positioning data to the subsequent intelligent analysis algorithm; once the intelligent analysis algorithm identifies a potential dangerous target, the FPGA processing module sends the result to the display device of the server, and sends an alarm to the host part installed near the rail through the wireless communication module of the server.

[0014] Furthermore, the host FPGA processing module is used to receive and process the return signal data from the millimeter wave radar SoC module, including demodulation, filtering and denoising operations on the signal, specifically including: the radar will transmit a frequency modulated continuous wave, that is, a continuous wave signal whose frequency changes linearly with time. The frequency of the signal reflected by the target object will also change with time. The time delay generated by the distance is used to compare the difference frequency between the transmitted signal and the received signal to calculate the target distance; in addition, the radar will use an array antenna for beamforming and direction measurement; by comparing the phase difference of the received signals of different antennas, the azimuth and pitch angle of the target can be calculated; the radar can also obtain the speed of the target through the Doppler effect. Since the relative motion of the target will cause the frequency of the echo signal to change, the speed of the target is calculated by analyzing the frequency change of the received signal; by analyzing the characteristics of the signal, possible foreign objects are identified, and frame data suspected of having foreign objects are marked; once the data frame suspected of having foreign objects is determined, the FPGA processing module packages these frame data and transmits them to the railway control center service end through the wireless communication module.

[0015] Furthermore, the intelligent analysis algorithm specifically includes:

[0016] Data preprocessing: Preprocess the original radar signal to remove noise and extract useful information. Target detection: Use a constant false alarm rate target detection algorithm to detect the presence of foreign objects. Feature extraction: Extract key features such as position, speed, etc. from the detected target signal. Classification and recognition: Use a deep learning model to classify and recognize targets. Trajectory initialization: Initialize the trajectory for the newly detected target. State update: Use the Kalman filter method to update the state estimate of the target. Data association: Determine the association between radar measurements and targets. Anomaly detection: Monitor whether the target behavior is abnormal and trigger an alarm if there is an abnormality.

[0017] Furthermore, the millimeter-wave radar SoC module uses a multi-chip MIMO radar sparse array based on an E-band SoC, each MIMO unit includes 3 transmitting antennas and 4 receiving antennas, the transmitting antennas are evenly arranged at 4 times the wavelength, and the receiving antennas are evenly arranged at 1 times the wavelength; when each transmitting antenna is working, all receiving antennas work at the same time, and a total of 12 equivalent sampling points can be obtained, that is, the midpoint of the line connecting the transmitting and receiving antennas, that is, the equivalent phase center point, and the equivalent sampling points are spaced apart by half a wavelength; then, a scanning line array is designed based on the MIMO unit; in order to ensure that equivalent sampling points are evenly arranged at half a wavelength in the array direction, two MIMO units are used to form an array unit, and the two MIMO units in each array unit are 180 degrees rotationally symmetric and the nearest transceiver unit interval between adjacent MIMO units is half a wavelength.

[0018] Furthermore, the host FPGA processing module uses the Hummingbird E203SoC based on the RISC-V instruction set architecture as a platform, and uses the high scalability of the RISC-V instruction set to control the positioning acceleration through the Nice bus to process the original radar signal, and on this basis, the slide control system and millimeter wave data acquisition circuit can be added according to needs, so as to achieve three-dimensional imaging of the target. The slide control system uses the FPGA end as the main control and uses a counter to generate a pulse signal with a fixed clock cycle. It is connected to the motor controller through the pins of the FPGA to control the motor to accurately push the millimeter wave radar sensor to slide at the millimeter level. Accurate motion control of the radar sensor is achieved. The millimeter wave data acquisition circuit transmits the echo signal of the target to be measured collected by the radar sensor back to the FPGA through the high-speed SPI interface for processing, and finally achieves three-dimensional imaging of the target.

[0019] Furthermore, after receiving the original radar frame data from the millimeter-wave radar SoC module, the server-side FPGA processing module performs a fast Fourier transform on the radar data to extract the target's distance and speed information, and obtains the target's distance, Doppler frequency shift and angle information by converting the time domain signal to the frequency domain; then, a constant false alarm rate target detection algorithm is used to analyze the frequency domain data, identify possible target echo signals, and extract relevant parameters; multiple-input multiple-output MIMO technology is used to further calculate the target's angle information, and the signals of multiple receiving antennas are spatially processed to determine the target's azimuth and elevation; then, the detected target distance, speed and angle information are integrated to generate the target's three-dimensional spatial coordinates to form point cloud data, where each point cloud represents the spatial position of a target; in addition, the FPGA will also perform post-processing on the generated point cloud data, including noise removal and clustering analysis.

[0020] Furthermore, the constant false alarm rate detection target detection algorithm specifically includes:

[0021] Select a target unit, select reference units around it and exclude units in the protection band to avoid the influence of the target signal; use the data of the reference unit to estimate the statistical characteristics of the background noise; calculate the detection threshold according to the predetermined false alarm probability; finally, compare the signal strength of the target unit with the threshold. If the signal strength is greater than the threshold, it is determined that the target signal exists in the target unit, otherwise it is determined that the target signal does not exist; repeat this process for all units to be detected.

[0022] Furthermore, the host wireless communication module uses an e22 communication module based on the ROLA protocol. The ROLA protocol achieves low power consumption and long-distance wireless communication through linear frequency modulation spread spectrum technology and time division multiple access technology.

[0023] A railway foreign body monitoring method based on any of the above systems comprises the following steps:

[0024] a) Signal transmission and reception: Millimeter wave radar SoC module: transmits millimeter wave signals and receives signals reflected from the track;

[0025] b) Preliminary signal processing: FPGA processing module: demodulates, filters and denoises the received signal, identifies possible foreign objects, and marks frame data suspected of foreign objects;

[0026] c) Data transmission: Host wireless communication module: Packs the marked frame data and transmits it to the service end of the railway control center over a long distance;

[0027] d) Data reception: The server-side wireless communication module receives the packaged data from the host part and transmits it to the server-side FPGA processing module;

[0028] e) Data aggregation and analysis: The server-side FPGA processing module aggregates and analyzes the data received from multiple hosts to generate synthetic positioning data; it performs intelligent analysis on the synthetic positioning data to identify potential dangerous targets;

[0029] f) Result output and alarm: The analysis results are displayed on the display device, and an alarm signal is sent to the host part through the wireless communication module of the server.

[0030] The advantages and beneficial effects of the present invention are as follows:

[0031] The present invention adopts a multi-chip MIMO radar sparse array based on E-band SoC, and each MIMO unit contains 3 transmitting antennas and 4 receiving antennas. The array unit composed of two MIMO units ensures that the equivalent sampling points are evenly distributed at half the wavelength, optimizes the array design, and improves the accuracy of angle estimation. This design can more accurately estimate the location information of the target, including distance, azimuth and pitch angle, thereby improving the positioning accuracy of the system, thereby helping to timely detect and locate foreign objects on the track and ensure the safety of train operation.

[0032] The host FPGA processing module uses a SoC based on the RISC-V instruction set architecture as a platform and integrates intelligent analysis algorithms. Through real-time signal processing and intelligent algorithms, it can quickly identify and classify targets. Using the high scalability of the RISC-V instruction set, other functions can be expanded as needed, improving the accuracy of detection.

[0033] The system adopts a distributed architecture of the host part and the server side, and realizes data transmission through the wireless communication module. It can realize remote monitoring and management, and facilitate centralized control and maintenance. The use of the communication module based on the ROLA protocol ensures low power consumption and long-distance wireless communication, enhancing the practicality and stability of the system. In the railway foreign body monitoring, the entire system can remotely monitor the monitoring sites distributed in different locations, improving the management and maintenance efficiency of the system.

[0034] The millimeter wave radar technology used can maintain efficient working ability even in adverse weather conditions. The millimeter wave radar has strong penetrating power and can effectively penetrate environmental factors such as rain, snow, fog, etc. to ensure all-weather monitoring. This is very important for railway foreign body monitoring, because no matter how the weather conditions change, the system can continue to work effectively and ensure the safe operation of the railway. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a schematic diagram of the installation layout of a preferred embodiment provided by the present invention;

[0036] Figure 2 It is a work flow chart of the present invention;

[0037] Figure 3 It is a schematic diagram of the host part of the present invention.

[0038] Figure 4 Schematic diagram of the detection effect of the present invention DETAILED DESCRIPTION

[0039] The following will describe the technical solutions in the embodiments of the present invention in detail in conjunction with the accompanying drawings in the embodiments of the present invention. The described embodiments are only part of the embodiments of the present invention.

[0040] The technical solution of the present invention to solve the above technical problems is:

[0041] Embodiment: The present invention provides a railway foreign body monitoring system, such as Figure 1 As shown, it includes: a host part installed near the rails and a service end installed in the railway control center;

[0042] The host part includes: millimeter wave radar SoC module, FPGA processing module, wireless communication module;

[0043] The server includes: a wireless communication module, an FPGA processing module, and a display device;

[0044] The millimeter wave radar SoC module is used to transmit millimeter wave signals and receive the returned reflected waves. This module integrates the RF front end, signal processing and control logic, and can generate high-frequency millimeter wave signals and accurately receive the reflected signals, so as to accurately detect and locate foreign objects on the railway track.

[0045] The FPGA processing module of the host part is used to receive and process the return signal data from the millimeter wave radar SoC module, including demodulating, filtering and denoising the signal to enhance the clarity and accuracy of the signal. At the same time, the FPGA processing module will analyze the time domain and frequency domain characteristics of the signal, identify possible foreign objects, and mark the frame data suspected of having foreign objects. Once the data frame suspected of having foreign objects is determined, the FPGA processing module will package the frame data and transmit it to the railway control center service end through the wireless communication module.

[0046] The wireless communication module of the host part is used to transmit the data processed by the FPGA processing module of the host part to the wireless communication module installed at the railway center service end over a long distance, and receive control and feedback signals from the service end.

[0047] The wireless communication module of the server is used to receive radar signals processed by the FPGA processing module of the host part, transmit these data to the FPGA processing module of the server for further processing and analysis, and send control and feedback signals of the server to the host part.

[0048] The FPGA processing module on the server side is used to receive and analyze the data transmitted from the host part, summarize and integrate the radar signals from multiple hosts, and pass the synthesized positioning data to the subsequent intelligent analysis algorithm. Once the intelligent analysis algorithm identifies a potential dangerous target, the FPGA processing module sends the result to the display device on the server side and sends an alarm to the host part installed near the rail through the wireless communication module on the server side.

[0049] The host part is installed on both sides of the railway at intervals of 80 meters. The red shaded area indicates the track range covered by a radar device. The effective range of each radar is 100 meters, and the coverage ranges of two adjacent radars will overlap by 20 meters.

[0050] The millimeter wave radar SoC module uses a multi-chip MIMO radar sparse array based on E-band SoC. The chip peak power consumption is as low as 50W, and the continuous power consumption is less than 2W. Each MIMO unit contains 3 transmitting antennas (Tx1, Tx2, Tx3) and 4 receiving antennas (Rx1, Rx2, Rx3, Rx4). The transmitting antennas are evenly arranged at 4 times the wavelength, and the receiving antennas are evenly arranged at 1 times the wavelength. When each transmitting antenna is working, all receiving antennas work at the same time, and a total of 12 equivalent sampling points (the midpoint of the connecting line of the transmitting and receiving antennas, that is, the equivalent phase center point) can be obtained, and the equivalent sampling points are spaced at half a wavelength. Then, the scanning line array is designed based on the MIMO unit. In order to ensure that the equivalent sampling points are evenly arranged at half a wavelength in the array direction, this project will cleverly use two MIMO units to form an array unit. The two MIMO units in each array unit are 180 degrees rotationally symmetrical and the nearest transceiver unit interval between adjacent MIMO units is half a wavelength.

[0051] The FPGA processing module uses the Hummingbird E203 SoC based on the RISC-V instruction set architecture as a platform, and uses the high scalability of the RISC-V instruction set to control the positioning acceleration through the Nice bus to process the original radar signal. It includes modules such as data reception, data analysis, detection and recognition algorithms, and display enhancement. On this basis, a slide control system and a millimeter-wave data acquisition circuit can be added according to needs to form an integrated millimeter-wave edge detection system for signal acquisition, data processing, and result warning.

[0052] After receiving the raw radar frame data from the TI6843 millimeter-wave radar SoC module, the FPGA processing module performs a fast Fourier transform on the radar data to extract the distance and speed information of the target, and obtains the distance, Doppler shift and angle information of the target by converting the time domain signal to the frequency domain. After that, the FPGA uses target detection algorithms such as constant false alarm rate detection to analyze the frequency domain data, identify possible target echo signals, and extract relevant parameters. Using multiple-input multiple-output (MIMO) technology, the angle information of the target is further calculated, and the signals of multiple receiving antennas are spatially processed to determine the azimuth and elevation of the target. Subsequently, the detected target distance, speed and angle information are integrated to generate the three-dimensional spatial coordinates of the target to form point cloud data, each point cloud representing the spatial position of a target. In addition, the FPGA will also post-process the generated point cloud data, including removing noise points, clustering analysis, etc., to ensure the accuracy and reliability of the point cloud data.

[0053] The detection algorithm and target recognition algorithm described above, combined with artificial intelligence deep learning technology, can achieve efficient detection and recognition of foreign objects on railways. In practical applications such as image analysis and speech understanding, pattern recognition processing requires convolution calculation of each pixel of the image in a very short time, and analysis and comparison of the calculation results to obtain reliable detection conclusions. Traditional radar recognition methods usually rely on software of ordinary computers to process static data. Although theoretical results can be obtained, their response speed is not enough to meet the needs of actual engineering. To address this problem, the use of high-speed parallel computing architecture can significantly improve the efficiency and judgment ability of traditional recognition methods. Through the local processing power of edge computing, the FPGA processing module can ensure high-speed response and high accuracy of the recognition process. In complex external environments, the system can still reliably identify targets.

[0054] The wireless communication module uses the e22 communication module based on the ROLA protocol. The ROLA protocol realizes low-power and long-distance wireless communication through linear frequency modulation spread spectrum technology and time division multiple access technology. Its working principle involves signal modulation and spread spectrum, effective allocation of time, and error detection and correction during data transmission. All these mechanisms together ensure the efficiency and reliability of the ROLA protocol in low-power wide area networks. The host device installed every 80 meters can transmit data such as warnings at all locations on all rails within one kilometer to the server side at a frequency of 400MHz through the wireless communication module. Each server is responsible for summarizing the data within one kilometer, and then sending it to the main monitoring room through optical fiber or other communication means, thereby realizing foreign object detection on the entire track.

[0055] The overall system workflow is:

[0056] 1. Signal transmission and reception: Millimeter wave radar SoC module: transmits millimeter wave signals and receives signals reflected from the track.

[0057] 2. Preliminary signal processing: FPGA processing module: demodulates, filters and denoises the received signal, identifies possible foreign objects, and marks frame data suspected of being foreign objects.

[0058] 3. Data packaging and transmission: Wireless communication module (host part): Pack the marked frame data and transmit it over long distances to the server of the railway control center.

[0059] 4. Data reception: Wireless communication module (server): Receives packaged data from the host part and transmits it to the FPGA processing module on the server side.

[0060] 5. Data aggregation and analysis: FPGA processing module (server): Aggregate and analyze the data received from multiple hosts to generate synthetic positioning data. Perform intelligent analysis on the synthetic positioning data to identify potential dangerous targets.

[0061] 6. Result output and alarm: The analysis results are displayed on the display device, and an alarm signal is sent to the host part through the wireless communication module of the server.

[0062] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions.

[0063] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0064] The above embodiments should be understood to be only used to illustrate the present invention and not to limit the protection scope of the present invention. After reading the contents of the present invention, technicians can make various changes or modifications to the present invention, and these equivalent changes and modifications also fall within the scope defined by the claims of the present invention.

Claims

1. A networkable millimeter wave radar railway foreign body monitoring system, characterized in that: include: The main unit installed near the rails and the server installed in the railway control center; The host part includes: a millimeter wave radar SoC module, a host FPGA processing module, and a host wireless communication module; the server includes: a server wireless communication module, a server FPGA processing module, and a display device; wherein, The millimeter wave radar SoC module is used to transmit millimeter wave signals and receive returned reflected waves; the host FPGA processing module is used to receive and process the return signal data from the millimeter wave radar SoC module, including demodulating, filtering and denoising the signal; the host wireless communication module is used to transmit the data processed by the host FPGA processing module to the wireless communication module installed at the railway center service end over a long distance, and receive control and feedback signals from the service end; The server-side wireless communication module is used to receive radar signals processed by the FPGA processing module of the host part, and transmit these data to the FPGA processing module of the server for further processing and analysis, and send the control and feedback signals of the server to the host part; the server-side FPGA processing module is used to receive and analyze the data transmitted from the host part, summarize and integrate the radar signals from multiple hosts, and pass the synthesized positioning data to the subsequent intelligent analysis algorithm; once the intelligent analysis algorithm identifies a potential dangerous target, the FPGA processing module sends the result to the display device of the server, and sends an alarm to the host part installed near the rail through the wireless communication module of the server.

2. A networkable millimeter wave radar railway foreign body monitoring system according to claim 1, characterized in that: The host FPGA processing module is used to receive and process the return signal data from the millimeter wave radar SoC module, including demodulation, filtering and denoising operations on the signal, specifically including: the radar will transmit a frequency modulated continuous wave, that is, a continuous wave signal whose frequency changes linearly with time, and the frequency of the signal reflected by the target object will also change with time. The time delay generated by the distance is used to compare the difference frequency between the transmitted signal and the received signal to calculate the target distance; in addition, the radar will use an array antenna for beamforming and direction measurement; by comparing the phase difference of the received signals of different antennas, the azimuth and pitch angle of the target can be calculated; the radar can also obtain the speed of the target through the Doppler effect. Since the relative motion of the target will cause the frequency of the echo signal to change, the speed of the target is calculated by analyzing the frequency change of the received signal; by analyzing the characteristics of the signal, possible foreign objects are identified, and frame data suspected of having foreign objects are marked; once the data frame suspected of having foreign objects is determined, the FPGA processing module packages these frame data and transmits them to the railway control center service end through the wireless communication module.

3. A networkable millimeter wave radar railway foreign body monitoring system according to claim 1, characterized in that: The intelligent analysis algorithm specifically includes: Data preprocessing: preprocess the original radar signal, remove noise, and extract useful information. Target detection: use constant false alarm rate target detection algorithm to detect whether there is foreign matter; feature extraction: extract key features from the detected target signal, including position and speed; classification and recognition: use deep learning model to classify and recognize targets; trajectory initialization: initialize trajectory for newly detected targets; state update: use Kalman filter method to update the state estimate of the target; data association: determine the association between radar measurement and target; anomaly detection: monitor whether the target behavior is abnormal, and trigger an alarm if there is an abnormality.

4. The networkable millimeter wave radar railway foreign body monitoring system according to claim 1 is characterized in that: The millimeter-wave radar SoC module uses a multi-chip MIMO radar sparse array based on an E-band SoC, each MIMO unit includes three transmitting antennas and four receiving antennas, the transmitting antennas are evenly arranged at four times the wavelength, and the receiving antennas are evenly arranged at one time the wavelength; when each transmitting antenna is working, all receiving antennas work at the same time, and a total of 12 equivalent sampling points can be obtained, that is, the midpoint of the line connecting the transmitting and receiving antennas, that is, the equivalent phase center point, and the equivalent sampling point interval is half a wavelength; then, a scanning line array is designed based on the MIMO unit; in order to ensure that equivalent sampling points are evenly arranged at half a wavelength in the array direction, two MIMO units are used to form an array unit, and the two MIMO units in each array unit are 180 degrees rotationally symmetrical and the nearest transceiver unit interval between adjacent MIMO units is half a wavelength.

5. The networkable millimeter wave radar railway foreign body monitoring system according to claim 1 is characterized in that: The host FPGA processing module uses the Hummingbird E203 SoC based on the RISC-V instruction set architecture as a platform, and uses the high scalability of the RISC-V instruction set to control the positioning acceleration through the Nice bus to process the original radar signal. On this basis, the slide control system and the millimeter wave data acquisition circuit can be added according to the needs, so as to realize the three-dimensional imaging of the target. The slide control system uses the FPGA end as the main control and uses a counter to generate a pulse signal with a fixed clock cycle. It is connected to the motor controller through the pins of the FPGA to control the motor to accurately push the millimeter wave radar sensor to slide at the millimeter level. Accurate motion control of the radar sensor is realized. The millimeter wave data acquisition circuit transmits the echo signal of the target to be measured collected by the radar sensor back to the FPGA through the high-speed SPI interface for processing, and finally realizes the three-dimensional imaging of the target.

6. The networkable millimeter wave radar railway foreign body monitoring system according to claim 1 is characterized in that: After receiving the original radar frame data from the millimeter-wave radar SoC module, the server-side FPGA processing module performs a fast Fourier transform on the radar data to extract the distance and speed information of the target, and obtains the distance, Doppler frequency shift and angle information of the target by converting the time domain signal to the frequency domain; then, a constant false alarm rate target detection algorithm is used to analyze the frequency domain data, identify possible target echo signals, and extract relevant parameters; multiple-input multiple-output MIMO technology is used to further calculate the angle information of the target, and the signals of multiple receiving antennas are spatially processed to determine the azimuth and elevation of the target; Subsequently, the detected target distance, speed and angle information are integrated to generate the three-dimensional spatial coordinates of the target, forming point cloud data, where each point cloud represents the spatial position of a target; in addition, FPGA will also perform post-processing on the generated point cloud data, including noise removal and cluster analysis.

7. A networkable millimeter wave radar railway foreign body monitoring system according to claim 6, characterized in that: The constant false alarm rate detection target detection algorithm specifically includes: Select a target unit, select reference units around it and exclude units in the protection band to avoid the influence of the target signal; use the data of the reference unit to estimate the statistical characteristics of the background noise; calculate the detection threshold according to the predetermined false alarm probability; finally, compare the signal strength of the target unit with the threshold. If the signal strength is greater than the threshold, it is determined that the target signal exists in the target unit, otherwise it is determined that the target signal does not exist; repeat this process for all units to be detected.

8. The networkable millimeter wave radar railway foreign body monitoring system according to claim 6, characterized in that: The host wireless communication module uses an e22 communication module based on the ROLA protocol. The ROLA protocol achieves low power consumption and long-distance wireless communication through linear frequency modulation spread spectrum technology and time division multiple access technology.

9. A railway foreign body monitoring method based on the system according to any one of claims 1 to 8, characterized in that: The following steps are involved: a) Signal transmission and reception: Millimeter wave radar SoC module: transmits millimeter wave signals and receives signals reflected from the track; b) Preliminary signal processing: FPGA processing module: demodulates, filters and denoises the received signal. Identify possible foreign objects and mark the frame data of suspected foreign objects; c) Data transmission: Host wireless communication module: Packs the marked frame data and transmits it to the service end of the railway control center over a long distance; d) Data reception: The server-side wireless communication module receives the packaged data from the host part and transmits it to the server-side FPGA processing module; e) Data aggregation and analysis: The server-side FPGA processing module aggregates and analyzes the data received from multiple hosts to generate synthetic positioning data; performs intelligent analysis on the synthetic positioning data. Identify potentially dangerous targets; f) Result output and alarm: The analysis results are displayed on the display device, and an alarm signal is sent to the host part through the wireless communication module of the server.

Citation Information

Patent Citations

  • Method and device of railway roadblock detection and alarm based on radar return characteristics

    CN103033808A

  • Airport runway foreign object debris detection device and airport runway foreign object debris detection method

    CN105549110A

  • Railway locomotive running ground environment sensing technology

    CN111681383A

  • 3D radar scanning imaging system based on MEMS

    CN112859038A

  • High-dynamic railway environment radar sensing method based on fixed grid

    CN114924274A

Cited By

  • Millimeter wave radar FOD detection system based on centralized operation

    CN121878688A