Networking method and device in moving process of target equipment

Abstract

The invention relates to a networking method and device in the moving process of target equipment. The method comprises the steps that the equipment signal state, multiple pieces of position data and network data matched with each piece of position data of the target equipment in the moving process are collected in real time; processing the equipment signal state, the plurality of pieces of position data and network data matched with each piece of position data through a pre-trained target generation model, and obtaining optimal network data matched with the next position data when predicting that the next position of the target equipment in the moving route has network abnormity; and transmitting the optimal network data to the target device, so that the target device adopts the optimal network data to perform networking. The time consumption of network recovery can be shortened.

Description

Technical Field

[0001] The present application relates to the field of neural network technology, and in particular to a method and apparatus for networking a target device during movement. Background Art

[0002] With the increasing popularity of smart wearable devices and smart cars, these smart devices that are highly dependent on cellular networks have placed extremely high demands on network stability in mobile scenarios. Since they are not used in a fixed location, users often use these devices in changing mobile scenarios. This feature causes the devices to frequently switch from one base station to another during movement. However, this frequent change between base stations often causes network equipment to drop out of the network, unstable signals, and other mobility phenomena, which directly affects the user experience.

[0003] Traditional solutions are mostly limited by the established solution framework in the 3GPP (3rd Generation Partnership Project) agreement. For example, re-searching or cell reselection after network disconnection may take a long time for network recovery. Summary of the invention

[0004] The present application provides a method and apparatus for networking a target device during movement, so as to solve the problem that network recovery takes a long time.

[0005] In a first aspect, the present application provides a method for connecting a target device to a network during movement, the method comprising:

[0006] Real-time collection of device signal status, multiple location data, and network data matching each location data of the target device during movement;

[0007] The device signal state, the multiple location data, and the network data matched by each location data are processed by a pre-trained target generation model, and when it is predicted that a network anomaly will occur at the next location of the target device in the moving route, the optimal network data matched by the next location data is obtained;

[0008] The optimal network data is transmitted to the target device, so that the target device uses the optimal network data to connect to the network.

[0009] Optionally, when it is predicted that a network anomaly will occur at the next position of the target device in the moving route, obtaining optimal network data matching the next position data includes:

[0010] generating a target route according to a plurality of location data and network data matching each of the location data;

[0011] If it is predicted that the next position in the target route has a network anomaly according to the current device signal state, then it is determined whether the target route belongs to a preset set of learned routes, wherein the set of learned routes includes multiple complete routes with network anomalies learned by the target generation model;

[0012] If the target route belongs to a preset set of learned routes, a set route matching the target route is selected from the set of learned routes;

[0013] The next location data in the set route is determined according to the current location data of the target device, and the optimal network data matching the next location data is determined, wherein the set route includes the optimal network data matching the next location data of each abnormal location.

[0014] Optionally, after determining whether the target route belongs to a preset set of learned routes, the method further includes:

[0015] If the target route does not belong to the preset set of learned routes, the network data of the next location data is determined by cell reselection or network loss re-search.

[0016] Optionally, after transmitting the optimal network data to the target device, the method further includes:

[0017] Obtaining a device signal status result after the target device is connected to the network using the optimal network data, wherein the device signal status result is used to indicate whether to shorten the network recovery time of the target device;

[0018] After the target device moves to a next location, determining a plurality of updated location data and network data matching each of the location data;

[0019] Model self-learning is performed according to the device signal status result, the updated multiple location data, and the network data matching each location data.

[0020] Optionally, the pre-training process of the target generation model includes:

[0021] When the sample device moves repeatedly on the set route for multiple times, obtaining a behavior log of the sample device at each movement, wherein the behavior log includes a signal state of the sample device, a plurality of sample location data, and network data matched by each of the sample location data;

[0022] If it is detected that a network anomaly occurs on the set route according to the signal states of the plurality of sample devices, and the number of times the network anomaly occurs exceeds a preset threshold, then determining the abnormal location data in the set route;

[0023] Determine the candidate network data that matches the next location data of the abnormal location data, wherein the candidate network data that matches the next location data is not completely the same each time the sample device moves;

[0024] The sample device signal state corresponding to each candidate network data is determined, and the sample network data corresponding to the best device signal state is selected as the optimal network data for matching the next position data.

[0025] Optionally, determining abnormal location data in the set route includes:

[0026] Determine the route position of the sample device in the set route each time a network anomaly occurs, and record the number of occurrences of each route position;

[0027] If the number of occurrences of the route location is greater than a preset number, the data of the route location is used as abnormal location data.

[0028] Optionally, predicting that there is a network anomaly on the target route according to the current device signal state includes:

[0029] If it is detected that the target device has at least one of the following behaviors: cell reselection, network loss and re-search, signal strength less than a strength threshold, or device power consumption greater than a power consumption threshold, it is predicted that there is a network anomaly on the target route.

[0030] In a second aspect, the present application provides a network connection device during the movement of a target device, the device comprising:

[0031] A collection module, used for collecting in real time the device signal status of the target device during movement, multiple location data, and network data matching each of the location data;

[0032] A processing module, used to process the device signal state, the multiple location data and the network data matched by each location data through a pre-trained target generation model, and obtain the optimal network data matched by the next location data when it is predicted that a network anomaly will occur at the next location of the target device in the moving route;

[0033] The transmission module is used to transmit the optimal network data to the target device so that the target device adopts the optimal network data to connect to the network.

[0034] In a third aspect, the present application provides an electronic device comprising: at least one communication interface; at least one bus connected to the at least one communication interface; at least one processor connected to the at least one bus; and at least one memory connected to the at least one bus.

[0035] In a fourth aspect, the present application further provides a computer storage medium storing computer executable instructions, wherein the computer executable instructions are used to execute the networking method during the movement of the target device as described in any one of the above items of the present application.

[0036] The above-mentioned technical scheme provided by the embodiment of the present application has the following advantages compared with the prior art: when the target generation model predicts that the target device will have a network anomaly at the next location based on the collected data, the optimal network data at the next location is transmitted to the target device in advance, and the target device directly adopts the optimal network data to connect to the network, avoiding unnecessary steps such as cell reselection (such as signal monitoring, information collection, cell reselection condition judgment, and calculation of cell priority) and re-searching the network after disconnection (such as detecting and confirming service interruption, initializing the reconnection process, and searching for available networks), thereby shortening the time taken for network recovery. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0038] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0039] One or more embodiments are exemplarily described by pictures in the corresponding drawings, and these exemplified descriptions do not constitute limitations on the embodiments. Elements with the same reference numerals in the drawings represent similar elements, and unless otherwise stated, the figures in the drawings do not constitute proportional limitations.

[0040] Figure 1 A schematic diagram of a system for networking a target device during movement provided by an embodiment of the present application;

[0041] Figure 2 A flow chart of a method for networking a target device during movement provided by an embodiment of the present application;

[0042] Figure 3 A flow chart comparing the optimized solution provided in the embodiment of the present application with the traditional cell reselection solution;

[0043] Figure 4 A flow chart comparing the optimized solution provided in the embodiment of the present application with the traditional network loss re-search solution;

[0044] Figure 5A schematic diagram of the overall process of using the target generation model provided in the embodiment of the present application;

[0045] Figure 6 A schematic diagram of the structure of a networking device during the movement of a target device provided in an embodiment of the present application;

[0046] Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0048] The disclosure below provides many different embodiments or examples to implement different structures of the present invention. In order to simplify the disclosure of the present invention, the parts and settings of specific examples are described below. Of course, they are only examples, and the purpose is not to limit the present invention. In addition, the present invention can repeat reference numbers and / or letters in different examples. This repetition is for the purpose of simplification and clarity, and does not itself indicate the relationship between the various embodiments and / or settings discussed.

[0049] In order to solve the problem mentioned in the background technology, according to one aspect of an embodiment of the present application, an embodiment of a method for networking a target device during movement is provided.

[0050] Optionally, in the embodiment of the present application, the above-mentioned method for networking during the movement of the target device can be applied to the following: Figure 1 In the hardware environment composed of the device 101 and the server 103 shown in FIG. Figure 1 As shown, the server 103 is connected to the device 101 via a network and can be used to provide services for the device. A database 105 can be set on the server or independently of the server to provide data storage services for the server 103. The above network includes but is not limited to: a wide area network, a metropolitan area network or a local area network.

[0051] The device provided in the embodiment of the present application may specifically be a module capable of realizing a communication function or a terminal device including the module, etc. The device relies on a cellular network during movement. The terminal device is, for example, a mobile terminal, a smart wearable device, a smart car or a shared bicycle. The mobile terminal may specifically be at least one of a mobile phone, a tablet computer, a laptop computer, etc. The smart wearable device may specifically be a smart watch. The module may specifically be a wireless communication module, such as any one of a 2G communication module, a 3G communication module, a 4G communication module, a 5G communication module, and a NB-IOT communication module.

[0052] The present application provides a method for connecting a target device to a network during its movement, which can be applied to a server or a target device to shorten the time required for network recovery.

[0053] The present application embodiment is explained in detail by taking application to a server as an example. Figure 2 As shown, the method comprises the following steps:

[0054] Step 201: collecting in real time the device signal status, multiple location data, and network data matching each location data of the target device during movement;

[0055] Step 202: Process the device signal state, multiple location data, and network data matching each location data through a pre-trained target generation model, and obtain the optimal network data matching the next location data when it is predicted that a network anomaly will occur at the next location of the target device in the moving route;

[0056] Step 203: Transmit the optimal network data to the target device, so that the target device uses the optimal network data to connect to the network.

[0057] When the target device moves along a route, the server collects the device signal status, multiple location data and network data matching each location data in the route in real time, and then inputs the above data into a pre-trained target generation model. The target generation model determines the current location of the target device and predicts that the target device will have network anomalies at the next location. The target generation model then processes the above data to obtain the optimal network data at the next location, and transmits the optimal network data to the target device's modem bottom layer. The target device directly uses the optimal network data to connect to the Internet.

[0058] Optionally, the mobile route includes multiple location data, each location data has corresponding network data, and the network data is at least one of a public land mobile network (PLMN), a frequency point or a physical cell identifier (PCI). PLMN is a network established and operated by a mobile network operator in a mobile communication system. It includes components such as a core network, a wireless access network, and user terminals. PLMN provides the public with a wide range of mobile communication services, such as voice calls, text messages, and data transmission. Frequency points refer to the specific frequencies used by wireless signals. Different frequency points correspond to different channels, which are used to distinguish and manage multiple wireless communication channels to avoid interference and conflicts. By selecting appropriate frequency points, the quality and efficiency of wireless communication can be improved. PCI is used to uniquely identify a cell in an LTE (Long-Term Evolution Network) network. Each cell has a unique PCI value, which is used to distinguish and manage different cells. PCI plays an important role in the LTE network, helping user equipment identify and connect to the correct cell to achieve efficient wireless communication.

[0059] The device signal status includes network status, serving cell information, signal strength, device power consumption, etc. The following is a detailed explanation of each device signal status.

[0060] The network status refers to the connection status between the device and the communication network, which usually includes the device's access status (such as connected, not connected), network type (such as 4G, 5G), IP address allocation, and data transmission rate. The network status reflects whether the device can successfully connect to the network and can stably transmit data.

[0061] Serving cell information refers to the detailed information of the wireless cell to which the device is currently connected. In mobile communications, a cell is usually covered by a base station or a sector of a base station. Serving cell information includes the cell identifier (such as PCI), frequency, etc. This information is crucial for device mobility and handover management, because it helps the device to smoothly switch between different cells and maintain communication continuity.

[0062] Signal strength refers to the strength of the wireless signal received by the device. If there is a network anomaly, the signal strength of the target device will be greatly attenuated.

[0063] Device power consumption refers to the amount of electricity consumed by a device during operation. If a network anomaly occurs, the power consumption of the target device will increase significantly.

[0064] In the present application, when the target generation model predicts that the target device will experience a network anomaly at the next location based on the collected data, the optimal network data at the next location is transmitted to the target device in advance. The target device directly uses the optimal network data to connect to the network, avoiding unnecessary steps such as cell reselection (such as signal monitoring, information collection, cell reselection condition judgment, and calculation of cell priority) and re-searching the network after disconnection (such as detecting and confirming service interruption, initializing the reconnection process, and searching for available networks), thereby shortening the time required for network recovery.

[0065] The embodiments of the present application omit some unnecessary steps in terms of cell reselection and network re-search, as follows.

[0066] The traditional 3GPP steps for cell reselection are: 1. Signal monitoring: The mobile station continuously monitors the signal strength and system messages of the current serving cell and its neighboring cells. 2. Information collection: Record and update the reception level, system messages and control messages of each neighboring cell. 3. Condition judgment: Determine whether cell reselection is required based on the preset trigger conditions and evaluation criteria. 4. Algorithm calculation: If the reselection conditions are met, the mobile station will use the cell reselection algorithm to calculate the priority of each candidate cell. 5. Decision execution: Based on the algorithm calculation results, the neighboring cell with the highest priority is selected as the new serving cell. Executing cell reselection includes updating internal status, sending necessary signaling messages, etc. 6. Status maintenance: The mobile station resides in the new serving cell and continues to monitor signal quality and system messages so that it can reselect the cell again when necessary.

[0067] For cell reselection, the embodiment of the present application replaces the key steps in the traditional steps (1. signal monitoring, 2. information collection, 3. condition judgment, 4. algorithm calculation) with an AI model, and instructs the execution module to pass the optimal network information obtained by the route learning module directly to the modem bottom layer, so that the modem can select the optimal cell in advance, and directly select the cell based on the cell information such as PCI and frequency recommended by the AI ​​model, eliminating the steps including signal monitoring, information collection, condition judgment, and algorithm calculation, and avoiding the repetition of reselection actions.

[0068] Figure 3 FIG. 1 is a flow chart comparing the optimized solution with the traditional cell reselection solution. It can be seen that the four steps in the traditional solution are replaced by the control module in the embodiment of the present application to reduce the duration of network disconnection.

[0069] The traditional 3GPP steps for reconnecting after a network outage are: 1. Detect and confirm service interruption: The modem will continuously monitor its connection status with the network. Once a service interruption is detected, the modem will identify that it is currently in the OOS (Out Of Service) state and prepare to start the reconnection process. 2. Initialize the reconnection process: Once the service interruption is confirmed, the modem will automatically start the reconnection process and try to reconnect to the network. 3. Search for available networks: The modem will search for available networks based on a preset network priority list (such as a PLMN priority list). Scanning the frequency band and selecting a network may take some time, depending on the hardware performance of the modem and the current network environment. 4. Initiate a network registration request: After finding an available network, the modem will send a registration request to the network. 5. Restore network service: Once the registration is successful, the modem will regain network services, including data transmission, voice calls and other functions. At this time, the modem will exit the OOS state and resume normal use. 6. Status maintenance: The mobile station resides in the new service cell and continues to monitor the network status.

[0070] Regarding network re-search after network disconnection, the embodiment of the present application replaces the key steps in the traditional steps (1. Detect and confirm service interruption, 2. Initialize the reconnection process, 3. Search for available networks) with an AI model, instructing the execution module to use the optimal network cell as the registration object, and directly search and register with the PLMN, PCI, frequency, etc. recommended by the AI ​​model, eliminating the steps of detecting and confirming service interruption, initializing the reconnection process, and searching for available networks, thereby avoiding repeated reselection actions.

[0071] Figure 4 The figure is a flow chart comparing the optimized solution with the traditional network loss re-search solution. It can be seen that the three steps in the traditional solution are replaced by the control module in the embodiment of the present application to improve the rapid recovery time after the device is disconnected from the network, minimize the user's perception of the device disconnecting from the network, and improve the user experience.

[0072] Optionally, obtaining optimal network data matching the next location data includes the following steps.

[0073] Step S11: generating a target route according to a plurality of location data and network data matching each location data;

[0074] Step S12: if it is predicted that the next position in the target route has a network anomaly according to the current device signal state, it is determined whether the target route belongs to a preset set of learned routes, wherein the set of learned routes includes multiple complete routes with network anomalies learned by the target generation model;

[0075] Step S13: if the target route belongs to a preset set of learned routes, a set route matching the target route is selected from the set of learned routes;

[0076] Step S14: Determine the next location data in the set route according to the current location data of the target device, and determine the optimal network data matching the next location data, wherein the set route includes the optimal network data matching the next location data of each abnormal location.

[0077] The target generation model generates a target route based on multiple location data and the network data that matches each location data. These location data represent key locations that the target device may pass through during movement, and the matching network data includes key parameters such as the public land mobile network, frequency point, and physical cell identifier. The target route generated by the target generation model is not necessarily a complete route. For example, a user needs to pass through five intersections from home to work. When the user reaches the second intersection, a target route can be generated. This target route is only a part of the complete route.

[0078] The target generation model analyzes the current device signal status. If it detects that the target device has at least one of the following behaviors: cell reselection, network loss and re-search, signal strength less than the strength threshold, or device power consumption greater than the power consumption threshold, it predicts that there will be a network anomaly at the next location on the target route.

[0079] By learning and analyzing a large amount of data, the target generation model can identify multiple routes with network anomalies and save these routes into a learned route set. This set not only contains the routes themselves, but also records the specific locations where network anomalies may occur on each route and the corresponding optimal network data.

[0080] When the target device moves along a target route, the target generation model will determine in real time whether the current target route belongs to the set of learned routes. If not, it means that the target route is a new route, and the target generation model cannot predict the data of the next location and the matching network data. In this case, the target device can only rely on traditional cell reselection or network re-search to determine the network data of the next location.

[0081] However, if the current target route belongs to the set of learned routes, even though the route may not be complete, the target generation model can still select a complete set route that matches the target route from the set. This set route not only includes all the locations that the target device may pass through, but also specifically marks the optimal network data that matches the next location of each abnormal location.

[0082] For example, suppose the route is set from home to company. This route contains five intersections, of which the third and fourth intersections are prone to network anomalies. When the target device is at the second intersection, the target generation model can accurately predict that the third intersection will have a network anomaly and transmit the optimal network data of the third intersection to the target device in advance. In this way, when the device reaches the third intersection, it no longer needs to go through the tedious process of re-searching for the network or service cell after the network is disconnected, but can directly connect to the network to ensure the continuity and stability of communication.

[0083] This dynamic network optimization method based on the target generation model and a set of learned routes can predict network anomalies at a certain location and transmit the optimal network data for the next location to the device in advance. The device does not need to go through the process of re-searching for the network or service cell after being disconnected from the network, and can directly connect to the network, thus shortening the duration of network anomalies and improving user experience.

[0084] Specifically, the target generation model includes five modules: device information recording module, business monitoring module, route learning module, control module and execution module. The specific functions of each module are as follows.

[0085] Device information recording module: deployed in the target device, responsible for collecting the location data of the target device when it moves on the route (the location coordinates of the device during movement) and the network data of each location data (PLMN, frequency, PCI).

[0086] Service monitoring module: real-time monitoring of terminal device status (such as network status, service cell information, signal strength, device power consumption, etc.).

[0087] Route learning module: includes three functions. The first function is to build a network route through the network information and location data collected by the device information recording module; the second function is to record the status of the device on the route (such as network status, service cell information, signal strength, device power consumption, etc.) according to the business monitoring module. This function is used to determine which position on which route the device moves to where the device status will be abnormal, and send the result of the network abnormality to the control module.

[0088] Control module: determines whether the current route belongs to the route that has been learned by the route learning module. If so, when there is a potential network abnormality risk on the route, it issues an instruction to the execution module and passes the relevant network data to the execution module.

[0089] Execution module: Receives instructions from the control module and executes corresponding measures. Before the device is about to reach an area that may trigger cell reselection or network disconnection, the optimal network information including PLMN, frequency, PCI, etc. is passed to the device's modem bottom layer. This operation is intended to enable the modem to set this optimal network information as the priority for network selection or cell residence, thereby avoiding unnecessary steps such as network measurement and re-search after network disconnection in advance, ensuring the continuity and stability of the device network.

[0090] As an optional implementation manner, after transmitting the optimal network data to the target device, the method further includes the following steps:

[0091] Step S21: obtaining a device signal status result after the target device is connected to the network using the optimal network data, wherein the device signal status result is used to indicate whether to shorten the network recovery time of the target device;

[0092] Step S22: after the target device moves to the next location, determining the updated plurality of location data and the network data matching each location data;

[0093] Step S23: Perform model self-learning according to the device signal status result, the updated multiple location data, and the network data matching each location data.

[0094] After the target device moves to the next location and connects to the network, it obtains the device signal status result. This signal status result is a key indicator that reflects the network connection quality of the target device at that location. If the signal status result shows that the network recovery time has been shortened, it means that the current network connection strategy is effective and can be continued or fine-tuned. On the contrary, if the signal status result is not ideal, the network connection strategy needs to be adjusted and optimized.

[0095] Next, after the target device moves to the next location, the updated multiple location data and the network data matching each location data are determined. The updated location data is an additional location data, and the other original location data remain unchanged. They will serve as input data for model self-learning.

[0096] Finally, the model self-learns based on the device signal status results, multiple updated location data, and the network data that matches each location data. In this process, the target generation model uses machine learning algorithms to analyze and process the input data to identify which factors have a significant impact on the network recovery time. By continuously adjusting and optimizing model parameters, the target generation model can gradually improve the accuracy and reliability of predictions, thereby providing a more accurate and efficient network connection strategy for the target device.

[0097] In the prior art, when a network device moves on a fixed line, it is possible to encounter a network anomaly at a fixed location each time it moves on the route, and then completely perform a cell reselection or network re-search for the anomaly. The present application uses a continuous self-learning mechanism of the model to improve itself against abnormal network phenomena, thereby continuously optimizing and improving the network connection strategy of the target device, improving the quality and efficiency of wireless communications, and enhancing the user experience.

[0098] Figure 5 The overall process diagram for generating the model for the target includes the following execution steps.

[0099] 1. Device registers the network. As the device moves, it may enter a new network coverage area or leave the original service cell. In this case, the device needs to register the network multiple times in order to establish a connection with the new network and maintain communication.

[0100] 2. The device information recording module is responsible for collecting network data (PLMN, frequency, PCI, etc.) and location data (location coordinates during device movement).

[0101] 3. The service monitoring module monitors the terminal device status in real time (such as network status, service cell information, signal strength, device power consumption, etc.).

[0102] 4. The route learning module constructs a network route based on the information collected by the device information recording module and the business monitoring module, detects whether there is a potential network anomaly risk at a certain location, and sends the network anomaly results to the control module.

[0103] 5. If the control module determines that there is no network anomaly, it returns to step 1. If it determines that there is a network anomaly, it determines whether the current route belongs to the route learned by the route learning module. If so, it sends an instruction to the execution module and passes the network data of the next location to the execution module.

[0104] 6. Before the device is about to reach an area that may trigger cell reselection or network disconnection, the execution module transmits the optimal network information including PLMN, frequency, PCI, etc. to the device's modem bottom layer.

[0105] 7. After the execution module is executed, the device information recording module records the latest location data and network data, the business monitoring module monitors the device signal status results in real time, and feeds the results back to the route learning module in step 4 and the control module in step 5 to train the machine learning algorithm and improve the intelligent model.

[0106] This application uses AI algorithms to monitor the operating routes of equipment and the quality of network services in real time and make intelligent decisions. It can help users select the optimal network data in advance on abnormal network routes, effectively reduce the number of cell reselections, shorten the time for re-searching after network loss, and effectively reduce users' perception of network anomalies, thereby improving the user experience.

[0107] In addition, in the traditional solution, the solution to the deterioration of the serving cell signal is to trigger the cell reselection mechanism, which involves steps such as terminal measurement, network strategy and terminal decision-making. Frequent measurement and reselection is not only time-consuming but also seriously increases the power consumption of the device. The solution to network connection interruption is to trigger the network loss re-search mechanism. The entire network loss re-search process involves detecting network loss, triggering re-search, searching for available networks, attempting to connect, and processing successful or failed connections. Each network loss re-search must execute the complete steps. This mechanism also has the defects of a long execution process and increased device power consumption. The present application can also reduce device power consumption by shortening the time it takes to restore the network.

[0108] As an optional implementation, the pre-training process of the target generation model includes the following steps.

[0109] Step S31: when the sample device repeatedly moves on the set route for multiple times, a behavior log of the sample device at each movement is obtained, wherein the behavior log includes a signal state of the sample device, multiple sample location data, and network data matching each sample location data;

[0110] Step S32: if it is detected that a network anomaly occurs on the set route according to the signal states of the multiple sample devices, and the number of times the network anomaly occurs exceeds a preset threshold, then the abnormal location data in the set route is determined;

[0111] Step S33: determining the candidate network data that matches the next location data of the abnormal location data, wherein the candidate network data that matches the next location data of the sample device is not completely the same each time it moves;

[0112] Step S34: Determine the sample device signal state corresponding to each candidate network data, and select the sample network data corresponding to the best device signal state as the optimal network data for matching the next location data.

[0113] In order to ensure the accuracy and reliability of the target generation model, it needs to be fully pre-trained. This process usually involves using sample devices to perform multiple repeated movements on a set route and collecting behavior logs generated during each movement. These logs contain rich information, such as the signal status of the sample device, multiple sample location data, and network data matched by each location data, providing a solid foundation for subsequent analysis and training.

[0114] It is worth noting that each time the sample device moves, the same location may use different network data. This diversity can record the device signal status under different network data at the location, making it easier to select the optimal network data for the same location later.

[0115] The server will obtain the signal status of multiple sample devices collected when the sample devices move multiple times. By analyzing the signal status of these devices, if abnormal conditions such as multiple cell reselections, multiple network drops and re-searches, multiple reductions in device signals, or multiple increases in device power consumption are detected, it can be inferred that multiple network anomalies have occurred on the set line. In order to further determine the location of the anomaly, the server will query the location of the sample device each time a network anomaly occurs, and use the location that appears multiple times as the abnormal location.

[0116] To illustrate with a specific example, suppose the sample device traveled on the set route 100 times, and network anomalies occurred 80 times. Of these 80 times, 77 network anomalies occurred at the third intersection, and the remaining three network anomalies occurred at the first intersection, the second intersection, and the fourth intersection. Through analysis, it can be considered that these three times may be accidental phenomena, and the third intersection is determined to be the abnormal location.

[0117] Once the abnormal location on the set route is determined, the optimal network data needs to be selected for the next location. Since the same location may match multiple candidate network data, the server will query the signal status of the sample device at that location each time the device moves. By comparing the device signal status under different network data, the candidate network data with the best sample device signal status can be used as the optimal network data.

[0118] During the pre-training process of the target generation model, the repeated movements of sample devices on the set route are recorded and analyzed to find out the locations where network anomalies may occur and determine the optimal network data for the next location.

[0119] Among them, the setting location of the target generation model is flexible. It can be integrated inside the target device or deployed on a remote server, such as the cloud. This design enables the target generation model to adapt to different usage scenarios and needs.

[0120] When the target generation model is set in the target device, the device itself becomes a combination of the sample device and the target device. In this case, the target generation model is trained based on the routes that the device has repeatedly traveled. After training, the target generation model can be directly applied to the target device to provide services for the device itself. This self-learning and self-optimization capability greatly improves the stability and communication efficiency of the target device in a complex network environment.

[0121] If the target generation model is set on the server, the sample device and the target device can be different devices. In this case, the target generation model can be trained based on the network route data of multiple sample devices to obtain a more comprehensive and generalized network optimization model. This model can not only provide services for a single target device, but also provide similar services for other devices, such as tourist routes, high-speed rail routes, etc. By deploying the target generation model on a cloud server, resources can be shared and optimized, and multiple devices can use this model together to improve the quality and stability of their network connections. At the same time, the powerful computing power and storage capacity of the cloud server also provide strong support for the training and application of the target generation model.

[0122] Based on the same technical concept, the present application provides a network connection device during the movement of a target device, such as Figure 6 As shown, the device comprises:

[0123] The acquisition module 601 is used to collect the device signal status, multiple location data and network data matching each location data of the target device in real time during the movement process;

[0124] The processing module 602 is used to process the device signal state, multiple location data, and network data matched by each location data through a pre-trained target generation model, and obtain the optimal network data matched by the next location data when it is predicted that a network anomaly will occur at the next location of the target device in the moving route;

[0125] The transmission module 603 is used to transmit the optimal network data to the target device so that the target device uses the optimal network data to connect to the network.

[0126] Optionally, the processing module 602 is used to:

[0127] generating a target route according to the plurality of location data and the network data matching each location data;

[0128] If it is predicted that the next position in the target route has a network anomaly based on the current device signal state, then it is determined whether the target route belongs to a preset set of learned routes, where the set of learned routes includes multiple complete routes with network anomalies learned by the target generation model;

[0129] If the target route belongs to a preset set of learned routes, a set route matching the target route is selected from the set of learned routes;

[0130] The next location data in the set route is determined according to the current location data of the target device, and the optimal network data matching the next location data is determined, wherein the set route includes the optimal network data matching the next location data of each abnormal location.

[0131] Optionally, the device is also used for:

[0132] If the target route does not belong to the preset set of learned routes, the network data of the next location data is determined by cell reselection or network loss re-search.

[0133] Optionally, the device is also used for:

[0134] Obtaining a device signal status result after the target device is connected to the network using the optimal network data, wherein the device signal status result is used to indicate whether to shorten the network recovery time of the target device;

[0135] After the target device moves to a next location, determining a plurality of updated location data and network data matching each location data;

[0136] The model is self-learned based on the device signal status results, the updated multiple location data, and the network data matching each location data.

[0137] Optionally, the device is also used for:

[0138] When the sample device moves repeatedly on the set route for multiple times, a behavior log of the sample device at each movement is obtained, wherein the behavior log includes a signal state of the sample device, a plurality of sample location data, and network data matching each sample location data;

[0139] If a network anomaly is detected on a set route based on the signal status of multiple sample devices, and the number of network anomalies exceeds a preset threshold, abnormal location data in the set route is determined;

[0140] Determine the candidate network data that matches the next location data of the abnormal location data, wherein the candidate network data that matches the next location data of the sample device is not completely the same each time it moves;

[0141] Determine the sample device signal state corresponding to each candidate network data, and select the sample network data corresponding to the best device signal state as the optimal network data for matching the next location data.

[0142] Optionally, the device is also used for:

[0143] Determine the route position of the sample device in the set route each time a network anomaly occurs, and record the number of occurrences of each route position;

[0144] If the number of occurrences of the route location is greater than a preset number, the data of the route location is regarded as abnormal location data.

[0145] Optionally, the device is also used for:

[0146] If it is detected that the target device has at least one of the following behaviors: cell reselection, network loss and re-search, signal strength less than a strength threshold, or device power consumption greater than a power consumption threshold, it is predicted that there is a network anomaly on the target route.

[0147] like Figure 7 As shown, an embodiment of the present application provides an electronic device, including a processor 701, a communication interface 702, a memory 703 and a communication bus 704, wherein the processor 701, the communication interface 702, and the memory 703 communicate with each other through the communication bus 704.

[0148] The memory 703 is used to store computer programs.

[0149] In one embodiment of the present application, the processor 701 is used to implement the networking method during the movement of the target device provided by any one of the aforementioned method embodiments when executing the program stored in the memory 703.

[0150] An embodiment of the present application further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the method for networking a target device during movement provided by any of the aforementioned method embodiments are implemented.

[0151] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0152] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a general hardware platform, and of course, by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the relevant technology can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0153] It should be understood that the terms used in the text are only for the purpose of describing specific example embodiments, and are not intended to be limiting. Unless the context clearly indicates otherwise, the singular forms "one", "an" and "said" as used in the text may also be meant to include plural forms. The terms "include", "comprise", "contain", and "have" are inclusive, and therefore specify the existence of stated features, steps, operations, elements and / or parts, but do not exclude the existence or addition of one or more other features, steps, operations, elements, parts, and / or combinations thereof. The method steps, processes, and operations described herein are not interpreted as necessarily requiring them to be performed in the specific order described or illustrated, unless the execution order is clearly indicated. It should also be understood that additional or alternative steps may be used.

[0154] The foregoing is merely a specific embodiment of the present invention, which enables those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A method for networking a target device during movement, characterized in that: The method comprises: Real-time collection of device signal status, multiple location data, and network data matching each location data of the target device during movement; The device signal state, the multiple location data, and the network data matched by each location data are processed by a pre-trained target generation model, and when it is predicted that a network anomaly will occur at the next location of the target device in the moving route, the optimal network data matched by the next location data is obtained; The optimal network data is transmitted to the target device, so that the target device uses the optimal network data to connect to the network.

2. The method according to claim 1, characterized in that When it is predicted that a network anomaly will occur at the next location of the target device in the moving route, obtaining optimal network data matching the next location data includes: generating a target route according to a plurality of location data and network data matching each of the location data; If it is predicted that the next position in the target route has a network anomaly according to the current device signal state, then it is determined whether the target route belongs to a preset set of learned routes, wherein the set of learned routes includes multiple complete routes with network anomalies learned by the target generation model; If the target route belongs to a preset set of learned routes, a set route matching the target route is selected from the set of learned routes; The next location data in the set route is determined according to the current location data of the target device, and the optimal network data matching the next location data is determined, wherein the set route includes the optimal network data matching the next location data of each abnormal location.

3. The method according to claim 2, characterized in that After determining whether the target route belongs to a preset set of learned routes, the method further includes: If the target route does not belong to the preset set of learned routes, the network data of the next location data is determined by cell reselection or network loss re-search.

4. The method according to claim 1, characterized in that: After transmitting the optimal network data to the target device, the method further includes: Obtaining a device signal status result after the target device is connected to the network using the optimal network data, wherein the device signal status result is used to indicate whether to shorten the network recovery time of the target device; After the target device moves to a next location, determining a plurality of updated location data and network data matching each of the location data; Model self-learning is performed according to the device signal status result, the updated multiple location data, and the network data matching each location data.

5. The method according to claim 1, characterized in that The pre-training process of the target generation model includes: When the sample device moves repeatedly on the set route for multiple times, obtaining a behavior log of the sample device at each movement, wherein the behavior log includes a signal state of the sample device, a plurality of sample location data, and network data matched by each of the sample location data; If it is detected that a network anomaly occurs on the set route according to the signal states of the plurality of sample devices, and the number of times the network anomaly occurs exceeds a preset threshold, then determining the abnormal location data in the set route; Determine the candidate network data that matches the next location data of the abnormal location data, wherein the candidate network data that matches the next location data is not completely the same each time the sample device moves; The sample device signal state corresponding to each candidate network data is determined, and the sample network data corresponding to the best device signal state is selected as the optimal network data for matching the next position data.

6. The method according to claim 5, characterized in that Determining abnormal location data in the set route includes: Determine the route position of the sample device in the set route each time a network anomaly occurs, and record the number of occurrences of each route position; If the number of occurrences of the route location is greater than a preset number, the data of the route location is used as abnormal location data.

7. The method according to claim 2, characterized in that According to the current device signal status, network anomalies are predicted on the target route, including: If it is detected that the target device has at least one of the following behaviors: cell reselection, network loss and re-search, signal strength less than a strength threshold, or device power consumption greater than a power consumption threshold, it is predicted that there is a network anomaly on the target route.

8. A network connection device for a target device during movement, characterized in that: The device comprises: A collection module, used for collecting in real time the device signal status of the target device during movement, multiple location data, and network data matching each of the location data; A processing module, used to process the device signal state, the multiple location data and the network data matched by each location data through a pre-trained target generation model, and obtain the optimal network data matched by the next location data when it is predicted that a network anomaly will occur at the next location of the target device in the moving route; The transmission module is used to transmit the optimal network data to the target device so that the target device adopts the optimal network data to connect to the network.

9. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; Memory, used to store computer programs; A processor, for implementing any of the methods described in claims 1-7 when executing a program stored in a memory.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images