OTA grading upgrading method and system based on network resource optimization and medium

Abstract

The invention relates to an OTA grading upgrading method and system based on network resource optimization and a medium, and relates to the field of OTA upgrade.The method comprises the steps that a user upgrading willingness dynamic evaluation model is built; establishing a network resource grading occupation mechanism; an intelligent bandwidth dynamic allocation algorithm is adopted; and establishing a three-dimensional matching model of the user demand, the network state and the upgrade content to carry out OTA upgrade. According to the method, the OTA upgrading success rate and the network resource utilization rate are improved, so that the user complaint rate is reduced.

Description

Technical Field

[0001] The present invention relates to the field of OTA upgrades, and in particular to an OTA hierarchical upgrade method, system, and medium based on network resource optimization. Background Art

[0002] With the development of intelligent new energy vehicles, the complexity of in-vehicle software systems continues to increase. OTA (Over-The-Air) upgrade technology has become a key means of ensuring continuous optimization of vehicle functions and safety. Currently, mainstream OTA upgrade solutions use a fixed-priority queue polling mechanism to push upgrade packages to vehicles in batches according to preset time windows and geographical regions. This approach alleviates network congestion to a certain extent and improves the success rate of upgrades.

[0003] However, the existing OTA upgrade strategy still has some shortcomings. First, the fixed priority mechanism cannot dynamically identify users' actual upgrade needs, resulting in some high-demand users being unable to obtain critical updates in a timely manner. Second, simple timestamp queuing can easily allow malicious users to preempt resources through repeated requests. Furthermore, emergency security patches and regular functional updates share the same transmission channel, affecting the timeliness of critical updates. In addition, the practice of dividing the upgrade order by vehicle grade puts users of basic models at a long-term disadvantage in updating. Finally, the existing strategy does not fully consider the dynamic changes in users' actual needs and network status, resulting in insufficient resource utilization and a long average delay time for emergency security updates. Summary of the Invention

[0004] In view of the shortcomings of the prior art described above, the purpose of the present invention is to provide an OTA hierarchical upgrade method, system and medium based on network resource optimization, which improves the success rate of OTA upgrades and network resource utilization, thereby reducing user complaint rates.

[0005] To achieve the above objectives, the present invention adopts the following technical solutions.

[0006] In a first aspect, the present invention provides an OTA hierarchical upgrade method based on network resource optimization, which adopts the following technical solutions: Build a dynamic evaluation model for user upgrade willingness; Establish a hierarchical network resource occupation mechanism; Adopt intelligent bandwidth dynamic allocation algorithm; Establish a three-dimensional matching model of user needs, network status and upgrade content for OTA upgrades.

[0007] Furthermore, in the above-mentioned upgrade method, the step of constructing a dynamic evaluation model of user upgrade willingness includes: Collect vehicle usage data, including average daily mileage, charging behavior, interactive behavior, and environmental factors; and establish an upgrade willingness index calculation model; A user upgrade willingness index is calculated based on the vehicle usage data and the upgrade willingness index calculation model.

[0008] Furthermore, in the above-mentioned upgrade method, the upgrade willingness index calculation model is: Among them, W u is the upgrade willingness index, αi is the weight coefficient of each vehicle usage data, Xi is the actual value of each vehicle usage data, μi is the mean of each vehicle usage data, σi is the standard deviation of each vehicle usage data, Sc is the system urgency coefficient, and β is the safety weight factor.

[0009] Furthermore, in the above-mentioned upgrade method, the establishment of a hierarchical network resource occupation mechanism includes: The upgrade tasks are divided into three transmission channels: security emergency, function optimization and system maintenance.

[0010] Furthermore, the above-mentioned upgrade method further includes: Initial bandwidth ratios are set for the three transmission channels of security emergency, function optimization and system maintenance respectively.

[0011] Furthermore, in the above-mentioned upgrade method, the intelligent bandwidth dynamic allocation algorithm is adopted, including: Monitor real-time network quality; Calculate the priority score of each vehicle to be upgraded; Based on the real-time network quality and the limit score of each vehicle, the transmission rate of each transmission channel is dynamically adjusted.

[0012] Furthermore, in the above-mentioned upgrade method, the priority score is the product of the user's upgrade willingness and a preset service quality coefficient.

[0013] Furthermore, the above-mentioned upgrade method further includes: Predict the future user vehicle usage status through LSTM neural network.

[0014] Furthermore, in the above-mentioned upgrade method, the method of predicting the future vehicle usage status of the user through the LSTM neural network includes: Collect historical vehicle usage data; Train the LSTM neural network model; The trained LSTM neural network model is used to predict the user's vehicle usage status within a preset time period in the future.

[0015] Furthermore, in the above-mentioned upgrade method, the step of establishing a three-dimensional matching model of user needs, network status, and upgrade content includes: Establish a user demand perception layer to collect vehicle status data in real time; Establish a network status monitoring layer to monitor network parameters; A dynamic resource allocation layer is established, and an improved Hungarian algorithm is used for multi-dimensional resource matching.

[0016] Furthermore, in the above-mentioned upgrading method, the vehicle status data includes SOC value, parking location and recent driving trajectory.

[0017] Furthermore, the above-mentioned upgrade method further includes: Select an upgrade window based on user behavior prediction results.

[0018] In a second aspect, the present invention provides an OTA hierarchical upgrade system based on network resource optimization, which adopts the following technical solutions: User upgrade willingness evaluation module, used to build a dynamic evaluation model for user upgrade willingness; Network resource classification module, used to establish a network resource classification occupation mechanism; Bandwidth allocation module, used to adopt intelligent bandwidth dynamic allocation algorithm; A three-dimensional matching module, used to establish a three-dimensional matching model of user needs, network status and upgrade content; and An upgrade execution module is used to execute OTA upgrade based on the three-dimensional matching model.

[0019] In a third aspect, the present invention provides a readable storage medium, which adopts the following technical solution: A readable storage medium stores computer instructions, which, when executed by a processor, implement the OTA hierarchical upgrade method based on network resource optimization as described in any one of the first aspects above.

[0020] In summary, compared with the prior art, the present invention has at least one of the following beneficial technical effects: By building a dynamic evaluation model for user upgrade willingness, we can more accurately identify users' real upgrade needs and avoid wasting network resources; by establishing a hierarchical network resource occupation mechanism and adopting an intelligent bandwidth dynamic allocation algorithm, we can improve network resource utilization and ensure the timeliness of key updates; by establishing a three-dimensional matching model, we can achieve dynamic matching of user needs, network status and upgrade content, and improve the success rate of OTA upgrades and user satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0022] Figure 1 It is a flowchart of a specific embodiment of the OTA hierarchical upgrade method based on network resource optimization of the present invention.

[0023] Figure 2 It is a flowchart of a specific embodiment of the OTA hierarchical upgrade method based on network resource optimization of the present invention.

[0024] Figure 3 It is a principle diagram of a specific embodiment of the OTA hierarchical upgrade method based on network resource optimization of the present invention.

[0025] Figure 4 It is a structural diagram of a specific embodiment of the OTA hierarchical upgrade system based on network resource optimization of the present invention. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of this application. In addition, it should be understood that the specific embodiments described herein are only used to illustrate and explain the present application and are not used to limit the present application.

[0027] It should be noted that the order of description of the following embodiments does not limit the preferred order of the embodiments of the present application. In addition, in the following embodiments, the description of each embodiment has its own focus. For parts not described in detail in one embodiment, please refer to the relevant description of other embodiments.

[0028] The method steps described in the embodiments of the present invention may be executed in the order described in the specific implementation manner, or the execution order of each step may be adjusted according to actual needs, provided that the technical problem can be solved. The execution order of each step will not be listed here one by one.

[0029] Reference Figure 1 An embodiment of the present invention provides an OTA hierarchical upgrade method based on network resource optimization, which includes the following main steps.

[0030] S1: Build a dynamic evaluation model for user upgrade willingness. This model collects vehicle usage data, such as average daily mileage, charging behavior, interactive behavior, and environmental factors, to establish an upgrade willingness index calculation model and calculate the user upgrade willingness index. This dynamic evaluation method can reflect users' actual upgrade needs in real time.

[0031] S2: Establish a hierarchical network resource allocation mechanism. This mechanism divides upgrade tasks into three transmission channels: security emergency response, function optimization, and system maintenance, and sets an initial bandwidth allocation for each channel. This hierarchical mechanism helps to rationally allocate network resources and ensure the timely delivery of critical updates.

[0032] S3 utilizes an intelligent dynamic bandwidth allocation algorithm. This algorithm monitors real-time network quality, including base station load, channel quality index, and bit error rate, and dynamically adjusts the transmission rate of each channel based on the priority of each vehicle to be upgraded. This approach improves bandwidth utilization and adapts to changing network conditions.

[0033] S4 establishes a three-dimensional matching model for user needs, network status, and upgrade content. This model includes a user needs perception layer, a network status monitoring layer, and a dynamic resource allocation layer. The user needs perception layer collects real-time vehicle status data, such as SOC value, parking location, and recent driving trajectory. The network status monitoring layer monitors network parameters. The dynamic resource allocation layer uses an improved Hungarian algorithm for multi-dimensional resource matching.

[0034] During OTA updates, a three-dimensional matching model comprehensively considers user needs, network status, and update content to select the optimal upgrade time and method for each user. During the upgrade process, network conditions and user behavior are continuously monitored, and upgrade strategies are adjusted in real time to ensure a smooth upgrade. After the upgrade is complete, the results are recorded and the user upgrade willingness model is updated to provide data support for subsequent upgrade optimization.

[0035] According to the OTA hierarchical upgrade method based on network resource optimization described in an embodiment of the present invention, by constructing a dynamic evaluation model of user upgrade willingness, establishing a hierarchical network resource occupancy mechanism, adopting an intelligent bandwidth dynamic allocation algorithm, and establishing a three-dimensional matching model, it is possible to achieve accurate identification of user needs, efficient use of network resources, and intelligent allocation of upgrade content. This multi-dimensional collaborative upgrade strategy can improve the success rate of OTA upgrades, reduce network resource waste, shorten the response time of key updates, and improve user experience. In addition, the dynamic adjustment mechanism of the method can adapt to network conditions and user behavior changes in different scenarios, thereby maximizing the utilization efficiency of network resources while ensuring the quality of the upgrade.

[0036] Furthermore, as an embodiment of the present invention, step S1, constructing a dynamic evaluation model of user upgrade willingness, includes: Collect vehicle usage data, including average daily mileage, charging behavior, interactive behavior, and environmental factors; and establish an upgrade willingness index calculation model; A user upgrade willingness index is calculated based on the vehicle usage data and the upgrade willingness index calculation model.

[0037] In some embodiments, the collected vehicle usage data includes average daily mileage, charging behavior, interaction behavior, and environmental factors. Average daily mileage reflects the user's frequency of use. Charging behavior, including the percentage of home charging stations used and the number of fast charging times per week, reflects the user's charging habits. Interaction behavior, including the number of active screen operations and the response rate to voice commands, reflects the user's reliance on the in-vehicle system. Environmental factors include the network quality in the user's usual area, which can affect network conditions for upgrades.

[0038] The upgrade willingness index calculation model takes into account multiple influencing factors and performs weighted calculations on each indicator.

[0039] Based on collected vehicle usage data and an established upgrade willingness index calculation model, a user's upgrade willingness index is calculated. This index reflects the user's acceptance and urgency for upgrading. In some embodiments, users are categorized based on the calculated upgrade willingness index to provide a basis for subsequent upgrade strategy formulation.

[0040] By building a dynamic evaluation model of user upgrade willingness, we can comprehensively evaluate users' upgrade needs and provide data support for formulating personalized upgrade strategies, thereby improving the success rate of upgrades and user satisfaction.

[0041] In some embodiments, the expression of the upgrade willingness index calculation model is: Wu represents the upgrade willingness index, a comprehensive indicator measuring user acceptance of OTA upgrades. αi represents the weighting coefficient for each vehicle usage metric, reflecting the importance of each metric in the evaluation. Xi represents the actual value of each vehicle usage metric, μi represents the mean of each metric, and σi represents the standard deviation of each metric. Normalization using (Xi - μi) / σi eliminates dimensional differences between different metrics, making them comparable. Sc represents the system urgency coefficient, reflecting the urgency of the upgrade content. β represents the safety weighting factor, which adjusts the impact of system urgency on the final upgrade willingness index.

[0042] As shown in Table 1, the model considers several key user behavior indicators. Among them, average daily mileage and frequency of rapid acceleration reflect driving habits; the proportion of home charging stations used and the number of fast charging times per week reflect charging behavior; the number of active screen operations and voice command response rate reflect user interaction; and the resident area network quality considers the impact of environmental factors on upgrades. The weight coefficients αi for these indicators are set based on their impact on upgrade intention. For example, the weight of active screen operations is 0.2, reflecting the significant impact of users' reliance on the in-vehicle system on upgrade intention.

[0043] Table 1:

[0044] In some embodiments, the security weighting factor β ranges from 0.3 to 0.5. When the upgrade involves a critical security update, β takes a larger value, increasing the impact of system urgency on the upgrade willingness index. When the upgrade primarily involves functional optimization, β takes a smaller value, reducing the impact of system factors and determining upgrade willingness more based on user behavior characteristics.

[0045] As shown in Table 2, the solution using this upgrade willingness index calculation model exhibits significant advantages over traditional solutions. The upgrade success rate increased from 0.682 to 0.935, the user complaint rate decreased from 0.224 to 0.051, and resource utilization increased from 0.617 to 0.882. These data demonstrate that this model can more accurately assess user upgrade willingness, leading to the development of more effective upgrade strategies and improved upgrade efficiency and user satisfaction.

[0046] Table 2: Solution Type Upgrade success rate User complaint rate Resource utilization Traditional solutions 0.682 0.224 0.617 This program 0.935 0.051 0.882

[0047] Furthermore, in step S2, a hierarchical network resource occupation mechanism is established, which includes dividing the upgrade task into three transmission channels: security emergency, function optimization, and system maintenance.

[0048] In some implementations, the safety emergency channel is used to transmit urgent updates directly related to vehicle safety, such as brake system optimization, battery management system repair, etc. This channel has the highest priority, ensuring that critical safety updates can be pushed to user vehicles in a timely manner.

[0049] The function optimization channel is used to transmit functional updates that enhance the user experience, such as in-car entertainment system upgrades and navigation map updates. This channel has a lower priority than the safety and emergency channel, but a higher priority than the system maintenance channel.

[0050] The system maintenance channel is used to transmit updates related to routine maintenance and performance optimization, such as background log optimization and system stability improvement. This channel has the lowest priority and is usually transmitted when network resources are sufficient.

[0051] In some implementations, initial bandwidth allocations are set for the security emergency channel, function optimization channel, and system maintenance channel. For example, the initial bandwidth allocation for the security emergency channel is 50%, for the function optimization channel is 30%, and for the system maintenance channel is 20%. This initial allocation ensures that critical updates receive sufficient network resources while also reserving appropriate bandwidth for other types of updates.

[0052] By establishing a hierarchical network resource occupation mechanism, priority management and resource allocation for different types of upgrade tasks are achieved, the efficiency of network resource utilization is improved, and the timely push of important updates is ensured.

[0053] In some implementations, initial bandwidth usage is set for each of the three transmission channels: security emergency response, function optimization, and system maintenance. The initial bandwidth usage is set based on the importance and urgency of different types of upgrade tasks.

[0054] The safety emergency channel is used to transmit urgent updates directly related to vehicle safety, such as brake system optimizations and battery management system repairs. Given the critical impact of these updates on vehicle safety, the initial bandwidth allocation for the safety emergency channel is set at 50%. This higher bandwidth allocation ensures that critical safety updates can be delivered to users' vehicles quickly and promptly.

[0055] The feature optimization channel is used to transmit updates that enhance the user experience, such as in-car entertainment system upgrades and navigation map updates. While these updates are less urgent than security updates, they are still crucial for improving user satisfaction. Therefore, the initial bandwidth usage of the feature optimization channel can be set to 30%. This ensures a certain transmission speed while reserving resources for other types of updates.

[0056] The system maintenance channel is used to transmit updates related to routine maintenance and performance optimization, such as background log optimization and system stability improvements. These updates are typically not time-critical and can be transmitted when network resources are sufficient. With this in mind, the initial bandwidth usage for the system maintenance channel can be set to 20%.

[0057] By setting different initial bandwidth usage for the three transmission channels, we can prioritize and allocate resources for different types of upgrade tasks. This allocation scheme ensures that important updates are pushed in a timely manner while also reserving appropriate bandwidth for other types of updates, improving the efficiency of network resource utilization.

[0058] In some implementations, the initial bandwidth share is not fixed but dynamically adjusted based on actual conditions. For example, when there are no urgent security updates, some of the bandwidth of the security emergency channel can be temporarily allocated to other channels. Similarly, when network resources are sufficient, the bandwidth share of the function optimization and system maintenance channels can be appropriately increased to speed up the transmission of non-urgent updates. This dynamic adjustment mechanism further improves the utilization efficiency of network resources and makes bandwidth allocation more flexible and efficient.

[0059] Furthermore, step S3, using an intelligent bandwidth dynamic allocation algorithm includes the following steps: Monitor real-time network quality; Calculate the priority score of each vehicle to be upgraded; Based on the real-time network quality and the limit score of each vehicle, the transmission rate of each transmission channel is dynamically adjusted.

[0060] Specifically, the current network status can be assessed by monitoring multiple network parameters such as base station load rate, channel quality index (CQI), transmission bit error rate, etc. These parameters can reflect the degree of network congestion and transmission quality in real time.

[0061] Dynamically adjust the transmission rate of each transmission channel. Based on the real-time network quality and the priority score of each vehicle, dynamically adjust the bandwidth allocation of the three transmission channels of safety emergency, function optimization and system maintenance. The specific adjustment method is as follows: First, set the initial bandwidth usage based on the urgency and importance of each channel's current transmission tasks. For example, the security emergency channel can be set to 50%, the function optimization channel to 30%, and the system maintenance channel to 20%. Then, based on the real-time network quality, the TCP window size is dynamically adjusted. When the network quality is good, the TCP window size is appropriately increased to improve transmission efficiency. When the network quality is poor, the TCP window size is reduced to reduce the packet loss rate. Finally, based on the priority score of each vehicle to be upgraded, fine-grained bandwidth allocation is performed within each channel while ensuring the initial bandwidth share. Vehicles with higher priority scores will receive more bandwidth resources, ensuring that critical updates can be pushed to the user end in a timely manner.

[0062] Reference Figure 2 ,The dynamic allocation process includes the following steps: The network quality detection module reports the current channel status; Calculate the priority score P of each vehicle to be upgraded; Arrange in descending order of P value and dynamically adjust the TCP window size to achieve bandwidth allocation.

[0063] Through the above steps, intelligent dynamic allocation of network resources is achieved, bandwidth utilization is improved, and the timeliness of important updates and user experience are guaranteed.

[0064] In some embodiments, the priority score may be calculated using the following formula: P = Wu × QoS; Where P is the priority score, Wu is the user upgrade willingness index, and QoS is the service quality coefficient. The user upgrade willingness index reflects the user's actual demand for upgrades, while the service quality coefficient considers the impact of network conditions on the upgrade experience.

[0065] The User Upgrade Willingness Index (Wu) reflects users' actual desire for upgrades. This index is dynamically calculated by analyzing multiple dimensions of information, including vehicle usage data and interaction behaviors. A higher WU indicates a more urgent need for upgrades.

[0066] The preset Quality of Service (QoS) factor takes into account the impact of network conditions on the upgrade experience. This factor is pre-set based on the current network quality, such as the Channel Quality Index (CQI) and the bit error rate. A higher QoS value indicates better network conditions and is suitable for OTA upgrades.

[0067] By multiplying the user's upgrade willingness index by a preset service quality coefficient, the priority score P comprehensively considers two key factors: user demand and network conditions. Vehicles with higher scores receive higher priority in resource allocation, ensuring that critical updates are delivered to users in a timely manner while avoiding large-scale upgrade operations during poor network conditions.

[0068] In some implementations, the OTA hierarchical upgrade method based on network resource optimization further includes: Predict the future user vehicle usage status through LSTM neural network.

[0069] The LSTM neural network is used to predict the future vehicle usage status of users, including the following steps: Collect historical vehicle usage data; Train the LSTM neural network model; The trained LSTM neural network model is used to predict the user's vehicle usage status within a preset time period in the future.

[0070] In some embodiments, the collected data includes vehicle driving status, charging status, parking location, vehicle system usage records, etc. These data are collected in real time through vehicle sensors and communication modules and uploaded to the server regularly.

[0071] In some embodiments, the input layer of the LSTM neural network model includes historical data from multiple time steps, the hidden layer is composed of multiple LSTM units, and the output layer predicts vehicle usage status for the next eight hours. During model training, a backpropagation algorithm is used to optimize network parameters and minimize prediction error.

[0072] The trained LSTM neural network model is used to predict the user's vehicle usage status within a preset timeframe. In some embodiments, the preset timeframe is 8 hours. The prediction results include information such as whether the vehicle is in motion, its expected parking location, and charging requirements.

[0073] In some embodiments, the structure of the LSTM neural network model is as follows: h t =σ(W ih x t +W hh h t-1 +b h ); o t =σ(W i0 x t +W h0 h t-1 +b0); h t =o t ⊙tanh(c t ); Among them, h t is the hidden state, c t is the unit state, x t is the input, W is the weight matrix, b is the bias term, σ is the activation function, and ⊙ represents element-by-element multiplication.

[0074] Optimize the OTA upgrade process based on the prediction results of the LSTM neural network.

[0075] In some embodiments, the optimization method includes: Choose the appropriate upgrade window: Based on the predicted vehicle usage, perform OTA upgrades during periods when the vehicle is parked for long periods and network conditions are good, minimizing the impact on users' normal use. Adjust the order of push updates: Based on the predicted vehicle usage needs, prioritize push updates related to features that users are likely to use soon, improving user experience. Optimize network resource allocation: Based on the predicted vehicle location information, network resources are rationally allocated to avoid network congestion caused by large-scale upgrades of multiple vehicles in the same area at the same time.

[0076] In some embodiments, the prediction accuracy of the LSTM neural network model is evaluated by the following metrics: Among them, TP is a true positive example, TN is a true negative example, FP is a false positive example, and FN is a false negative example. The accuracy of the prediction is maintained by regularly evaluating the model performance and updating the model with newly collected data.

[0077] The LSTM neural network is used to predict the future vehicle usage status of users, providing data support for the OTA upgrade process, realizing the intelligent and personalized upgrade strategy, and improving upgrade efficiency and user satisfaction.

[0078] Furthermore, the three-dimensional matching model of user demand, network status and upgrade content established in step S4 includes the following three levels: user demand perception layer, network status monitoring layer and resource dynamic allocation layer.

[0079] Reference Figure 3 The user demand perception layer, located at the bottom of the model, is responsible for collecting real-time vehicle status data. In some implementations, real-time driving data is acquired via the CAN bus, with a sampling frequency of 10 Hz. The collected data includes SOC values, parking locations, and recent driving trajectories. A sliding window algorithm is used to extract dynamic features, with a window size of 30 minutes. This approach can capture short-term changes in user behavior, providing a more accurate data foundation for subsequent analysis.

[0080] The network status monitoring layer, located in the middle of the model, is responsible for monitoring network parameters. These parameters include seven network metrics, including base station load factor, channel quality index (CQI), and bit error rate. These parameters comprehensively reflect the current network status and provide a basis for resource allocation decisions.

[0081] The dynamic resource allocation layer is located at the top of the model and uses a modified Hungarian algorithm for multi-dimensional resource matching. This algorithm considers three dimensions: user demand, network status, and upgrade content to achieve optimal resource allocation. In some embodiments, the objective function of the algorithm is defined as: Among them, c ij represents the utility value of allocating resource j to user i, x ij It is a 0-1 variable, indicating whether allocation is performed. The algorithm searches for a resource allocation solution that maximizes the objective function through iterative optimization.

[0082] There is a close information exchange between the three layers. The user demand perception layer and the network status monitoring layer continuously provide real-time data to the dynamic resource allocation layer. The dynamic resource allocation layer makes decisions based on this data and feeds the results back to the other two layers, forming a closed-loop optimization mechanism.

[0083] In some implementations, the user demand perception layer employs deep learning models to analyze user behavior patterns. The network status monitoring layer utilizes time series prediction algorithms to estimate future network load. The dynamic resource allocation layer incorporates reinforcement learning methods to continuously optimize allocation strategies. This multi-layered, multi-algorithm collaborative mechanism improves the adaptability and accuracy of the three-dimensional matching model.

[0084] In some implementations, vehicle status data includes SOC value, parking location, and recent driving trajectory. This data is collected in real time through onboard sensors and communication modules, providing important reference for the OTA upgrade process.

[0085] The SOC value reflects the remaining charge of the vehicle's battery. In some implementations, the appropriate upgrade timing is selected based on the SOC value. When the SOC value is high, upgrades that consume more power are prioritized. When the SOC value is low, large upgrades are postponed, prioritizing critical security updates.

[0086] Parking location information is used to assess network environment and charging conditions. In some implementations, the optimal update window is selected based on parking location. For example, when a vehicle is parked near a home charging station, large upgrade tasks are prioritized to take advantage of a stable network connection and ample power supply. When a vehicle is parked in a public area, small security updates are prioritized to avoid prolonged use of public charging stations.

[0087] Recent driving trajectory data is used to analyze user habits and predict future travel. In some implementations, a user behavior model is constructed based on driving trajectory data. This model is used to predict the next time a user will use the vehicle, thereby selecting an appropriate upgrade window to minimize impact on normal user use.

[0088] In some implementations, a personalized OTA upgrade strategy is developed by comprehensively analyzing the SOC value, parking location, and recent driving trajectory. For example, for users who frequently drive long distances, the upgrade task can be immediately initiated upon detecting that the user has returned to a fixed parking spot and the SOC value is sufficient. For users who primarily use the device for short commutes, the upgrade can be performed on weekday evenings to ensure that it does not affect the user's use the next day.

[0089] The frequency and accuracy of vehicle status data collection are crucial for OTA upgrade optimization. In some implementations, the SOC value is updated every minute with an accuracy of 0.1%. Parking location is acquired via a GPS module with an accuracy of within 10 meters. Recent driving trajectory records driving data for the past seven days, including start and end times, mileage, average speed, and other information.

[0090] Through in-depth analysis and utilization of vehicle status data, OTA upgrades have made the upgrade process intelligent and personalized, improved the upgrade success rate, reduced the impact on users' normal use, and ensured the timely delivery of key security updates.

[0091] In some implementations, the upgrade method further includes: selecting an upgrade window based on the user behavior prediction result.

[0092] Specifically, the LSTM neural network's prediction results are used to optimize the selection of upgrade timing. In some embodiments, the LSTM neural network model predicts the user's vehicle usage status for the next 8 hours, including information such as whether the vehicle is in motion, its expected parking location, and charging requirements.

[0093] In some implementations, the order in which updates are pushed is adjusted based on predicted vehicle usage needs. Updates related to features the user is likely to use soon are prioritized to improve the user experience. For example, if the prediction indicates that the user is likely to make a long drive within the next two hours, updates related to the navigation system are prioritized.

[0094] Based on predicted vehicle location information, network resources are allocated rationally to avoid network congestion caused by multiple vehicles undergoing large-scale upgrades simultaneously in the same area. In some implementations, upgrades are staggered for vehicles predicted to be parked in the same area for extended periods of time to ensure balanced use of network resources.

[0095] Dynamically adjust the TCP window size to achieve bandwidth allocation. In some implementations, the TCP window size is dynamically adjusted based on the predicted vehicle usage and network conditions. When the vehicle is predicted to be parked for an extended period and network conditions are good, the TCP window size is increased to improve data transmission efficiency. When the vehicle is predicted to be in imminent use or network conditions are unstable, the TCP window size is decreased to reduce data transmission error rates.

[0096] In some implementations, the TCP window size adjustment formula is: W = W0·(1+α·Y+β·Q); Where W is the adjusted TCP window size, W0 is the baseline window size, Y is the predicted vehicle stationary time ratio, Q is the network quality score, and α and β are adjustment coefficients.

[0097] In some implementations, the priority of upgrade tasks is dynamically adjusted based on the prediction results. For vehicles predicted to be in use for extended periods, the priority of safety-related updates is increased to ensure timely updates of critical features. For vehicles predicted to be parked for extended periods, the priority of major feature updates is appropriately increased to fully utilize idle time for upgrades.

[0098] By selecting the upgrade window based on user behavior prediction results, the OTA upgrade process is made intelligent and personalized, which improves upgrade efficiency and user satisfaction while reducing the impact on users' normal use.

[0099] The embodiment of the present invention also discloses an OTA hierarchical upgrade system based on network resource optimization.

[0100] Reference Figure 4 The OTA hierarchical upgrade system based on network resource optimization includes an upgrade willingness evaluation module 1, a network resource classification module 2, a bandwidth allocation module 3, a three-dimensional matching module 4 and an upgrade execution module 5.

[0101] Upgrade Willingness Assessment Module 1 is used to construct a dynamic assessment model for user upgrade willingness. This module collects vehicle usage data, such as average daily mileage, charging behavior, interactive behavior, and environmental factors, to establish an upgrade willingness index calculation model. Based on this collected vehicle usage data and the established calculation model, Upgrade Willingness Assessment Module 1 dynamically calculates the user upgrade willingness index, thereby accurately assessing user upgrade needs.

[0102] Network Resource Classification Module 2 establishes a hierarchical network resource allocation mechanism. This module divides upgrade tasks into three transmission channels: security emergency response, function optimization, and system maintenance, and sets an initial bandwidth allocation for each channel. This hierarchical mechanism enables the system to more effectively manage and allocate network resources.

[0103] Bandwidth Allocation Module 3 employs an intelligent dynamic bandwidth allocation algorithm. This module monitors real-time network quality, calculates the priority score for each vehicle to be upgraded, and dynamically adjusts the transmission rate of each channel based on this information. Bandwidth Allocation Module 3 also implements a dynamic coupling model between upgrade tasks and network status, enabling the system to flexibly adjust based on real-time network conditions and the characteristics of the upgrade task.

[0104] The three-dimensional matching module 4 establishes a three-dimensional matching model based on user needs, network status, and upgrade content. This module includes a user needs perception layer, a network status monitoring layer, and a dynamic resource allocation layer. The user needs perception layer collects vehicle status data in real time, the network status monitoring layer monitors network parameters, and the dynamic resource allocation layer uses an improved Hungarian algorithm for multi-dimensional resource matching.

[0105] Upgrade Execution Module 5 performs OTA updates based on a three-dimensional matching model. This module implements a smooth resource transition technology for vehicle-side hardware, minimizing the impact of the upgrade process on normal vehicle operation. It also implements a Service Level Agreement (SLA) guarantee mechanism for emergency updates, prioritizing critical security updates.

[0106] Through the collaborative work of these modules, the system achieves accurate identification of user upgrade needs, efficient use of network resources, and intelligent scheduling of upgrade tasks, thereby improving the success rate of OTA upgrades and user satisfaction.

[0107] The embodiment of the present invention also discloses a readable storage medium.

[0108] A readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the OTA hierarchical upgrade method based on network resource optimization described in any of the above embodiments. The computer-readable storage medium may include: any entity or device capable of carrying a computer program, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), and a software distribution medium, etc. The computer program includes a computer program code. The computer program code may be in source code form, object code form, an executable file, or some intermediate form, etc. The computer-readable storage medium may include: any entity or device capable of carrying a computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), and a software distribution medium, etc.

[0109] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may be performed out of the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present invention pertain.

[0110] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, apparatus or device (such as a computer-based system, a system including a processing module, or other system that can fetch instructions from an instruction execution system, apparatus or device and execute instructions), or used in conjunction with such instruction execution systems, apparatuses or devices.

[0111] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. An OTA hierarchical upgrade method based on network resource optimization, characterized in that: include: Build a dynamic evaluation model for user upgrade willingness; Establish a hierarchical network resource occupation mechanism; Adopt intelligent bandwidth dynamic allocation algorithm; Establish a three-dimensional matching model of user needs, network status and upgrade content for OTA upgrades.

2. The upgrade method according to claim 1, characterized in that: The construction of a dynamic evaluation model for user upgrade willingness includes: Collect vehicle usage data, including average daily mileage, charging behavior, interactive behavior, and environmental factors; and establish an upgrade willingness index calculation model; A user upgrade willingness index is calculated based on the vehicle usage data and the upgrade willingness index calculation model.

3. The upgrade method according to claim 2, characterized in that: The calculation model of the upgrade willingness index is: Among them, W u is the upgrade willingness index, αi is the weight coefficient of each vehicle usage data, Xi is the actual value of each vehicle usage data, μi is the mean of each vehicle usage data, σi is the standard deviation of each vehicle usage data, Sc is the system urgency coefficient, and β is the safety weight factor.

4. The upgrade method according to claim 1, characterized in that: The establishment of a hierarchical network resource occupation mechanism includes: dividing the upgrade task into three transmission channels: security emergency, function optimization and system maintenance.

5. The upgrade method according to claim 4, characterized in that: Also includes: Initial bandwidth ratios are set for the three transmission channels of security emergency, function optimization and system maintenance respectively.

6. The upgrade method according to claim 1, characterized in that: The intelligent dynamic bandwidth allocation algorithm includes: monitoring real-time network quality; Calculate the priority score of each vehicle to be upgraded; Based on the real-time network quality and the limit score of each vehicle, the transmission rate of each transmission channel is dynamically adjusted.

7. The upgrade method according to claim 6, characterized in that: The priority score is the product of the user's upgrade willingness and a preset service quality coefficient.

8. The upgrading method according to claim 1, characterized in that: Also includes: Predict the future user vehicle usage status through LSTM neural network.

9. The upgrading method according to claim 8, characterized in that: The method of predicting the future vehicle usage status of a user through the LSTM neural network includes: Collect historical vehicle usage data; Train the LSTM neural network model; The trained LSTM neural network model is used to predict the user's vehicle usage status within a preset time period in the future.

10. The upgrade method according to claim 1, characterized in that: The establishment of a three-dimensional matching model of user needs, network status, and upgrade content includes: Establish a user demand perception layer to collect vehicle status data in real time; Establish a network status monitoring layer to monitor network parameters; A dynamic resource allocation layer is established, and an improved Hungarian algorithm is used for multi-dimensional resource matching.

11. The upgrade method according to claim 10, characterized in that: The vehicle status data includes SOC value, parking location and recent driving trajectory.

12. The upgrade method according to claim 1, characterized in that: Also includes: Select an upgrade window based on user behavior prediction results.

13. An OTA hierarchical upgrade system based on network resource optimization, characterized in that: The system comprises: User upgrade willingness evaluation module, used to build a dynamic evaluation model for user upgrade willingness; Network resource classification module, used to establish a network resource classification occupation mechanism; Bandwidth allocation module, used to adopt intelligent bandwidth dynamic allocation algorithm; A three-dimensional matching module, used to establish a three-dimensional matching model of user needs, network status and upgrade content; and An upgrade execution module is used to execute OTA upgrade based on the three-dimensional matching model.

14. A readable storage medium, characterized in that The readable storage medium stores computer instructions, and when the computer instructions are executed by a processor, the upgrading method according to any one of claims 1 to 12 is implemented.

Images