Vehicle-mounted network switching method, vehicle-mounted network switching device and vehicle-mounted network switching system

Abstract

The invention relates to a vehicle-mounted network switching method, a vehicle-mounted network switching device and a system, and the method comprises the steps: obtaining multi-dimensional original data reported to a cloud end by a vehicle end, and obtaining multi-dimensional feature data based on the multi-dimensional original data; performing at least one of network coverage prediction, user demand prediction and network performance prediction on the multi-dimensional feature data to obtain a prediction result; according to the prediction result and the current network use state, determining an optimal network switching strategy, and determining a network switching opportunity; generating a network switching instruction according to the optimal network switching strategy and the network switching opportunity; and issuing a network switching instruction to the vehicle end. According to the invention, the operation task of vehicle-mounted network switching is transferred to the cloud for execution, so that the computing resources of the cloud can be more fully utilized, the computing burden of vehicle-side hardware and software is reduced, and the switching efficiency is improved at the same time.

Description

Technical Field

[0001] The present application relates to the field of vehicles, and in particular to a vehicle network switching method, a vehicle network switching device, and a vehicle network switching system. Background Art

[0002] As a crucial component of vehicle intelligence, in-vehicle networks must achieve efficient and stable communications even in high-speed mobile scenarios to meet user needs for a variety of applications, including navigation, entertainment, remote vehicle diagnostics, and autonomous driving. Currently, in related technologies, the generation of in-vehicle network switching solutions is implemented on the vehicle side. Specifically, implementing in-vehicle network switching on the vehicle side relies on the vehicle's hardware and software, which performs network switching based on threshold judgments based on network signal-related parameters. Having the vehicle side calculate the network switching strategy increases the hardware cost and energy consumption of the vehicle-side equipment, thereby increasing the computational burden on the vehicle-side hardware and software, and resulting in low switching efficiency.

[0003] There is currently no effective solution to the problems of high computational burden and low switching efficiency of vehicle-side software and hardware in related technologies. Summary of the Invention

[0004] In this embodiment, a vehicle network switching method, a vehicle network switching device and a system are provided to solve the problems of high computational burden and low switching efficiency of vehicle-side software and hardware in related technologies.

[0005] In a first aspect, this embodiment provides a vehicle network switching method for use in the cloud, the method comprising:

[0006] Acquire multidimensional raw data reported by the vehicle end to the cloud, and obtain multidimensional feature data based on the multidimensional raw data;

[0007] Performing at least one of network coverage prediction, user demand prediction, and network performance prediction on the multi-dimensional feature data to obtain a prediction result;

[0008] Determining an optimal network switching strategy and a network switching timing based on the prediction results and the current network usage status;

[0009] generating a network switching instruction according to the optimal network switching strategy and the network switching timing;

[0010] Send the network switching instruction to the vehicle end.

[0011] In some embodiments, obtaining multidimensional raw data reported by the vehicle to the cloud and obtaining multidimensional feature data based on the multidimensional raw data include:

[0012] Integrating the multidimensional raw data according to a preset time sequence and / or spatial sequence to obtain integrated data;

[0013] Feature extraction is performed on the integrated data to obtain the multi-dimensional feature data.

[0014] In some embodiments, the multi-dimensional raw data includes at least one of the following: vehicle status data, network status data, navigation data, user behavior data, and environmental data;

[0015] The multi-dimensional feature data includes at least one of the following: a vehicle status change trend, a network status change trend, a driving trajectory change trend, a user behavior pattern change trend, and an environment change trend.

[0016] In some embodiments, performing at least one of network coverage prediction, user demand prediction, and network performance prediction on the multi-dimensional feature data to obtain a prediction result includes:

[0017] According to the environmental change trend, combined with historical network coverage data and driving trajectory change trends, network coverage is predicted to obtain a network coverage prediction result;

[0018] According to the changing trend of user behavior patterns, predict the user behavior in the preset time period and obtain the user demand prediction results;

[0019] The network performance of the preset time period is predicted based on the network status change trend to obtain a network performance prediction result.

[0020] In some embodiments, determining an optimal network switching strategy and determining a network switching timing based on the prediction result and the current network usage status includes:

[0021] Determining an optimal network switching strategy based on the trained strategy calculation model according to the prediction results and the current network usage status;

[0022] Determine the network switching timing based on the vehicle's real-time driving information.

[0023] In some embodiments, generating a network switching instruction according to the optimal network switching strategy and the network switching timing includes:

[0024] Determining a target network to be switched and switching parameters according to the optimal network switching strategy;

[0025] The target network, the switching parameters, and the network switching opportunity are assembled into a network switching instruction.

[0026] In some embodiments, obtaining the multi-dimensional raw data reported by the vehicle to the cloud includes:

[0027] Based on a publish-subscribe mode message transmission protocol, the multi-dimensional original data reported by the vehicle side is received.

[0028] In some embodiments, sending the network switching instruction to the vehicle terminal includes:

[0029] Determine whether the communication status of the data link between the cloud and the vehicle is normal; if so, issue the network switching instruction to the vehicle based on the data link;

[0030] Otherwise, the network switching instruction is sent to the vehicle end via text signaling.

[0031] In a second aspect, an in-vehicle network switching device is provided in this embodiment for use in the cloud. The device includes: an acquisition module, a prediction module, a strategy determination module, an instruction generation module, and an instruction delivery module.

[0032] The acquisition module is used to acquire the multi-dimensional original data reported by the vehicle end to the cloud, and obtain multi-dimensional feature data based on the multi-dimensional original data;

[0033] The prediction module is configured to perform at least one of network coverage prediction, user demand prediction, and network performance prediction on the multi-dimensional feature data to obtain a prediction result;

[0034] The strategy determination module is used to determine the optimal network switching strategy and the network switching timing based on the prediction result and the current network usage status;

[0035] The instruction generation module is used to generate a network switching instruction according to the optimal network switching strategy and the network switching timing;

[0036] The instruction issuing module is used to issue the network switching instruction to the vehicle end.

[0037] In a third aspect, a vehicle network switching system is provided in this embodiment, including: a vehicle side and a cloud side;

[0038] The vehicle side is communicatively connected with the cloud side;

[0039] The vehicle side is used to collect multi-dimensional raw data of the vehicle and report the collected multi-dimensional raw data to the cloud; the vehicle side is also used to receive and execute the network switching instruction issued by the cloud;

[0040] The cloud is used to execute the vehicle network switching method described in the first aspect above.

[0041] Compared to related technologies, this embodiment provides an in-vehicle network switching method, an in-vehicle network switching device, and a system. The in-vehicle network switching method obtains multidimensional raw data reported from the vehicle to the cloud, and obtains multidimensional feature data based on the multidimensional raw data; performs at least one of network coverage prediction, user demand prediction, and network performance prediction on the multidimensional feature data to obtain a prediction result; determines the optimal network switching strategy and the timing of network switching based on the prediction result and the current network usage status; generates a network switching instruction based on the optimal network switching strategy and the timing of network switching; and issues the network switching instruction to the vehicle. By moving the computational tasks of in-vehicle network switching to the cloud for execution, it can more fully utilize the cloud's computing resources, reduce the computational burden on the vehicle's hardware and software, and improve switching efficiency.

[0042] The details of one or more embodiments of the present application are set forth in the following drawings and description to make other features, objects, and advantages of the present application more readily apparent. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0044] Figure 1 This is a hardware structure block diagram of a terminal of the vehicle network switching method according to an embodiment of the present application;

[0045] Figure 2 is a flow chart of the vehicle network switching method according to an embodiment of the present application;

[0046] Figure 3 is a flowchart of a vehicle network switching method according to some embodiments of the present application;

[0047] Figure 4 This is a flow chart of a vehicle-side data collection and reporting method according to some embodiments of the present application;

[0048] Figure 5 This is a flow chart of a method for processing multi-dimensional raw data in the cloud according to some embodiments of the present application;

[0049] Figure 6 This is a flow chart of a method for issuing a network switching instruction on the cloud side according to some embodiments of the present application;

[0050] Figure 7 is a flowchart of a method for executing network switching on a vehicle side according to some embodiments of the present application;

[0051] Figure 8 This is a structural block diagram of a vehicle network switching device for the cloud according to an embodiment of the present application;

[0052] Figure 9 It is a structural diagram of the vehicle network switching system of an embodiment of the present application. DETAILED DESCRIPTION

[0053] In order to more clearly understand the purpose, technical solutions and advantages of the present application, the present application is described and illustrated below in conjunction with the accompanying drawings and embodiments.

[0054] Unless otherwise defined, technical or scientific terms used in this application shall have the ordinary meanings as understood by persons of ordinary skill in the art to which this application belongs. The terms "a," "an," "the," "these," and similar expressions in this application do not denote limitations on quantity and may be singular or plural. The terms "comprise," "include," "have," and any variations thereof, as used in this application, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or device comprising a series of steps or modules (units) is not limited to the listed steps or modules (units) but may include unlisted steps or modules (units) or other steps or modules (units) inherent to the process, method, product, or device. The terms "connected," "connected," "coupled," and similar expressions used in this application are not limited to physical or mechanical connections but may include electrical connections, whether direct or indirect. As used in this application, "plurality" means two or more. "And / or" describes an association between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can mean: A exists alone; A and B exist simultaneously; or B exists alone. Generally, the character " / " indicates that the objects in the preceding and following relationship are in an "or" relationship. The terms "first", "second", "third", etc. involved in this application are only used to distinguish similar objects and do not represent a specific ordering of the objects.

[0055] The method embodiment provided in this embodiment can be executed in a terminal, a computer or a similar computing device. For example, running on a terminal, Figure 1 FIG. 1 is a block diagram of the hardware structure of the terminal of the vehicle network switching method of this embodiment. Figure 1 As shown, the terminal may include one or more ( Figure 1 The processor 102 (only one is shown) and a memory 104 for storing data, wherein the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA. The terminal may also include a transmission device 106 for communication functions and an input / output device 108. It will be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above terminal. Figure 1 More or fewer components than shown, or with Figure 1 Different configurations shown.

[0056] The memory 104 can be used to store computer programs, such as software programs and modules of application software, such as the computer program corresponding to the vehicle network switching method in this embodiment. The processor 102 executes the computer program stored in the memory 104 to execute various functional applications and data processing, thereby implementing the above-mentioned method. The memory 104 may include high-speed random access memory and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories may be connected to the terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0057] Transmission device 106 is used to receive or transmit data via a network. This network may include a wireless network provided by the terminal's communications provider. In one embodiment, transmission device 106 includes a network interface controller (NIC), which can be connected to other network devices via a base station to enable communication with the Internet. In another embodiment, transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0058] In this embodiment, a vehicle network switching method is provided. Figure 2 is a flow chart of the vehicle network switching method of this embodiment. Figure 2 As shown, the process includes the following steps:

[0059] In step S210 , the multi-dimensional original data reported by the vehicle to the cloud is obtained, and multi-dimensional feature data is obtained based on the multi-dimensional original data.

[0060] Different networks (such as 4G, 5G, and satellite) have varying coverage. For example, cellular signals may be interrupted in some areas, but satellite communications can maintain service. Therefore, in-vehicle networks need to adapt as the vehicle moves to different locations, as driving time changes, as the weather changes, or as the network needs of vehicle users change. This ensures efficient and stable communication in mobile scenarios, meeting the needs of applications such as navigation, entertainment, remote vehicle diagnostics, and autonomous driving. 4G, 5G, and satellite communications can be used as in-vehicle networks.

[0061] In this embodiment, the task of calculating in-vehicle network switching is transferred from the vehicle to the cloud (Telematics Service Provider, TSP). The vehicle only needs to collect multidimensional raw data and report it to the cloud, enabling the cloud to perform in-vehicle network switching calculations based on the multidimensional raw data. The multidimensional raw data can be data that affects the in-vehicle network and can include driving status data such as the vehicle's location, speed, and driving direction; network status data such as signal strength, bandwidth, and latency; and external environmental data such as weather conditions, traffic conditions, and geographic location that can affect signals. The dimensions of the multidimensional raw data can be multidimensional data representing driving status data, multidimensional data representing network status, or multidimensional data representing weather conditions; or a combination of one or more driving status data, one or more network status data, and other types of data. Those skilled in the art can analyze the factors to be considered for network switching based on actual application scenarios and vehicle performance, and then determine the dimensions that comprise the multidimensional raw data based on these factors. Therefore, this embodiment does not specifically limit the specific type of multidimensional raw data.

[0062] The vehicle can report collected multidimensional raw data to the cloud at preset intervals, using a pre-set communication transmission protocol. For example, multidimensional raw data can be reported to the cloud every 100 milliseconds (ms), enabling the cloud to adjust network switching strategies in a timely manner, improving the efficiency and stability of network switching. Alternatively, the vehicle can trigger data reporting when it detects a data change.

[0063] After receiving the multidimensional raw data reported by the vehicle, the cloud will store it on disk. The cloud can also perform further data preprocessing and feature extraction on the multidimensional raw data, extracting multidimensional feature data that can characterize data change trends. For example, vehicle status change trends can be extracted from vehicle status data, network status change trends from network status data, and environmental change trends from environmental data, thereby forming a more comprehensive and clearer basis for data analysis.

[0064] Step S220 , performing at least one of network coverage prediction, user demand prediction, and network performance prediction on the multi-dimensional feature data to obtain a prediction result.

[0065] After obtaining multidimensional feature data, data prediction can be performed in at least one of the following three aspects: network coverage, user demand, and network performance. This data serves as the basis for calculating network switching strategies. Network coverage prediction can be based on some or all of the multidimensional feature data to predict network coverage for the vehicle's future road sections. For example, it can predict the 4G and 5G signal strengths for the road section the vehicle will travel in the next 10 seconds. User demand prediction can predict the service requirements a user may need in the future, such as whether a user might switch to video calls during peak network usage hours. Network performance prediction can predict network load in future time periods, such as whether the vehicle will pass through densely populated areas, experience network congestion, and predict network latency and packet loss rates. Data prediction can be achieved using methods such as time series prediction, regression analysis, or machine learning. Network coverage prediction, user demand prediction, or network performance prediction can be performed independently to obtain a prediction result. Alternatively, predictions can be performed for two or all of these separately to obtain a comprehensive prediction result.

[0066] Step S230: Determine the optimal network switching strategy and the network switching timing based on the prediction result and the current network usage status.

[0067] The current network usage status may include at least one of current network coverage, current user demand (or behavior), and current network performance. The current network usage status is combined with the aforementioned prediction results to determine the optimal network switching strategy. For example, the current network usage status can be combined with the network coverage prediction results, user demand prediction results, and network performance prediction results to determine the optimal network switching strategy. For example, if the 5G network coverage on a future road section is predicted to meet preset coverage conditions and the user has a video call requirement, the network switching strategy can be determined to switch to the 5G network during a preset time period in the future. If the 4G network signal is predicted to be stable and the 5G network is disconnected on the future road section, the 4G connection can be maintained. Alternatively, the optimal network switching timing can be determined based on the vehicle's driving status, such as its location and speed. For example, a network switch can be triggered when the vehicle approaches the network coverage boundary.

[0068] For example, deep learning or reinforcement learning can be used, with the above prediction results and the current network usage status as input data, and according to pre-set optimization goals (such as minimizing network switching delay and signal fluctuations), the corresponding network switching strategy (target network to be switched, switching frequency band, signal threshold, etc.) can be output.

[0069] Step S240: Generate a network switching instruction according to the optimal network switching strategy and network switching timing.

[0070] Step S250: Send a network switching instruction to the vehicle.

[0071] The cloud can assemble the optimal network switching strategy into a network switching instruction according to a preset data format and send it to the vehicle. After receiving the network switching instruction, the vehicle will parse it and execute the corresponding in-vehicle network switching based on the parsed result.

[0072] In related technologies, the generation of in-vehicle network switching strategies is implemented on the vehicle side. Specifically, implementing in-vehicle network switching on the vehicle side relies on the vehicle's hardware and software, which uses thresholds determined based on network signal parameters to determine network switching. Having the vehicle side calculate the network switching strategy increases the hardware cost and energy consumption of the vehicle-side equipment, thereby increasing the computational burden on the vehicle-side hardware and software, and resulting in low switching efficiency.

[0073] In this embodiment, by offloading the generation of in-vehicle network switching plans to the cloud, the vehicle only needs to collect and report data and execute the network switching policy issued by the cloud. This reduces hardware costs and energy consumption on the vehicle side, and reduces the computational burden on the vehicle's software and hardware. Furthermore, by leveraging the cloud's richer and more powerful computing resources to complete complex data processing and decision-making calculations, the accuracy of in-vehicle network switching is further improved, ensuring real-time switching and enhancing switching efficiency.

[0074] Thus, through the above steps S210 to S250, the multidimensional raw data reported by the vehicle to the cloud is obtained, and multidimensional feature data is obtained based on the multidimensional raw data; a prediction of network coverage, user demand, or network performance is performed on the multidimensional feature data to obtain a prediction result; based on the prediction result and the current network usage status, the optimal network switching strategy and the network switching timing are determined; based on the optimal network switching strategy and the network switching timing, a network switching instruction is generated; and the network switching instruction is issued to the vehicle. By moving the computational tasks of on-board network switching to the cloud for execution, the cloud's computing resources can be more fully utilized, reducing the computational burden on the vehicle's hardware and software, while improving switching efficiency.

[0075] In one embodiment, based on the above step S210, obtaining the multi-dimensional raw data reported by the vehicle to the cloud, and obtaining the multi-dimensional feature data based on the multi-dimensional raw data may specifically include:

[0076] According to a preset time sequence and / or spatial sequence, the multi-dimensional original data is integrated to obtain integrated data; and features are extracted from the integrated data to obtain multi-dimensional feature data.

[0077] In particular, the historical multi-dimensional raw data and the currently reported real-time multi-dimensional raw data can be combined to perform data integration according to the time series and / or spatial series. The spatial series can be the spatial location associated with the data. According to the changes in the spatial location of the vehicle, the data can be integrated in the dimension of the spatial series. For example, the spatial flow data of the network signal coverage can be integrated according to the spatial location. The time series can be the time associated with the data. The data can be integrated in the dimension of the time series according to the time axis of the vehicle's travel, for example, the vehicle speed, vehicle location, signal quality, road status and other data at a certain timestamp can be integrated, or the same data can be integrated in chronological order. By integrating the multi-dimensional raw data reported by the vehicle end, a comprehensive view representing information such as the vehicle's driving trajectory and network coverage can be constructed. Next, the integrated data can be preprocessed, such as cleaning and normalizing the integrated data and performing feature extraction to obtain the above-mentioned multi-dimensional feature data.

[0078] This embodiment can achieve data standardization and normalization based on data integration processing, thereby improving the accuracy of data prediction as a whole.

[0079] In one embodiment, the multidimensional original data includes at least one of the following: vehicle status data, network status data, navigation data, user behavior data and environmental data; the multidimensional feature data includes at least one of the following: vehicle status change trend, network status change trend, driving trajectory change trend, user behavior pattern change trend and environmental change trend.

[0080] The vehicle status data may include the vehicle's location, speed, and direction of travel during driving; network status data may include 4G / 5G network signal strength, such as Reference Signal Received Power (RSRP), Reference Signal Received Quality (RSRQ), bandwidth, latency, and packet loss rate; navigation data may include the estimated driving route, next intersection, and estimated arrival time obtained by the in-vehicle navigation system; user behavior data may include the type of services currently used by the user, such as voice calls, video playback, and data downloads, as well as the user's historical behavior patterns, such as usage habits during peak hours; and environmental data may include weather conditions, such as rain, snow, fog, and sunny days, and traffic conditions, such as congestion, smooth traffic, and construction. The vehicle side can obtain the above environmental data by accessing third-party traffic data providers and using the weather application programming interface (API).

[0081] After obtaining multi-dimensional raw data, we can extract the changing trends of various types of data. For example, we can extract the changing trends of vehicle status (such as the changing trends of vehicle speed) from vehicle status data; extract the changing trends of network status (such as the changing trends of signal strength) from network status data; extract the changing trends of driving trajectory from navigation data; extract the changing trends of user behavior patterns from user behavior data; and extract the changing trends of environment from environmental data.

[0082] In some of the embodiments, real-time traffic flow, road conditions, and weather condition data can be introduced and integrated with vehicle status and network status data to construct more comprehensive raw data and improve the accuracy of data prediction and network switching strategy calculation.

[0083] In one embodiment, performing at least one of network coverage prediction, user demand prediction, and network performance prediction on the multi-dimensional feature data to obtain a prediction result may include:

[0084] Based on the trend of environmental changes, combined with historical network coverage data and driving trajectory change trends, the network coverage situation is predicted to obtain the network coverage prediction result; based on the trend of user behavior pattern changes, the user behavior in the preset time period is predicted to obtain the user demand prediction result; based on the trend of network status changes, the network performance in the preset time period is predicted to obtain the network performance prediction result.

[0085] Based on a pre-built network coverage prediction model, environmental change trends (including map data), historical network coverage, and vehicle trajectory change trends, network coverage can be predicted for a road section that the vehicle will reach within a preset time period. This prediction can yield a network coverage prediction result, for example, a prediction of the 4G and 5G signal strengths for the road section the vehicle will reach within the next 10 seconds. This network coverage prediction model can be derived using machine learning, regression algorithms, or time series algorithms, and is not specifically limited in this embodiment.

[0086] Additionally, future service demands can be predicted based on changing trends in user behavior patterns, such as the transition from historical user behavior patterns to current behavior. For example, it can be predicted whether a user will switch to video calls, perform over-the-air (OTA) upgrades, or perform other needs at a certain time in the future. Furthermore, network performance can be predicted for future periods based on changing trends in network status, such as future network load conditions (whether there will be network congestion in user-dense areas), network latency, and packet loss rates. Accordingly, user demand prediction can be based on pre-built user demand prediction models, and network performance prediction can be based on pre-built network performance prediction models. These user demand prediction models and network performance prediction models can also be implemented based on technologies such as machine learning and regression algorithms.

[0087] This embodiment can respectively complete network coverage prediction, user demand prediction, and network performance prediction based on multi-dimensional feature data, thereby providing accurate and comprehensive data basis for the calculation of vehicle network switching strategy.

[0088] Additionally, in one embodiment, determining an optimal network switching strategy and determining a network switching timing based on the prediction results and the current network usage status may include:

[0089] Based on the prediction results and the current network usage status, the optimal network switching strategy is determined based on the trained strategy calculation model; the network switching timing is determined based on the vehicle's real-time driving information.

[0090] The policy calculation model can be a deep learning model or a reinforcement learning model. During the training phase, an objective function can be constructed that includes network coverage, user demand, network performance, and vehicle network-related indicators, and different weights are assigned to network coverage, user demand, and network performance. By training the model, the above weights are adjusted and optimized. The training process can be based on a variety of constraints, such as considering indicators such as bandwidth, latency, network resource utilization, and device energy consumption, and the constraints can be flexibly adjusted according to actual conditions. Ultimately, the trained policy calculation model can output adapted vehicle network indicators such as the target network, frequency band, network, switching time, etc. based on the input network coverage prediction results, user demand prediction results, and network performance prediction results.

[0091] Furthermore, after determining the optimal network switching strategy, the optimal time to initiate a network switch can be determined based on the vehicle's real-time location and speed. For example, a network switch can be triggered when the vehicle approaches the coverage boundary of the currently used network.

[0092] This embodiment fully leverages the cloud's hardware and software capabilities, leveraging its powerful computing power to accurately calculate network switching strategies. This improves the accuracy and timeliness of network switching strategy generation, and enhances switching efficiency. Furthermore, by integrating multi-dimensional data such as vehicle status data, environmental data, and network status data, it enables data prediction across multiple dimensions and the final network switching strategy calculation, further enhancing the accuracy of network switching strategies.

[0093] Additionally, in one embodiment, generating a network switching instruction according to the optimal network switching strategy and network switching timing may include:

[0094] According to the optimal network switching strategy, the target network to be switched and the switching parameters are determined; the target network, the switching parameters and the network switching timing are assembled into a network switching instruction.

[0095] For example, the cloud can generate a network switching instruction containing the following content: target network (4G or 5G), switching timing (arrival location or time), frequency band, signal threshold, and switching conditions, etc. For example, it can be expressed in the form of JSON characters:

[0096] "json

[0097] {

[0098] "switch_instruction": {

[0099] "target_network": "5G",

[0100] "switchTiming": {

[0101] "type": "location_based",

[0102] "threshold": {"RSRP": -70},

[0103] },

[0104] "switch_params": {

[0105] "hysteresis": 5,

[0106] "timer": 2,

[0107] "band":"8"

[0108] }

[0109] }

[0110] }";

[0111] "switch_instruction" represents the switch instruction; "target_network" represents the target network; "switchTiming" represents the switch timing; "type" represents the switch type (in this example, location-based); and "threshold" represents the signal threshold (in this example, RSRP with a threshold of -70). "switch_params" represents the switch parameters, including hysteresis, timer, and band.

[0112] This embodiment converts the network switching strategy into a network switching instruction that can be parsed and recognized by the vehicle, ensuring that the vehicle can accurately and smoothly complete the network switch based on the network switching strategy. This predictive switching reduces network interruptions and service lags. Furthermore, preemptive switching at the edge of network coverage or in areas with weak signals ensures the stability of the vehicle's network connection.

[0113] In one embodiment, obtaining multi-dimensional raw data reported by the vehicle to the cloud may include:

[0114] Receive multi-dimensional raw data reported by the vehicle using a publish-subscribe messaging protocol. For example, the vehicle can exchange data with the cloud via Message Queuing Telemetry Transport (MQTT) or other publish-subscribe messaging protocols, thereby improving data link security, reducing data transmission latency and overhead, and further enhancing network switching efficiency.

[0115] Additionally, in one embodiment, sending a network switching instruction to the vehicle may include:

[0116] Determine whether the communication status of the data link between the cloud and the vehicle is normal; if so, send a network switching instruction to the vehicle based on the data link; otherwise, send a network switching instruction to the vehicle via text signaling.

[0117] The data link can be a persistent data channel established over a mobile network. This requires a data link to be established first, with session persistence capabilities, and capable of carrying IP protocol data such as HTTP, MQTT, and TCP / UDP. The text signaling can be discrete short messages transmitted over the mobile network signaling channel, using the cellular network control plane protocol. When the cloud sends a network switching command to the vehicle, it first needs to determine whether the vehicle's private network is online. Specifically, it needs to determine whether the communication status of the dedicated data link between the cloud and the vehicle, such as the Access Point Name (APN) link, is normal. This determination can be based on a heartbeat mechanism or persistent connection status detection. If the vehicle's private network is online, the network switching command can be sent to the vehicle via the private network. If the vehicle's private network is offline, the network switching command can be sent to the vehicle via text signaling, such as an SMS message.

[0118] This embodiment selects different channels to send network switching instructions to the vehicle side according to the online status of the private network on the vehicle side, so that the network switching instructions can be accurately and smoothly received by the vehicle side.

[0119] In some of the embodiments, after receiving the network switching instruction sent by the cloud, the vehicle side parses and checks the specific content, determines the switching time, and sets a countdown timer; when the timer indicates that the switching time has arrived, the vehicle side determines whether there is an emergency service such as ECALL using the current network at this moment. If not, the vehicle network can be switched directly according to the network switching strategy, or the relevant frequency band can be selected. If an emergency service is in progress at this time, the network switching instruction can be ignored. In addition, when switching the network, the continuity of the service can be maintained, for example, when switching from 5G to 4G network, the video playback can be ensured to be uninterrupted. Afterwards, the vehicle side can feedback the execution results of the network switch to the cloud. The cloud introduces a feedback mechanism to continuously optimize the switching strategy based on the switching results of the vehicle side and user feedback.

[0120] Figure 3 is a flow chart of a vehicle network switching method in some embodiments, such as Figure 3 As shown, the vehicle network switching method includes the following steps:

[0121] Step S301: The vehicle collects and reports multi-dimensional raw data;

[0122] In step S302, the cloud uses artificial intelligence (AI) technology to extract multidimensional feature data from the multidimensional raw data to perform network coverage prediction, user demand prediction, and network performance prediction, and obtain prediction results.

[0123] Step S303: The cloud determines the optimal network switching strategy based on the prediction results and the current network usage status, and generates a network switching instruction based on the optimal network switching strategy;

[0124] Step S304: The cloud sends a network switching instruction to the vehicle.

[0125] Step S305: The vehicle receives and executes the network switching instruction, and feeds back the switching result to the cloud.

[0126] The above steps S301 to S305 significantly reduce the computing requirements of vehicle-side hardware and software by transplanting the computing tasks of complex network switching strategies to the cloud, thereby reducing equipment costs and energy consumption; by fully utilizing the powerful computing power and optimized switching mechanism of the cloud, the real-time and accuracy of switching are significantly improved; the cloud can centrally manage and optimize switching strategies, thereby improving the robustness and adaptability of network switching.

[0127] Figure 4 Flowchart of the vehicle-side data collection and reporting method in some embodiments, such as Figure 4 As shown, the vehicle-side data collection and reporting method may include:

[0128] Step S401, obtaining user behavior data, vehicle status data and network status data; wherein, each type of data may be obtained once every 1 second.

[0129] Step S402: When changes are detected in the environmental data, navigation data, and sensor data, the changed data are reported.

[0130] Step S403: Packaging the collected data.

[0131] In step S404, the data collected in steps S401 and S402 are transmitted to the cloud via a private network using the MQTT protocol.

[0132] In step S405, the cloud receives the data reported by the vehicle and stores it in the database.

[0133] The above steps S401 to S405 implement a complete data interaction process between the vehicle and the cloud, which can ensure the real-time and accuracy of data transmission.

[0134] Figure 5 Flowchart of a method for processing multi-dimensional raw data in the cloud in some embodiments, such as Figure 5 As shown, the method for processing the multidimensional raw data includes:

[0135] Step S501: integrate the multi-dimensional raw data reported by the vehicle side into time series and / or spatial series, combine the historical multi-dimensional raw data and the current multi-dimensional raw data, and construct a comprehensive view of the vehicle's trajectory and network coverage.

[0136] Step S502 , cleaning, normalizing and feature extracting the data integrated in step S501 to obtain multi-dimensional feature data.

[0137] Step S503: Perform network coverage prediction, user demand prediction, and network performance prediction based on the multi-dimensional feature data.

[0138] Step S504 : Calculate the optimal switching strategy using the trained strategy calculation model according to the prediction result and the current network usage status.

[0139] Step S505: Calculate the switching timing.

[0140] The above steps S501 to S505 can improve the accuracy of prediction and thus improve the accuracy of network switching strategy calculation by integrating multi-dimensional data such as vehicle status, environmental data, and network status.

[0141] Figure 6 Flowchart of a method for issuing a network switching instruction to the cloud in some embodiments, such as Figure 6 As shown, the network switching instruction issuing method may include:

[0142] Step S601: Generate a network switching instruction based on the optimal network switching strategy; wherein the network switching instruction may include the target network to be switched (4G or 5G), switching timing, switching parameters, etc.

[0143] Step S602, determine whether the vehicle private network is online; if so, execute step S603; otherwise, execute step S604.

[0144] Step S603: Send a network switching instruction to the vehicle via MQTT.

[0145] Step S604: Send a network switching instruction to the vehicle via SMS.

[0146] The above steps S601 to S604 can accurately construct the network switching instruction based on the network switching strategy, and successfully send the network switching instruction to the vehicle side through a complete channel, so that the vehicle side can accurately and timely perform the network switching operation.

[0147] Figure 7 Flowchart of a method for executing network switching on a vehicle side in some embodiments, such as Figure 7 As shown, the vehicle-side network switching method may include:

[0148] Step S701: The vehicle receives a network switching instruction from the cloud.

[0149] In step S702, the vehicle-side ternary content addressable memory (TCAM) sets a countdown according to the switching timing sent by the cloud.

[0150] In step S703, when the vehicle side determines that the switching time has arrived, it determines whether the network switching can be executed immediately; if so, it executes step S704; otherwise, it executes step S705; among them, whether the network switching can be executed directly can be determined by determining whether there is an emergency task running at present.

[0151] Step S704: The vehicle performs network switching according to the network switching instruction.

[0152] Step S705: The vehicle ignores the network switching instruction and does not execute it.

[0153] In step S706, the vehicle feeds back the network status to the cloud.

[0154] The above steps S701 to S706 can avoid conflicts between network switching and tasks being executed on the vehicle side, which may lead to failure of vehicle-side task execution, and can enable the cloud to obtain network changes on the vehicle side in a timely manner, thereby optimizing the network switching strategy.

[0155] Optionally, in some other embodiments, in order to further reduce latency, some computing tasks can be deployed on an edge server close to the vehicle or a roadside unit (RSU) on the road based on edge computing technology, thereby reducing the delay in data transmission to the cloud and improving the response speed of network switching.

[0156] Optionally, in some other embodiments, the cloud can also dynamically allocate computing resources to the computing tasks of the network switching strategy based on the real-time network load and network status, optimize the switching strategy, and ensure the stable operation of the network switching strategy tasks during peak hours and complex driving scenarios.

[0157] Additionally, in some embodiments, the cloud can continuously optimize various prediction models and switching strategies based on historical multi-dimensional raw data and real-time multi-dimensional raw data, and realize online updates of models through edge computing and federated learning technologies, thereby improving the accuracy of predictions and the intelligence level of switching.

[0158] Next, we will use the vehicle network switching in tunnel scenarios, vehicle network switching on highways, and vehicle network switching on urban roads as examples to illustrate.

[0159] For tunnel scenarios: When a vehicle travels on a highway, it collects multi-dimensional raw data and reports it to the cloud in real time. The cloud predicts an underground tunnel 10 kilometers ahead of the vehicle. Because there's no 5G signal inside the tunnel, the cloud considers the vehicle's current speed and location and issues a network switch command to the vehicle, instructing it to switch to the 4G network in five minutes. Upon receiving this command, the vehicle switches to the 4G network before entering the tunnel, avoiding internet access issues after the vehicle enters the tunnel.

[0160] For highway scenarios: When a vehicle is traveling on a highway and the driver is using the in-vehicle system for navigation and online music playback, the vehicle's high-speed movement requires interaction between the cloud and the vehicle to switch between different base stations or networks to ensure connection stability and data continuity. Specifically, while the vehicle is driving, the in-vehicle system monitors the current network's signal strength, latency, and bandwidth in real time. When network conditions change—for example, when the vehicle approaches the edge of a base station's coverage, signal strength begins to decrease and latency increases—a seamless network switch is implemented based on a network switching strategy calculated in the cloud, ensuring continuity of navigation and music playback. The vehicle can then feed back the switching results to the cloud, which records performance data during the switching process and optimizes future switching strategies. This enables seamless network switching for both the driver and passengers, improving the user experience. The cloud enables rapid computation and decision-making, reducing switching latency. Deploying switching strategy calculations to the cloud reduces the use of in-vehicle hardware resources and enhances the stability of the in-vehicle system.

[0161] For urban road scenarios: On urban roads, vehicles may frequently travel between different network coverage areas, such as Wi-Fi hotspots, 4G / 5G base stations, and dedicated short-range communications (DSRC). Traditional handover solutions can cause network delays or interruptions, impacting in-vehicle entertainment, navigation, and vehicle-to-everything (V2X) features. In this embodiment, when a vehicle enters an urban area, the vehicle's system detects multiple network signal sources (such as roadside Wi-Fi, 4G networks, and DSRC). The cloud evaluates the network status of each network and dynamically plans the optimal network handover strategy based on real-time data and urban traffic conditions. The vehicle's in-vehicle system, following instructions from the cloud, prioritizes low-latency, high-bandwidth networks for handover, ensuring smooth operation of the vehicle's systems. The cloud monitors the vehicle's network status in real time and dynamically adjusts the handover strategy based on vehicle location and traffic changes. Multi-network collaboration enables full utilization of multiple network resources and improves connection stability. The cloud optimizes the handover strategy in real time, adapting to complex urban traffic conditions. Timely network handover ensures uninterrupted in-vehicle entertainment and navigation, improving user satisfaction.

[0162] This embodiment also provides an in-vehicle network switching device for implementing the above-mentioned embodiments and preferred implementations. Details already described will not be repeated. The terms "module," "unit," "subunit," etc. used below may refer to a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation using hardware, or a combination of software and hardware, is also possible and contemplated.

[0163] Figure 8 This is a structural block diagram of the vehicle network switching device 80 for the cloud in this embodiment. Figure 8 As shown, the vehicle network switching device 80 includes: an acquisition module 81, a prediction module 82, a strategy determination module 83, an instruction generation module 84 and an instruction issuing module 85; wherein:

[0164] The acquisition module 81 is used to obtain the multi-dimensional original data reported by the vehicle side to the cloud, and obtain multi-dimensional feature data based on the multi-dimensional original data; the prediction module 82 is used to perform at least one of network coverage prediction, user demand prediction and network performance prediction on the multi-dimensional feature data to obtain the prediction result; the strategy determination module 83 is used to determine the optimal network switching strategy and determine the network switching timing based on the prediction result and the current network usage status; the instruction generation module 84 is used to generate a network switching instruction based on the optimal network switching strategy and the network switching timing; the instruction issuing module 85 is used to issue the network switching instruction to the vehicle side.

[0165] It should be noted that the above modules can be functional modules or program modules, and can be implemented through software or hardware. For modules implemented through hardware, the above modules can be located in the same processor; or the above modules can be located in different processors in any combination.

[0166] It should be noted that, for specific examples in this embodiment, reference may be made to the examples described in the above embodiments and optional implementation modes, and will not be repeated in this embodiment.

[0167] In addition, this embodiment also provides a vehicle network switching system. Figure 9 FIG. 9 is a structural diagram of the vehicle network switching system 90 of this embodiment. Figure 9 As shown, the vehicle-mounted network switching system 90 includes: a vehicle-end 92 and a cloud-end 94; the vehicle-end 92 and the cloud-end 94 are communicatively connected; the vehicle-end 92 is used to collect multi-dimensional raw data of the vehicle and report the collected multi-dimensional raw data to the cloud-end 94; the vehicle-end 92 is also used to receive and execute the network switching instructions issued by the cloud-end; the cloud-end 94 is used to execute the vehicle-mounted network switching method provided by any of the above embodiments.

[0168] It should be understood that the specific embodiments described herein are only used to explain this application and are not used to limit it. Based on the embodiments provided in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0169] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0170] Obviously, the accompanying drawings are merely examples or embodiments of the present application. A person skilled in the art can also apply the present application to other similar situations based on these drawings without inventive effort. Furthermore, it is understandable that, although the work involved in this development process may be complex and lengthy, certain design, manufacturing, or production changes based on the technical content disclosed in this application are merely routine technical means for a person skilled in the art and should not be considered to constitute a deficiency in the disclosure of the present application.

[0171] The term "embodiment" as used in this application refers to specific features, structures, or characteristics described in conjunction with the embodiment that can be included in at least one embodiment of this application. The appearance of this phrase in various places in the specification does not necessarily mean that the embodiment is the same, nor does it mean that it is mutually exclusive with other embodiments and is independent or optional. It is understood, either explicitly or implicitly, by those skilled in the art that the embodiments described in this application can be combined with other embodiments when there is no conflict.

[0172] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of patent protection. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A vehicle network switching method, characterized in that: For use in the cloud, the method includes: Acquire multidimensional raw data reported by the vehicle end to the cloud, and obtain multidimensional feature data based on the multidimensional raw data; Performing at least one of network coverage prediction, user demand prediction, and network performance prediction on the multi-dimensional feature data to obtain a prediction result; Determining an optimal network switching strategy and a network switching timing based on the prediction results and the current network usage status; generating a network switching instruction according to the optimal network switching strategy and the network switching timing; Send the network switching instruction to the vehicle end.

2. The vehicle network switching method according to claim 1, characterized in that: Acquiring multidimensional raw data reported by the vehicle to the cloud, and obtaining multidimensional feature data based on the multidimensional raw data, including: Integrating the multidimensional raw data according to a preset time sequence and / or spatial sequence to obtain integrated data; Feature extraction is performed on the integrated data to obtain the multi-dimensional feature data.

3. The vehicle network switching method according to claim 2, wherein: The multi-dimensional raw data includes at least one of the following: vehicle status data, network status data, navigation data, user behavior data, and environmental data; The multi-dimensional feature data includes at least one of the following: a vehicle status change trend, a network status change trend, a driving trajectory change trend, a user behavior pattern change trend, and an environment change trend.

4. The vehicle network switching method according to claim 3, characterized in that: Performing at least one of network coverage prediction, user demand prediction, and network performance prediction on the multi-dimensional feature data to obtain a prediction result, including: According to the environmental change trend, combined with historical network coverage data and driving trajectory change trends, network coverage is predicted to obtain a network coverage prediction result; According to the changing trend of user behavior patterns, predict the user behavior in the preset time period and obtain the user demand prediction results; The network performance of the preset time period is predicted based on the network status change trend to obtain a network performance prediction result.

5. The vehicle network switching method according to claim 1, characterized in that: Determine the optimal network switching strategy and the timing of network switching based on the prediction results and the current network usage status, including: Determining an optimal network switching strategy based on the trained strategy calculation model according to the prediction results and the current network usage status; Determine the network switching timing based on the vehicle's real-time driving information.

6. The vehicle network switching method according to claim 1, characterized in that: Generating a network switching instruction according to the optimal network switching strategy and the network switching timing includes: Determining a target network to be switched and switching parameters according to the optimal network switching strategy; The target network, the switching parameters, and the network switching opportunity are assembled into a network switching instruction.

7. The vehicle network switching method according to claim 1, characterized in that: Obtain multi-dimensional raw data reported by the vehicle to the cloud, including: Based on a publish-subscribe mode message transmission protocol, the multi-dimensional original data reported by the vehicle side is received.

8. The vehicle network switching method according to any one of claims 1 to 7, characterized in that: Sending the network switching instruction to the vehicle terminal includes: Determine whether the communication status of the data link between the cloud and the vehicle is normal; if so, issue the network switching instruction to the vehicle based on the data link; Otherwise, the network switching instruction is sent to the vehicle end via text signaling.

9. A vehicle network switching device, characterized in that: Used in the cloud; the device includes: an acquisition module, a prediction module, a strategy determination module, an instruction generation module and an instruction issuing module; wherein: The acquisition module is used to acquire the multi-dimensional original data reported by the vehicle end to the cloud, and obtain multi-dimensional feature data based on the multi-dimensional original data; The prediction module is configured to perform at least one of network coverage prediction, user demand prediction, and network performance prediction on the multi-dimensional feature data to obtain a prediction result; The strategy determination module is used to determine the optimal network switching strategy and the network switching timing based on the prediction result and the current network usage status; The instruction generation module is used to generate a network switching instruction according to the optimal network switching strategy and the network switching timing; The instruction issuing module is used to issue the network switching instruction to the vehicle end.

10. A vehicle network switching system, characterized in that: include: Vehicle-side and cloud-side; The vehicle side is communicatively connected with the cloud side; The vehicle side is used to collect multi-dimensional raw data of the vehicle and report the collected multi-dimensional raw data to the cloud; the vehicle side is also used to receive and execute the network switching instruction issued by the cloud; The cloud is used to execute the vehicle network switching method according to any one of claims 1 to 8.

Images