A vehicle communication abnormal signal detection method and system in a vehicle networking environment
Patent Information
- Application Number
- CN202611016778.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-09
- Publication Date
- 2026-09-22
AI Technical Summary
但易受以下异常影响：恶意节点伪造虚假位置/速度信息（如伪造紧急刹车信号）、信道拥塞导致的信号丢包或畸变
[0056]本发明实施例还提供了一种车联网环境下的车辆通信异常信号检测方法及系统。
Smart Images
Figure CN122802636A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of vehicle communication technology, and more specifically, to a method and system for detecting abnormal vehicle communication signals in a vehicle networking environment. Background Technology
[0002] Vehicle-to-everything (V2X) systems refer to the collection, storage, and transmission of all vehicle operating conditions and static / dynamic information through onboard terminal devices installed on the vehicle's dashboard. V2X systems typically feature real-time scene tracking and utilize mobile networks for human-vehicle interaction. Vehicles broadcast basic safety messages (BSM) via technologies such as DSRC / C-V2X. Wireless radio frequency signals are acquired through V2X (Vehicle to Everything) communication technology within the V2X system. These signals include communication data between the vehicle and the V2X network. However, they are susceptible to the following anomalies: malicious nodes forging false location / speed information (such as spoofed emergency braking signals), and signal loss or distortion due to channel congestion. If any dimension of the information does not match the preset information, an anomaly in the V2X signal can be identified.
[0003] However, how to detect preset information and compare it with the actual information to determine whether there is an anomaly during the process of time change is a problem. Summary of the Invention
[0004] The purpose of this invention is to provide a method and system for detecting abnormal vehicle communication signals in a vehicle-to-everything (V2X) environment, in order to solve the above-mentioned problems existing in the prior art.
[0005] In a first aspect, embodiments of the present invention provide a method for detecting abnormal vehicle communication signals in a vehicle-to-everything (V2X) environment, comprising:
[0006] Acquire a vehicle location image; the vehicle location image represents an image marking the vehicle's location at n time points; the vehicle location represents the vehicle's position.
[0007] Based on the vehicle location image, a 3D vehicle image and a decrypted vehicle location image are obtained through encryption and decryption; the decrypted vehicle location image is the image that has been decrypted after being transmitted to the vehicle network.
[0008] Based on the vehicle's 3D image and the decrypted vehicle location image, determine whether the transmitted signal is abnormal;
[0009] If the transmission signal is normal, based on the decrypted vehicle location image, the data processing time is determined, and multiple vehicle images are obtained; the vehicle images represent images of the vehicle taken by the corresponding camera on the road after the data processing time has elapsed.
[0010] Based on the decrypted vehicle location image and multiple vehicle images, it is determined whether the time communication signal is abnormal.
[0011] Optionally, obtaining a 3D vehicle image and a decrypted vehicle position image based on the vehicle position image through encryption and decryption includes:
[0012] Based on the vehicle position image, different positions are segmented to obtain a 3D vehicle image and an interpolated vehicle image; the 3D vehicle image represents an image where the vehicle's position is continuous at adjacent time points;
[0013] In the width direction, the 3D image of the vehicle is cut to obtain multiple images at the cutting positions;
[0014] The image of the cutting location is converted into a cutting location vector; the values in the cutting location vector include vehicle pixel values and vehicle pixel positions; the vehicle pixel values represent the values in the cutting location image; the vehicle pixel positions represent the positions of the vehicle pixel values in the cutting location image;
[0015] The cutting position vector is encrypted to obtain an encrypted vehicle vector; multiple encrypted vehicle vectors are obtained corresponding to multiple cutting position images; the encrypted vehicle vector is a value that has not been transmitted to the vehicle network.
[0016] Transmit multiple encrypted vehicle vectors to the vehicle network;
[0017] Based on multiple encrypted vehicle vectors, decryption is performed to obtain the decrypted vehicle location image.
[0018] Optionally, if the transmission signal is normal, based on the decrypted vehicle location image, the data processing time is determined, and multiple vehicle images are obtained, including:
[0019] The transmission time length and decryption time point are obtained; the decryption time point is the time point when the vehicle location image is acquired and decrypted; the transmission time represents the length from the time point when the vehicle location image is acquired to the time point when decryption is performed.
[0020] Based on the decryption time point and the decrypted vehicle location image, the predicted vehicle location is obtained through a location discrimination network; the transmission time length represents the predicted vehicle location at the time point when the decrypted vehicle location image is acquired.
[0021] Based on the predicted vehicle location, a matching process is performed to obtain a road data table; the road data table stores images taken by camera equipment on the road where the predicted vehicle location is located at multiple time points;
[0022] The detection period is obtained based on the decryption time point and the transmission time length; the starting point of the detection period is the difference between the decryption time point and the quotient of the transmission time length divided by the segment value; the starting point of the detection period is the sum of the decryption time point and the quotient of the transmission time length divided by the segment value.
[0023] In the road database, multiple vehicle images for the detection time period were found.
[0024] Optionally, determining whether the time communication signal is abnormal based on the decrypted vehicle location image and multiple vehicle images includes:
[0025] The vehicle image is input into a target detection network to identify the user's vehicle, obtaining a discrimination value and the identified vehicle location; a discrimination value of 1 indicates the presence of the user's vehicle; a discrimination value of 0 indicates the absence of the user's vehicle.
[0026] If the discrimination value is 1, the time point corresponding to the vehicle location is taken as the user's time point; the user's time point represents the time point when the user's vehicle was captured in the vehicle image;
[0027] The determined vehicle location is converted into a geographic location to obtain the user-determined geographic location; the user-determined geographic location represents the geographic location of the vehicle detected in the vehicle image.
[0028] Based on multiple identical user time points, the user's geographical location is determined and the vehicle's location is predicted to determine whether the time communication signal is abnormal.
[0029] Optionally, determining whether the transmitted signal is abnormal based on the vehicle's 3D image and the decrypted vehicle position image includes:
[0030] With a stride of 1, a 2*2*m two-dimensional convolution kernel is convolved with the vehicle interpolation image to obtain the vehicle value feature vector.
[0031] With a stride of 1, a 2*2*m two-dimensional convolution kernel is convolved with the decrypted vehicle location image to obtain the decrypted vehicle feature vector.
[0032] The similarity between the vehicle value feature vector and the decrypted vehicle feature vector is calculated to obtain the feature similarity value;
[0033] If the feature similarity value is less than the similarity threshold, the transmission signal will be set to normal.
[0034] If the feature similarity value is greater than or equal to the similarity threshold, the transmission signal is set to abnormal.
[0035] Optionally, the step of determining whether the time communication signal is abnormal based on the user's geographical location and predicted vehicle location corresponding to multiple identical user time points includes:
[0036] The distance between the user's identified geographical location and the predicted vehicle location corresponding to the same user time point is calculated to obtain the distance difference; multiple distance differences are obtained for multiple user time points;
[0037] Multiple distance differences are clustered, and the cluster centers are used as distance influence deviation values; the distance influence deviation values represent the noise generated by the location discrimination network and the target detection network.
[0038] If the absolute value of the quotient of the difference between the distance difference and the distance influence deviation value divided by the distance influence deviation value is less than the vehicle distance difference threshold, the time communication signal is set to normal.
[0039] If the absolute value of the difference between the distance difference and the distance influence deviation value is greater than or equal to the vehicle distance difference threshold, it is set as an abnormal time communication signal.
[0040] Optionally, the step of segmenting different locations based on the vehicle position image to obtain a 3D vehicle image and a vehicle interpolated image includes:
[0041] Multiple vehicle positions in the vehicle position image are interpolated to obtain an interpolated vehicle image; the interpolated vehicle image includes multiple interpolated positions; the interpolated position represents the predicted position of the vehicle between two adjacent vehicle positions at two time points.
[0042] Optionally, in the vehicle interpolation image, the pixels between vehicle positions at adjacent time points are retained, while other pixels are deleted to obtain a vehicle segmentation image; n-1 vehicle segmentation images are obtained for n time points;
[0043] The n-1 vehicle segmentation images are superimposed to form a 3D vehicle image, arranged from morning to night according to time points.
[0044] Optionally, the step of obtaining the predicted vehicle position based on the decryption time point and the decrypted vehicle position image through a position discrimination network includes:
[0045] In chronological order, multiple vehicle positions from the decrypted vehicle position image are input into a position discrimination network to predict the position at the next time point, thus obtaining the first predicted position and the first predicted time point.
[0046] The schematic diagrams of the location discrimination network and the target detection network are shown below. Figure 2 As shown.
[0047] In chronological order, the first predicted position and multiple vehicle positions in the decrypted vehicle position image are input into the position discrimination network to predict the position at the next time point, thus obtaining the second predicted position and the second predicted time point.
[0048] By performing multiple convolutions through a location discrimination network, the predicted vehicle position and the corresponding predicted time point are obtained; the predicted time point is equal to the decryption time point.
[0049] Secondly, embodiments of the present invention provide a vehicle communication anomaly signal detection system in a vehicle-to-everything (V2X) environment, comprising:
[0050] The acquisition module is used to acquire vehicle location images; the vehicle location images represent images of vehicle locations marked at n time points; the vehicle location represents the vehicle's position.
[0051] The transmission module is used to obtain a three-dimensional image of the vehicle and a decrypted vehicle location image based on the vehicle location image through encryption and decryption; the decrypted vehicle location image is the image that has been decrypted after being transmitted to the vehicle network.
[0052] The transmission anomaly detection module is used to determine whether the transmission signal is abnormal based on the vehicle's 3D image and the decrypted vehicle position image.
[0053] The time change detection module is used to determine the data processing time based on the decrypted vehicle position image if the transmission signal is normal, and obtain multiple vehicle images; the vehicle images represent images of the vehicle taken by the corresponding camera on the road after the data processing time has elapsed;
[0054] The time communication anomaly detection module is used to determine whether the time communication signal is abnormal based on the decrypted vehicle location image and multiple vehicle images.
[0055] Compared with the prior art, the embodiments of the present invention achieve the following beneficial effects:
[0056] This invention also provides a method and system for detecting abnormal vehicle communication signals in a vehicle-to-everything (V2X) environment.
[0057] In this invention, the non-adjacent pixels in the vehicle location image at each time point are supplemented. Furthermore, a three-dimensional vehicle image is obtained by segmenting between adjacent time points. After segmenting the three-dimensional vehicle image along its width, it is encrypted in the order of top to bottom and left to right according to the segmentation location image. After encrypted transmission, the accuracy of the transmitted signal is determined by comparing it with the decrypted vehicle location image. If the transmitted signal is accurate, the vehicle's position at a future time point is predicted. This allows for the location of the user's vehicle at approximately the approximate time point. After more accurately locating the vehicle's position using the vehicle image, and after removing network noise from the predicted vehicle position, it is determined whether an anomaly occurred during the data transmission and processing time. This approach, considering both transmission anomalies and sudden anomalies during data processing, achieves a more accurate anomaly detection effect. Attached Figure Description
[0058] Figure 1 This is a flowchart of a method for detecting abnormal vehicle communication signals in a vehicle-to-everything (V2X) environment provided by an embodiment of the present invention.
[0059] Figure 2 This is a schematic diagram of the position discrimination network and the target detection network in a vehicle communication anomaly signal detection method in a vehicle-to-everything (V2X) environment provided by an embodiment of the present invention.
[0060] Figure 3 This is a schematic diagram illustrating the relationship between the length and width of a vehicle in a three-dimensional image and the road in a vehicle communication anomaly signal detection method provided in an embodiment of the present invention. Detailed Implementation
[0061] The present invention will now be described in detail with reference to the accompanying drawings.
[0062] Example 1
[0063] like Figure 1 As shown, this embodiment of the invention provides a method for detecting abnormal vehicle communication signals in a vehicle-to-everything (V2X) environment. The method includes:
[0064] S101: Obtain vehicle location image; the vehicle location image represents an image of the vehicle's location marked at n time points; the vehicle location represents the vehicle's position.
[0065] Where n is a positive integer.
[0066] The vehicle's location is determined using GPS.
[0067] The vehicle location image is a grayscale image; the vehicle location image contains multiple roads; in this embodiment, the background of the vehicle location image is white with a corresponding grayscale value of 255, the roads are gray with a corresponding grayscale value of 125, and the vehicle location is black with a corresponding grayscale value of 0.
[0068] S102: Based on the vehicle location image, a three-dimensional image of the vehicle and a decrypted vehicle location image are obtained through encryption and decryption; the decrypted vehicle location image is the image that has been decrypted after being transmitted to the vehicle network.
[0069] The decrypted vehicle location image is a three-dimensional image.
[0070] S103: Based on the encrypted and decrypted vehicle location images, determine whether the transmitted signal is abnormal;
[0071] S104: If the transmission signal is normal, based on the decrypted vehicle position image, determine the data processing time and obtain multiple vehicle images; the vehicle images represent images captured by the corresponding cameras on the road after the data processing time has elapsed.
[0072] One time point corresponds to one vehicle image.
[0073] S105: Based on the decrypted vehicle location image and multiple vehicle images, determine whether the time communication signal is abnormal.
[0074] Optionally, obtaining a 3D vehicle image and a decrypted vehicle position image based on the vehicle position image through encryption and decryption includes:
[0075] Based on the vehicle position image, different positions are segmented to obtain a three-dimensional vehicle image and a vehicle interpolated image; the three-dimensional vehicle image represents an image in which the vehicle position is continuous at adjacent time points.
[0076] In this 3D vehicle image, the length corresponds to the geographical location of the road, the width represents the geographical direction of travel on the road, and the height represents the continuous vehicle positions at multiple time points. The relationship between the length and width of the vehicle image and the road is as follows: Figure 3 As shown.
[0077] In the width direction, the 3D image of the vehicle is cut to obtain multiple images at the cutting positions;
[0078] The image at the cutting location is converted into a cutting location vector; the values in the cutting location vector include vehicle pixel values and vehicle pixel positions; the vehicle pixel values represent the values in the image at the cutting location; and the vehicle pixel positions represent the positions of the vehicle pixel values in the image at the cutting location.
[0079] In this process, the image of the cutting position is converted into a cutting position vector in the order from top to bottom and from left to right.
[0080] The cutting position vector is encrypted to obtain an encrypted vehicle vector; multiple encrypted vehicle vectors are obtained corresponding to multiple cutting position images; the encrypted vehicle vector is a value that has not been transmitted to the vehicle network.
[0081] In this embodiment, the encryption algorithm is the RSA algorithm.
[0082] Transmit multiple encrypted vehicle vectors to the vehicle network;
[0083] Based on multiple encrypted vehicle vectors, decryption is performed to obtain the decrypted vehicle location image.
[0084] The decrypted vehicle location image refers to an image of the same size as the vehicle's three-dimensional image, which is formed by decrypting the encrypted vehicle vector and combining it according to the cutting order.
[0085] Optionally, if the transmission signal is normal, based on the decrypted vehicle location image, the data processing time is determined, and multiple vehicle images are obtained, including:
[0086] The transmission time length and decryption time point are obtained; the decryption time point is the time point when the vehicle location image is acquired and decrypted; the transmission time represents the length from the time point when the vehicle location image is acquired to the time point when decryption is performed.
[0087] Based on the decryption time point and the decrypted vehicle location image, the predicted vehicle location is obtained through a location discrimination network; the transmission time length represents the predicted vehicle location at the time point when the decrypted vehicle location image is acquired.
[0088] The decryption time point is later than the time points corresponding to the multiple vehicle positions in the decrypted vehicle position image;
[0089] Based on the predicted vehicle location, a matching process is performed to obtain a road data table; the road data table stores images taken by camera equipment on the road where the predicted vehicle location is located at multiple time points.
[0090] One camera device is fixed in one location to detect vehicles in a fixed area along a road. Because the camera's field of view is limited, multiple cameras are used along a road to capture images of different areas. The images captured by different cameras are stored in different data tables, which serve as the road data table.
[0091] The detection period is obtained based on the decryption time point and the transmission time length; the starting point of the detection period is the difference between the decryption time point and the quotient of the transmission time length divided by the segment value; the starting point of the detection period is the sum of the decryption time point and the quotient of the transmission time length divided by the segment value.
[0092] In this embodiment, the segment value of 5 indicates that the judgment is made using 1 / 5 of the transmission time length.
[0093] In the road database, multiple vehicle images for the detection time period were found.
[0094] Optionally, determining whether the time communication signal is abnormal based on the decrypted vehicle location image and multiple vehicle images includes:
[0095] The vehicle image is input into the target detection network to identify the user's vehicle, and a discrimination value and the identified vehicle location are obtained; the discrimination value of 1 indicates that the user's vehicle exists; the discrimination value of 0 indicates that the user's vehicle does not exist.
[0096] In this embodiment, the target detection network is a YOLOv7 model trained on license plates. By detecting the location of the vehicle, the license plate is detected in the image containing the vehicle area, and the license plate number is extracted to determine whether it belongs to the user's vehicle. If it does, the vehicle's location is used as the vehicle's location.
[0097] If the discrimination value is 1, the time point corresponding to the vehicle location is taken as the user's time point; the user's time point represents the time point when the user's vehicle is captured in the vehicle image.
[0098] This is because the user's vehicle is also moving while the user is uploading encrypted vehicle vectors for data transmission and processing. By analyzing the changes in vehicle position in the vehicle location image, the vehicle's position at a future time point can be predicted. The predicted future vehicle position is then used to locate the corresponding camera device for the user's vehicle. Detecting the vehicle's position from the camera device's image allows for more accurate vehicle location determination.
[0099] The determined vehicle location is converted into a geographic location to obtain the user-determined geographic location; the user-determined geographic location represents the geographic location of the vehicle detected in the vehicle image.
[0100] In this process, since the vehicle location is determined by its position in the vehicle image, the position in the vehicle image can be associated with a geographic location. For example, based on the determined vehicle location, the geographic location of the corresponding road can be found, and then a mapping relationship can be constructed between the position in the vehicle image and the geographic location to convert the determined vehicle location into a geographic location.
[0101] Based on multiple identical user time points, the user's geographical location is determined and the vehicle's location is predicted to determine whether the time communication signal is abnormal.
[0102] Optionally, the step of determining whether the time communication signal is abnormal based on the user's geographical location and predicted vehicle location corresponding to multiple identical user time points includes:
[0103] The distance between the user's identified geographical location and the decrypted vehicle's location corresponding to the same user time point is calculated to obtain the distance difference; multiple distance differences are obtained for multiple user time points;
[0104] Clustering multiple distance differences yields the distance influence bias value.
[0105] The distance influence deviation value represents the deviation between the time point when the encrypted vehicle vector is issued for more accurate positioning and the time point when the vehicle is located.
[0106] If the absolute value of the quotient of the difference between the distance difference and the distance influence deviation value divided by the distance influence deviation value is less than the vehicle distance difference threshold, the time communication signal is considered normal.
[0107] In this embodiment, the vehicle distance difference threshold is 0.01.
[0108] If the absolute value of the quotient of the difference between the distance difference and the distance influence deviation value divided by the distance influence deviation value is greater than or equal to the vehicle distance difference threshold, it is set as a time communication signal abnormality.
[0109] Optionally, the step of segmenting different locations based on the vehicle position image to obtain a 3D vehicle image and a vehicle interpolated image includes:
[0110] Multiple vehicle positions in the vehicle position image are interpolated to obtain an interpolated vehicle image; the interpolated vehicle image includes multiple interpolated positions; the interpolated position represents the predicted position of the vehicle between two adjacent vehicle positions at two time points.
[0111] In this embodiment, a bilinear interpolation algorithm is used.
[0112] The vehicle interpolation image represents the position of multiple vehicles in the vehicle position image.
[0113] Since the vehicle positions in the vehicle position image are discrete points at adjacent time points, there will be cases where they are not adjacent when represented by pixels. Therefore, interpolation is used to supplement the pixels between the vehicle positions at two adjacent time points.
[0114] Since each pixel in the vehicle interpolation image has at least one adjacent pixel, it is equivalent to a continuous curve.
[0115] Optionally, in the vehicle interpolation image, the pixels between vehicle positions at adjacent time points are retained, while other pixels are deleted to obtain a vehicle segmentation image; n-1 vehicle segmentation images are obtained for n time points;
[0116] The n-1 vehicle segmentation images are superimposed to form a 3D vehicle image, arranged from morning to night according to time points.
[0117] Optionally, determining whether the transmitted signal is abnormal based on the vehicle's 3D image and the decrypted vehicle position image includes:
[0118] With a stride of 1, a 2*2*m two-dimensional convolution kernel is convolved with the vehicle interpolation image to obtain the vehicle value feature vector.
[0119] With a stride of 1, a 2*2*m two-dimensional convolution kernel is convolved with the decrypted vehicle location image to obtain the decrypted vehicle feature vector.
[0120] The similarity between the vehicle value feature vector and the decrypted vehicle feature vector is calculated to obtain the feature similarity value.
[0121] In this embodiment, Euclidean distance is used to calculate similarity.
[0122] If the feature similarity value is less than the similarity threshold, the transmission signal is set to normal.
[0123] In this embodiment, the similarity threshold is 0.1.
[0124] If the feature similarity value is greater than or equal to the similarity threshold, the transmission signal is set to abnormal.
[0125] Optionally, the step of obtaining the predicted vehicle position based on the decryption time point and the decrypted vehicle position image through a position discrimination network includes:
[0126] In chronological order, multiple vehicle positions from the decrypted vehicle position image are input into a position discrimination network to predict the position at the next time point, thus obtaining the first predicted position and the first predicted time point.
[0127] In this embodiment, the location discrimination network is a temporal convolutional network (TCN).
[0128] In chronological order, the first predicted position and multiple vehicle positions in the decrypted vehicle position image are input into the position discrimination network to predict the position at the next time point, thus obtaining the second predicted position and the second predicted time point.
[0129] By performing multiple convolutions through a location discrimination network, the predicted vehicle position and the corresponding predicted time point are obtained; the predicted time point is equal to the decryption time point.
[0130] Example 2
[0131] Based on the above-described method for detecting abnormal vehicle communication signals in a vehicle-to-everything (V2X) environment, this invention also provides a system for detecting abnormal vehicle communication signals in a V2X environment, the system comprising:
[0132] The acquisition module is used to acquire vehicle location images; the vehicle location images represent images of vehicle locations marked at n time points; the vehicle location represents the vehicle's position.
[0133] The transmission module is used to obtain a three-dimensional image of the vehicle and a decrypted vehicle location image based on the vehicle location image through encryption and decryption; the decrypted vehicle location image is the image that has been decrypted after being transmitted to the vehicle network.
[0134] The transmission anomaly detection module is used to determine whether the transmission signal is abnormal based on the vehicle's 3D image and the decrypted vehicle position image.
[0135] The time change detection module is used to determine the data processing time based on the decrypted vehicle position image if the transmission signal is normal, and obtain multiple vehicle images; the vehicle images represent images of the vehicle taken by the corresponding camera on the road after the data processing time has elapsed;
[0136] The time communication anomaly detection module is used to determine whether the time communication signal is abnormal based on the decrypted vehicle location image and multiple vehicle images.
[0137] The algorithms and displays provided herein are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems can also be used in conjunction with the teachings herein. The required structure for constructing such systems is apparent from the above description. Furthermore, this invention is not directed to any particular programming language. It should be understood that the contents of the invention described herein can be implemented using various programming languages, and the above description of specific languages is for the purpose of disclosing the best mode of implementation of the invention.
[0138] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.
[0139] The various component embodiments of the present invention can be implemented in hardware, or as software modules running on one or more processors, or a combination thereof. Those skilled in the art will understand that microprocessors or digital signal processors (DSPs) can be used in practice to implement some or all of the functions of some or all of the components in the apparatus according to embodiments of the present invention. The present invention can also be implemented as a device or apparatus program (e.g., a computer program and computer program product) for performing part or all of the methods described herein. Such programs implementing the present invention can be stored on a computer-readable medium or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.
Claims
1. A method for detecting abnormal vehicle communication signals in a vehicle-to-everything (V2X) environment, characterized in that, include: Acquire vehicle location image; The vehicle location image represents an image that marks the vehicle locations at n time points; The vehicle location refers to the vehicle's position. Based on the vehicle location image, a 3D vehicle image and a decrypted vehicle location image are obtained through encryption and decryption; the decrypted vehicle location image is the image that has been decrypted after being transmitted to the vehicle network. Based on the vehicle's 3D image and the decrypted vehicle location image, determine whether the transmitted signal is abnormal; If the transmission signal is normal, based on the decrypted vehicle location image, the data processing time is determined, and multiple vehicle images are obtained; the vehicle images represent images of the vehicle taken by the corresponding camera on the road after the data processing time has elapsed. Based on the decrypted vehicle location image and multiple vehicle images, it is determined whether the time communication signal is abnormal.
2. The method for detecting abnormal vehicle communication signals in a vehicle-to-everything (V2X) environment according to claim 1, characterized in that, The process of obtaining a 3D vehicle image and a decrypted vehicle position image based on the vehicle position image through encryption and decryption includes: Based on the vehicle position image, different positions are segmented to obtain a 3D vehicle image and an interpolated vehicle image; the 3D vehicle image represents an image where the vehicle's position is continuous at adjacent time points; In the width direction, the 3D image of the vehicle is cut to obtain multiple images at the cutting positions; The image of the cutting location is converted into a cutting location vector; the values in the cutting location vector include vehicle pixel values and vehicle pixel positions; the vehicle pixel values represent the values in the cutting location image; the vehicle pixel positions represent the positions of the vehicle pixel values in the cutting location image; The cutting position vector is encrypted to obtain an encrypted vehicle vector; multiple encrypted vehicle vectors are obtained corresponding to multiple cutting position images; the encrypted vehicle vector is a value that has not been transmitted to the vehicle network. Transmit multiple encrypted vehicle vectors to the vehicle network; Based on multiple encrypted vehicle vectors, decryption is performed to obtain the decrypted vehicle location image.
3. The method for detecting abnormal vehicle communication signals in a vehicle-to-everything (V2X) environment according to claim 1, characterized in that, If the transmission signal is normal, based on the decrypted vehicle location image, the data processing time is determined, and multiple vehicle images are obtained, including: The transmission time length and decryption time point are obtained; the decryption time point is the time point when the vehicle location image is acquired and decrypted; the transmission time represents the length from the time point when the vehicle location image is acquired to the time point when decryption is performed. Based on the decryption time point and the decrypted vehicle location image, the predicted vehicle location is obtained through a location discrimination network; the transmission time length represents the predicted vehicle location at the time point when the decrypted vehicle location image is acquired. Based on the predicted vehicle location, a matching process is performed to obtain a road data table; the road data table stores images taken by camera equipment on the road where the predicted vehicle location is located at multiple time points; The detection period is obtained based on the decryption time point and the transmission time length; the starting point of the detection period is the difference between the decryption time point and the quotient of the transmission time length divided by the segment value; the starting point of the detection period is the sum of the decryption time point and the quotient of the transmission time length divided by the segment value. In the road database, multiple vehicle images for the detection time period were found.
4. The method for detecting abnormal vehicle communication signals in a vehicle-to-everything (V2X) environment according to claim 3, characterized in that, The step of determining whether the time communication signal is abnormal based on the decrypted vehicle location image and multiple vehicle images includes: The vehicle image is input into a target detection network to identify the user's vehicle, obtaining a discrimination value and the identified vehicle location; a discrimination value of 1 indicates the presence of the user's vehicle; a discrimination value of 0 indicates the absence of the user's vehicle. If the discrimination value is 1, the time point corresponding to the vehicle location is taken as the user's time point; the user's time point represents the time point when the user's vehicle was captured in the vehicle image; The determined vehicle location is converted into a geographic location to obtain the user-determined geographic location; the user-determined geographic location represents the geographic location of the vehicle detected in the vehicle image. Based on multiple identical user time points, the user's geographical location is determined and the vehicle's location is predicted to determine whether the time communication signal is abnormal.
5. The method for detecting abnormal vehicle communication signals in a vehicle-to-everything (V2X) environment according to claim 1, characterized in that, The determination of whether the transmitted signal is abnormal based on the vehicle's 3D image and the decrypted vehicle position image includes: With a stride of 1, a 2*2*m two-dimensional convolution kernel is convolved with the vehicle interpolation image to obtain the vehicle value feature vector. With a stride of 1, a 2*2*m two-dimensional convolution kernel is convolved with the decrypted vehicle location image to obtain the decrypted vehicle feature vector. The similarity between the vehicle value feature vector and the decrypted vehicle feature vector is calculated to obtain the feature similarity value; If the feature similarity value is less than the similarity threshold, the transmission signal will be set to normal. If the feature similarity value is greater than or equal to the similarity threshold, the transmission signal is set to abnormal.
6. The method for detecting abnormal vehicle communication signals in a vehicle-to-everything (V2X) environment according to claim 4, characterized in that, The method of determining whether time communication signals are abnormal based on user geographic location identification and vehicle location prediction corresponding to multiple identical user time points includes: The distance between the user's identified geographical location and the predicted vehicle location corresponding to the same user time point is calculated to obtain the distance difference; multiple distance differences are obtained for multiple user time points; Multiple distance differences are clustered, and the cluster centers are used as distance influence deviation values; the distance influence deviation values represent the noise generated by the location discrimination network and the target detection network. If the absolute value of the quotient of the difference between the distance difference and the distance influence deviation value divided by the distance influence deviation value is less than the vehicle distance difference threshold, the time communication signal is set to normal. If the absolute value of the difference between the distance difference and the distance influence deviation value is greater than or equal to the vehicle distance difference threshold, it is set as an abnormal time communication signal.
7. The method for detecting abnormal vehicle communication signals in a vehicle-to-everything (V2X) environment according to claim 2, characterized in that, The step of segmenting different locations based on the vehicle position image to obtain a 3D vehicle image and a vehicle interpolated image includes: Multiple vehicle positions in the vehicle position image are interpolated to obtain an interpolated vehicle image; the interpolated vehicle image includes multiple interpolated positions; the interpolated position represents the predicted position of the vehicle between two adjacent vehicle positions at two time points.
8. The method for detecting abnormal vehicle communication signals in a vehicle-to-everything (V2X) environment according to claim 7, characterized in that, In the vehicle interpolation image, the pixels between the vehicle positions at adjacent time points are retained, while other pixels are deleted to obtain the vehicle segmentation image; n-1 vehicle segmentation images are obtained for n time points. The n-1 vehicle segmentation images are superimposed to form a 3D vehicle image, arranged from morning to night according to time points.
9. The method for detecting abnormal vehicle communication signals in a vehicle-to-everything (V2X) environment according to claim 3, characterized in that, The step of obtaining the predicted vehicle position based on the decryption time point and the decrypted vehicle position image through a position discrimination network includes: In chronological order, multiple vehicle locations in the decrypted vehicle location image are input into the location discrimination network to predict the location at the next time point, thus obtaining the first predicted location and the first predicted time point; In chronological order, the first predicted position and multiple vehicle positions in the decrypted vehicle position image are input into the position discrimination network to predict the position at the next time point, thus obtaining the second predicted position and the second predicted time point. By performing multiple convolutions through a location discrimination network, the predicted vehicle position and the corresponding predicted time point are obtained; the predicted time point is equal to the decryption time point.
10. A vehicle communication anomaly signal detection system in a vehicle-to-everything (V2X) environment, characterized in that, include: The acquisition module is used to acquire vehicle location images; The vehicle location image represents an image that marks the vehicle locations at n time points; The vehicle location refers to the vehicle's position. The transmission module is used to obtain a three-dimensional image of the vehicle and a decrypted vehicle location image based on the vehicle location image through encryption and decryption; the decrypted vehicle location image is the image that has been decrypted after being transmitted to the vehicle network. The transmission anomaly detection module is used to determine whether the transmission signal is abnormal based on the vehicle's 3D image and the decrypted vehicle position image. The time change detection module is used to determine the data processing time based on the decrypted vehicle position image if the transmission signal is normal, and obtain multiple vehicle images; the vehicle images represent images of the vehicle taken by the corresponding camera on the road after the data processing time has elapsed; The time communication anomaly detection module is used to determine whether the time communication signal is abnormal based on the decrypted vehicle location image and multiple vehicle images.