A video transmission management system and method based on cloud-edge collaboration
The cloud-edge collaboration system addresses real-time tracking and dynamic adjustment issues in edge computing by predicting storage space release times and optimizing edge device selection, enhancing video transmission efficiency and resource utilization.
Patent Information
- Application Number
- CN202411287814.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-14
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-09-14
AI Technical Summary
Existing edge computing video transmission systems in public transportation face challenges with real-time tracking and dynamic adjustment, low storage resource utilization, and inefficient resource scheduling, leading to low transmission efficiency and system performance.
A video transmission management system based on cloud-edge collaboration that includes real-time location tracking of terminal devices, dynamic identification of transmission lines, time series prediction for storage space release behavior, and intelligent scheduling to optimize resource utilization and transmission efficiency.
Enhances transmission efficiency and resource utilization by predicting storage space release times and dynamically selecting optimal edge devices for video transmission, improving real-time performance and reducing delays and resource waste.
Smart Images

Figure CN118803210B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of video surveillance technology, and specifically to a video transmission management system and method based on cloud-edge collaboration. Background Art
[0002] In recent years, with the development of intelligent transportation systems, the real-time transmission and management of public transportation surveillance videos have become an important part of urban traffic management. Traditional video transmission systems mainly rely on centralized data centers. This architecture has problems such as bandwidth bottlenecks, high latency, and high data packet loss when processing large amounts of video data. With the rise of edge computing technology, edge computing-based video transmission management systems have begun to be used in public transportation monitoring. Edge computing reduces the burden on central servers, reduces latency, and improves data processing efficiency by processing data at the edge of the network. However, existing edge computing video transmission systems still face some challenges in practical applications, especially when dealing with dynamic and complex urban traffic environments.
[0003] Although the edge computing video transmission system in the existing technology has solved some of the delay and bandwidth bottleneck problems, it still has deficiencies in real-time, reliability and scalability. For example, in the public transportation system, edge devices need to process a large amount of real-time video data and dynamically adjust according to the real-time location of public transportation. However, the existing systems usually lack effective real-time tracking and dynamic adjustment mechanisms, resulting in low transmission efficiency. At the same time, the existing systems lack intelligent prediction and optimization methods when dealing with the storage space release behavior of edge devices, resulting in low storage resource utilization. In addition, when multiple edge devices work together, traditional systems often lack effective scheduling algorithms and cannot fully utilize the resources of each edge device, affecting the overall system performance.
[0004] The existing video transmission management system based on cloud-edge collaboration has the problem of insufficient real-time tracking and dynamic adjustment of edge devices. The inventor provides a video transmission management method based on cloud-edge collaboration, which mainly solves these problems. This method significantly improves the efficiency and reliability of video file transmission by acquiring and tracking the location information of public transportation terminal devices in real time, dynamically identifying transmission lines and generating optimal transmission solutions. In addition, by using a time series prediction algorithm to model and predict the storage space release behavior of edge devices, the utilization of storage resources is optimized. Summary of the invention
[0005] The purpose of the present invention is to provide a video transmission management system and method based on cloud-edge collaboration to solve the problems raised in the above background technology.
[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a video transmission management method based on cloud-edge collaboration, comprising: dividing the public transportation road network into regions based on edge devices, and generating a planning area set for the edge devices based on the regional division results; obtaining the location information of all public transportation terminal devices in the planning area, and tracking the location information of all public transportation terminal devices in the planning area in real time; based on the location information of the public transportation terminal devices, dynamically identifying the transmission lines of the public transportation terminal devices in the transmission area where the current edge device is located, and generating a transmission plan according to the priority of the transmission line; based on the transmission plan, the terminal device transmits the video file to the edge device.
[0007] As a preferred solution of the video transmission management method based on cloud-edge collaboration described in the present invention, the regional division of the public transportation network based on edge devices includes: setting edge device fixed points based on the urban transportation route network, the edge device is used to receive video files uploaded by a video storage device, the video storage device is installed on public transportation, and at least one video acquisition device is also installed on the public transportation, and the video storage device is used to store video files of the video acquisition device installed on the public transportation, wherein one video storage device is installed on one public transportation.
[0008] Establish a cloud-edge archive library, which stores edge devices and video storage devices, and uniformly encode edge devices and video storage devices respectively, wherein any i-th edge device is recorded as , denote any a-th video storage device as .
[0009] Based on the urban traffic route network, we traverse all the bus routes in the city and record any k-th bus route as , and coordinate bus routes All public transportation on the route, generating a route traffic set , where A represents the total number of video storage devices;
[0010] Based on the effective signal receiving range of the edge device, the edge transmission area is formed, which is denoted as , when public transportation is on a bus route During the driving process, enter the edge device Edge transport zone When the video storage device corresponding to the public transportation is uploaded Video files collected in the network to edge devices middle.
[0011] As a preferred solution of the video transmission management method based on cloud-edge collaboration described in the present invention, wherein: based on the location information of the public transportation terminal device, the public transportation terminal device transmission line in the transmission area where the current edge device is located is dynamically identified, and according to the priority of the transmission line, a transmission plan is generated, including: calling the edge device The storage space release behavior refers to when the edge device The video files stored in the edge device fill up storage space, edge devices Upload edge devices to the cloud Video files stored in the edge device The time period during which the storage space release behavior occurs is recorded as ,in, represents the tth time period.
[0012] Based on the route traffic set and the edge transmission area , get all edge devices , and collect edge devices Select a time series prediction algorithm to model and train the historical release time data to predict the edge device The time period of all release behaviors in the next day; use the trained model to analyze the occurrence of each edge device Predict release times within the next day.
[0013] As a preferred solution of the video transmission management method based on cloud-edge collaboration described in the present invention, wherein: Forecasts for release times within the next day include:
[0014] From edge devices Obtain historical release time data, perform data cleaning, and integrate the processed data into a time series format to form a time series data set .
[0015] Select the time series prediction model Prophet, and organize the time series dataset Input the time series prediction model, perform fitting, and obtain the fitting results ; Adjust model parameters to optimize prediction results; divide data into training set and test set, use training set to train model, and use test set to verify model results and calculate error:
[0016] , where N is the number of data, is the actual value, is the predicted value;
[0017] Constructs a datetime index covering a time range one day into the future ,in, Ensure the continuity of the time series for future time points; input the future time index into the model to form a future data frame ; Use the trained Prophet model for future data frames Make predictions and generate prediction result data frame , and from the prediction result data frame Extract the column as the prediction result of release time, denoted as ;
[0018] The predicted results The time point above the set threshold θ is marked as the release time point: , where Z is the set of release time points, is the jth release time point, is the jth predicted value, and m is the number of predicted time points; the continuous release time points are merged into time periods to obtain the final predicted release time period list ;
[0019] According to the video storage device Real-time location of edge devices Whether it is within the predicted release time period.
[0020] As a preferred solution of the video transmission management method based on cloud-edge collaboration described in the present invention, the step of determining whether the edge device is in the predicted release time period includes:
[0021] Get video storage via GPS Real-time latitude and longitude coordinates; traverse all edge devices The location information and calculate the video storage device Current location and each edge device distance; select distance video storage device Recent Edge Devices As the current corresponding edge device; obtain the current time (for example, through the system time), and traverse the predicted release time period list; for each predicted release time period:
[0022] Checks whether the current time is within the start and end time of the time period.
[0023] If it is within the time period, it is determined that the edge device is within the predicted release time period.
[0024] If the device is not included in all time periods, the edge device is judged Not within the predicted release timeframe.
[0025] The result is an edge device. If the video storage device is not within the predicted release time period, The stored video files are transmitted to the edge device.
[0026] If the edge device If the data is in the predicted release time period, data transmission is temporarily suspended until the next inspection cycle.
[0027] According to the video storage device Current location, speed and route information, estimated video storage device Reaching edge devices The estimated arrival time is used to replace the current time, and the above judgment process is repeated.
[0028] This method can determine the video storage device in advance. Edge devices upon arrival Whether it is in a released state, thereby improving the responsiveness of the system.
[0029] As a preferred solution of the video transmission management method based on cloud-edge collaboration described in the present invention, wherein: based on the location information of the public transportation terminal device, the public transportation terminal device transmission line in the transmission area where the current edge device is located is dynamically identified, and the transmission plan is generated according to the priority of the transmission line, and further includes: from all edge devices , filter out the distance video storage device The first n edge devices that are less than or equal to the preset distance L , denoted as candidate edge devices Set M; for candidate edge devices Edge devices in set M Sort by distance from small to large; from the sorted candidate edge devices In the set, select the first m edge devices closest to , where m <n。
[0030] Check the first m edge devices The time period of release behavior If there is a period of time when the release behavior does not occur Edge devices within , then select the nearest edge device Transfer video files;
[0031] If the first m edge devices All in the time period of release behavior Within, the entire candidate edge device The set M is dynamically judged, and the specific judgment method is as follows:
[0032] Computational video storage With each edge device Dynamic distance changes between
[0033] The dynamic distance change formula is:
[0034] ;
[0035] in, Video storage device The x coordinate in three-dimensional space; Video storage device The y coordinate in three-dimensional space; Video storage device The z coordinate in three-dimensional space; For edge devices At current time The x-coordinate of the moment in three-dimensional space; For edge devices At current time The y coordinate of the moment in three-dimensional space; For edge devices At current time The z coordinate in three-dimensional space at the moment;
[0036] Consider edge devices Remaining release time ;
[0037] Design formula to choose the right edge device Acceptance is performed with the priority being: minimum dynamic distance change and minimum remaining release time;
[0038] The formula is: ;
[0039] Select Priority Rating Smallest edge device :
[0040] ;
[0041] Among them, α is the distance weight coefficient; Video storage device With edge devices At current time The dynamic distance change rate; β is the time weight coefficient; For edge devices The remaining release time of , used to identify candidate edge devices Each edge device in the set M ; For edge devices Priority score;
[0042] Stream video files to selected edge devices , and update edge devices Memory status and release time.
[0043] As a preferred solution of the video transmission management method based on cloud-edge collaboration described in the present invention, wherein: based on the transmission scheme, the terminal device transmits the video file to the edge device, including the following steps: obtaining the location information of all public transportation terminal devices in the current transmission area, and tracking the location information of the public transportation terminal devices in real time, the transmission node updates the terminal node information of the current transmission node in real time, and each terminal node corresponds to only one transmission node in the same time period, and each transmission node corresponds to n terminal nodes in the same time period; the terminal node transmits the video file to the transmission node corresponding to the current terminal node based on the transmission scheme provided by the above steps.
[0044] The present invention also provides a video transmission management system based on cloud-edge collaboration, including a regional division and planning module, which is used to divide the public transportation network into regions based on edge devices and generate a set of planned regions; a location information acquisition and tracking module, which is used to acquire and track the location information of public transportation terminal devices in real time; a dynamic transmission plan generation module, which is used to identify transmission lines and generate transmission plans based on the location information of terminal devices; an edge device storage space management module, which is used to manage the storage space release behavior of edge devices to ensure that video files can be uploaded to the cloud in time; a time series prediction module, which is used to predict the storage space release time of edge devices based on a time series prediction algorithm; a transmission node selection module, which is used to select a suitable edge device for video file transmission to avoid transmission within the release time period; an edge device dynamic evaluation module, which is used to dynamically evaluate the transmission applicability of edge devices, comprehensively considering dynamic distance changes and remaining release time; a real-time location and transmission node update module, which is used to update the terminal device information in the current transmission area in real time to ensure that each terminal node corresponds to a unique transmission node; a video file transmission module, which is used for the terminal node to transmit video files to the corresponding transmission node based on the transmission plan.
[0045] Compared with the prior art, the beneficial effects achieved by the present invention are: by retrieving the storage space release behavior of edge devices and predicting future release time, the video transmission management based on cloud-edge collaboration is optimized. The Prophet prediction algorithm is used to model the historical release time data, and the optimal edge device is dynamically selected for video transmission, which effectively improves the transmission efficiency and resource utilization of the system. Through real-time monitoring and intelligent scheduling, it ensures that video data is transmitted within the optimal time period, avoiding transmission delays and resource waste. These innovative steps effectively improve the real-time, reliability and overall performance of the system. BRIEF DESCRIPTION OF THE DRAWINGS
[0046] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:
[0047] Figure 1 A flow chart of a video transmission management method based on cloud-edge collaboration according to an embodiment of the present invention. DETAILED DESCRIPTION
[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0049] Example 1: Please refer to Figure 1 The present invention provides a video transmission management method based on cloud-edge collaboration, comprising:
[0050] S100: Divide the public transportation network into regions based on edge devices, and generate a set of planned regions for the edge devices based on the region division results, where the planned regions include unique regions and overlapping regions.
[0051] Furthermore, based on the urban traffic route network, edge device fixed points are set up, and the edge device is used to receive video files uploaded by a video storage device. The video storage device is installed on public transportation, and at least one video acquisition device is also installed on the public transportation. Moreover, the video storage device is used to store video files of the video acquisition device installed on the public transportation, wherein one video storage device is installed on one public transportation.
[0052] Establish a cloud-edge archive library, which stores edge devices and video storage devices, and uniformly encode edge devices and video storage devices respectively, wherein any i-th edge device is recorded as , denote any a-th video storage device as .
[0053] S101: Based on the urban transportation route network, traverse all bus routes in the city and record any k-th bus route as , and coordinate bus routes All public transportation on the route, generating a route traffic set , where A represents the total number of video storage devices.
[0054] Furthermore, based on the effective signal receiving range of the edge device, an edge transmission area is formed, which is denoted as , when public transportation is on a bus route During the driving process, enter the edge device Edge transport zone When the video storage device corresponding to the public transportation is uploaded Video files collected in the network to edge devices middle.
[0055] S200: Acquire the location information of all public transportation terminal devices in the planning area, and track the location information of all public transportation terminal devices in the planning area in real time.
[0056] Specifically, by using positioning technologies such as GPS to obtain location information, the location information such as the latitude and longitude coordinates of each public transportation terminal device is collected; by establishing a real-time data transmission and processing mechanism, the location information of each terminal device is continuously updated and recorded so as to grasp the dynamic location of the device at any time.
[0057] S300: Based on the location information of the public transportation terminal equipment, dynamically identify the transmission lines of the public transportation terminal equipment in the transmission area where the current edge device is located, and generate a transmission plan according to the priority of the transmission line.
[0058] S301: Retrieve edge devices The storage space release behavior refers to when the edge device The video files stored in the edge device fill up storage space, edge devices Upload edge devices to the cloud Video files stored in the edge device The time period during which the storage space release behavior occurs is recorded as ,in, represents the tth time period.
[0059] Based on the route traffic set and the edge transmission area , get all edge devices , and collect edge devices Select a time series prediction algorithm to model and train the historical release time data to predict the edge device The time period during which all releases will occur in the future day.
[0060] S302: Use the trained model to train each edge device Predict release times within the next day.
[0061] Specifically, from edge devices Obtain historical release time data, perform data cleaning, and integrate the processed data into a time series format to form a time series data set .
[0062] Select the time series prediction model Prophet, and organize the time series dataset Input the time series prediction model, perform fitting, and obtain the fitting results ; Adjust model parameters to optimize prediction results; divide data into training set and test set, use training set to train model, and use test set to verify model results and calculate error: , where N is the number of data, is the actual value, is the predicted value; according to the verification results, the model parameters are tuned to further improve the prediction accuracy of the model.
[0063] Furthermore, a datetime index is constructed that contains a time range of one day in the future. ,in, Ensure the continuity of the time series for future time points; input the future time index into the model to form a future data frame ; Use the trained Prophet model for future data frames Make predictions and generate prediction result data frame , and from the prediction result data frame Extract the column as the prediction result of release time, denoted as .
[0064] The predicted results The time point above the set threshold θ is marked as the release time point: , where Z is the set of release time points, is the jth release time point, is the jth predicted value, and m is the number of predicted time points; the continuous release time points are merged into time periods to obtain the final predicted release time period list .
[0065] According to the video storage device Real-time location of edge devices Whether it is within the predicted release time period.
[0066] Specifically, the video storage device is obtained through GPS Real-time latitude and longitude coordinates; traverse all edge devices The location information and calculate the video storage device Current location and each edge device distance; select distance video storage device Recent Edge Devices As the current corresponding edge device; obtain the current time (for example, through the system time), and traverse the predicted release time period list;
[0067] For each predicted release time period, check whether the current time is within the start and end time of the time period; if it is within the time period, determine whether the edge device If the device is not within the predicted release time period, then the edge device is judged to be Not within the predicted release timeframe.
[0068] The judgment result is an edge device If the video storage device is not within the predicted release time period, The stored video file is transmitted to the edge device; if the edge device If the data is in the predicted release time period, data transmission is temporarily suspended until the next inspection cycle.
[0069] According to the video storage device Current location, speed and route information, estimated video storage device Reaching edge devices The estimated arrival time is used to replace the current time, and the above judgment process is repeated.
[0070] It should be noted that the above content can determine in advance whether the video storage device Edge devices upon arrival Whether it is in a released state, thereby improving the responsiveness of the system.
[0071] S303: From all edge devices , filter out the distance video storage device The first n edge devices that are less than or equal to the preset distance L , denoted as candidate edge devices Set M; for candidate edge devices Edge devices in set M Sort by distance from small to large; from the sorted candidate edge devices In the set, select the first m edge devices closest to , where m <n。
[0072] Check the first m edge devices The time period of release behavior If there is a period of time when the release behavior does not occur Edge devices within , then select the nearest edge device Transmit video files, if the first m edge devices All in the time period of release behavior Within, the entire candidate edge device The set M is dynamically judged, and the specific judgment method is as follows:
[0073] Computational video storage With each edge device Dynamic distance changes between
[0074] The dynamic distance change formula is:
[0075] ;
[0076] in, Video storage device The x coordinate in three-dimensional space; Video storage device The y coordinate in three-dimensional space; Video storage device The z coordinate in three-dimensional space; For edge devices At current time The x-coordinate of the moment in three-dimensional space; For edge devices At current time The y coordinate of the moment in three-dimensional space; For edge devices At current time The z-coordinate of the moment in three-dimensional space.
[0077] Consider edge devices Remaining release time .
[0078] Design formula to choose the right edge device Acceptance is performed with priority being: minimum dynamic distance change and minimum remaining release time.
[0079] The formula is: ;
[0080] Select Priority Rating Smallest edge device :
[0081] ;
[0082] Among them, α is the distance weight coefficient; Video storage device With edge devices At current time The dynamic distance change rate; β is the time weight coefficient; For edge devices The remaining release time of , used to identify candidate edge devices Each edge device in the set M ; For edge devices Priority score.
[0083] Stream video files to selected edge devices , and update edge devices Memory status and release time.
[0084] S304: Record the actual release time, optimize the prediction model, and dynamically adjust it by combining GIS and traffic data.
[0085] Specifically, record each edge device ED i The actual release time is compared with the predicted results to evaluate the accuracy of the prediction model, and the prediction model is retrained regularly with new historical data to improve the prediction accuracy; the geographic information system (GIS) is introduced to optimize the spatial relationship calculation of routes and edge devices, and the expected arrival time of the bus is dynamically adjusted considering real-time traffic data.
[0086] S400: Based on the transmission solution, the terminal device transmits the video file to the edge device.
[0087] The location information of all public transportation terminal devices in the current transmission area is obtained, and the location information of the public transportation terminal devices is tracked in real time. The transmission node updates the terminal node information of the current transmission node in real time, and each terminal node corresponds to only one transmission node in the same time period, and each transmission node corresponds to n terminal nodes in the same time period; the terminal node transmits the video file to the transmission node corresponding to the current terminal node based on the transmission scheme provided in step S300.
[0088] Embodiment 2: This embodiment also provides a video transmission management system based on cloud-edge collaboration that implements the method of embodiment 1, including: a regional division and planning module, which is used to divide the public transportation road network into regions based on edge devices and generate a set of planned regions; a location information acquisition and tracking module, which is used to acquire and track the location information of public transportation terminal devices in real time; a dynamic transmission plan generation module, which is used to identify transmission lines and generate transmission plans based on the location information of terminal devices; an edge device storage space management module, which is used to manage the storage space release behavior of edge devices to ensure that video files can be uploaded to the cloud in a timely manner; a time series prediction module, which is used to predict the storage space release time of edge devices based on a time series prediction algorithm; a transmission node selection module, which is used to select a suitable edge device for video file transmission to avoid transmission during the release time period; an edge device dynamic evaluation module, which is used to dynamically evaluate the transmission applicability of edge devices, comprehensively considering dynamic distance changes and remaining release time; a real-time location and transmission node update module, which is used to update the terminal device information in the current transmission area in real time to ensure that each terminal node corresponds to a unique transmission node; a video file transmission module, which is used for the terminal node to transmit video files to the corresponding transmission node based on the transmission plan.
[0089] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article 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, article or device.
[0090] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.
Claims
1. A video transmission management method based on cloud-edge collaboration, characterized in that: The method comprises the following steps: Step S100: dividing the public transportation network into regions based on edge devices, and generating a planning region set for the edge devices based on the region division result; Step S200: acquiring the location information of all public transportation terminal devices in the planning area, and tracking the location information of all public transportation terminal devices in the planning area in real time; Step S300: Based on the location information of the public transportation terminal equipment, dynamically identify the public transportation terminal equipment transmission line in the transmission area where the current edge device is located, and generate a transmission plan according to the priority of the transmission line; Step S400: Based on the transmission scheme, the terminal device transmits the video file to the edge device; The step S100 includes: Based on the urban transportation route network, an edge device fixed point is set, the edge device is used to receive the video file uploaded by the video storage device, the video storage device is installed on the public transportation, and at least one video acquisition device is also installed on the public transportation, and the video storage device is used to store the video files of the video acquisition device installed on the public transportation, wherein one video storage device is installed on one public transportation; Establish a cloud-edge archive library, which stores edge devices and video storage devices, and uniformly encode edge devices and video storage devices respectively, wherein any i-th edge device is recorded as , denote any a-th video storage device as ; The step S100 further includes: Based on the urban traffic route network, we traverse all the bus routes in the city and record any k-th bus route as , and coordinate bus routes All public transportation on the route, generating a route traffic set , where A represents the total number of video storage devices; Based on the effective signal receiving range of the edge device, the edge transmission area is formed, which is denoted as , when public transportation is on a bus route During the driving process, enter the edge device Edge transport zone When the video storage device corresponding to the public transportation is uploaded Video files collected in the network to edge devices middle; The step S300 includes: Call the edge device The storage space release behavior refers to when the edge device The video files stored in the edge device fill up storage space, edge devices Upload edge devices to the cloud Video files stored in the edge device The time period during which the storage space release behavior occurs is recorded as ,in, represents the tth time period; Based on the route traffic set and the edge transmission area , get all edge devices , and collect edge devices Select a time series prediction algorithm to model and train the historical release time data to predict the edge device The time period of all release behaviors in the next day; use the trained model to analyze the occurrence of each edge device Predict the release time within the next day; For each edge device Forecasts for release times within the next day include: From edge devices Obtain historical release time data, perform data cleaning, and integrate the processed data into a time series format to form a time series data set ; Select the time series prediction model Prophet, and organize the time series dataset Input the time series prediction model, perform fitting, and obtain the fitting results ; Divide the data into training set and test set, use the training set to train the model, use the test set to verify the model effect, and calculate the error MSE: Where N is the number of data, is the actual value, is the predicted value; Constructs a datetime index covering a time range one day into the future ,in, is a future time point; the future time index is input into the model to form a future data frame ; Use the trained time series forecasting model to predict future data frames Make predictions and generate prediction result data frame , and from the prediction result data frame Extract the column as the prediction result of release time, denoted as ; The predicted results The time point above the set threshold θ is marked as the release time point: , where Z is the set of release time points, is the jth release time point, is the jth predicted value, is the number of predicted time points; the continuous release time points are combined into time periods to obtain the final predicted release time period list ; According to the video storage device Real-time location of edge devices Whether it is within the predicted release time period; Determining whether the edge device is within the predicted release time period includes: Computational video storage Current location and each edge device distance; select distance video storage device Recent Edge Devices As the current corresponding edge device; get the current time, and traverse the list of predicted release time periods; for each predicted release time period: Check whether the current time is within the start and end time of the time period; If it is within the time period, then the edge device is judged is within the predicted release time period; If the device is not included in all time periods, the edge device is determined to be Not within the predicted release time period and the video storage device The stored video files are transmitted to the edge device ; If the edge device If it is within the predicted release time period, data transmission will not be performed temporarily, and the next inspection cycle will be waited; The step S300 further includes: From all edge devices , filter out the distance video storage device The first n edge devices that are less than or equal to the preset distance L , denoted as candidate edge devices Set M; for candidate edge devices Edge devices in set M Sort by distance from small to large; from the sorted candidate edge devices In the set, select the first m edge devices closest to ; Check the first m edge devices The time period of release behavior If there is a period of time when the release behavior does not occur Edge devices within , then select the nearest edge device Transfer video files; If the first m edge devices All in the time period of release behavior Within, the entire candidate edge device The set M is dynamically judged, and the specific judgment method is as follows: Computational video storage With each edge device Dynamic distance changes between The dynamic distance change formula is: in, Video storage device The x coordinate in three-dimensional space; Video storage device The y coordinate in three-dimensional space; Video storage device The z coordinate in three-dimensional space; For edge devices At current time The x-coordinate of the moment in three-dimensional space; For edge devices At current time The y coordinate of the moment in three-dimensional space; For edge devices At current time The z coordinate in three-dimensional space at the moment; Consider edge devices Remaining release time ; Design formula to choose the right edge device Acceptance is performed with the priority being: minimum dynamic distance change and minimum remaining release time; The formula is: Select Priority Rating Smallest edge device : Among them, α is the distance weight coefficient; Video storage device With edge devices At current time The dynamic distance change rate; β is the time weight coefficient; For edge devices The remaining release time; i is the i-th edge device , used to identify candidate edge devices Each edge device in the set M ; For edge devices Priority score of Stream video files to selected edge devices , and update edge devices Memory status and release time; The step S400 includes: obtaining the location information of all public transportation terminal devices in the current transmission area, and tracking the location information of the public transportation terminal devices in real time, the transmission node updates the terminal node information of the current transmission node in real time, and each terminal node corresponds to only one transmission node in the same time period, and each transmission node corresponds to n terminal nodes in the same time period; the terminal node transmits the video file to the transmission node corresponding to the current terminal node based on the transmission scheme provided in step S300.
2. A video transmission management system based on cloud-edge collaboration, executing a video transmission management method based on cloud-edge collaboration as claimed in claim 1, characterized in that: The system includes: The regional division and planning module is used to divide the public transportation network into regions based on edge devices and generate a set of planned regions; Location information acquisition and tracking module, used to acquire and track the location information of public transportation terminal equipment in real time; A dynamic transmission scheme generation module, used to identify the transmission line and generate a transmission scheme based on the location information of the terminal device; The edge device storage space management module is used to manage the storage space release behavior of edge devices to ensure that video files can be uploaded to the cloud in a timely manner; The time series prediction module is used to predict the storage space release time of edge devices based on the time series prediction algorithm; The transmission node selection module is used to select edge devices for video file transmission to avoid transmission during the release time period; The edge device dynamic assessment module is used to dynamically assess the transmission suitability of edge devices, taking into account dynamic distance changes and remaining release time; Real-time location and transmission node update module, used to update the terminal device information in the current transmission area in real time to ensure that each terminal node corresponds to a unique transmission node; The video file transmission module is used for the terminal node to transmit the video file to the corresponding transmission node based on the transmission scheme.
