A wireless resource connection control method, device and electronic equipment

Abstract

The application discloses a wireless resource connection control method and device and electronic equipment. The method comprises the following steps: obtaining a target parameter in response to a target event; the target parameter represents the interaction state between the electronic equipment and the network; processing the target parameter based on a target model to obtain a target result; the target result represents whether the wireless resource control connection between the electronic equipment and the network is abnormal; if the target result represents abnormality and a preset termination condition is met, the wireless resource control connection between the electronic equipment and the network is interrupted.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to a wireless resource connection control method, apparatus and electronic device. Background Technology

[0002] Given the current highly complex and diverse network environment, electronic devices may experience abnormal power consumption under certain network conditions, affecting the user experience. Summary of the Invention

[0003] The technical solution provided in this application is as follows:

[0004] The first aspect of this application provides a wireless resource connection control method, comprising:

[0005] In response to a target event, target parameters are obtained; these target parameters characterize the interaction state between the electronic device and the network.

[0006] Based on the target model, the target parameters are processed to obtain the target result; the target result characterizes whether the wireless resource control connection between the electronic device and the network is abnormal.

[0007] If the target result is characterized as abnormal and a preset termination condition is met, the wireless resource control connection between the electronic device and the network is interrupted.

[0008] In one possible implementation, obtaining the target parameter includes at least one of the following:

[0009] Obtain a first type of parameter; the first type of parameter represents the correlation between user operation and the power consumption state of the electronic device;

[0010] Obtain a second type of parameter; the second type of parameter is used to characterize changes in the resource configuration of the wireless resource control connection between the electronic device and the network.

[0011] Obtain a third type of parameter; the third type of parameter is used to characterize the dynamic attributes of the radio resource control connection.

[0012] In one possible implementation, the first type parameter includes at least one of the following:

[0013] The first parameter, the second parameter, and the third parameter; the first parameter represents the continuous duration for which the screen of the electronic device remains off; the second parameter represents the total amount of data transmitted by the electronic device within a specified time period; the third parameter represents whether the electronic device is in a call state.

[0014] The second type of parameter includes: a fourth parameter, which characterizes the number of times the carrier aggregation configuration changes per unit time;

[0015] The third type of parameter includes: a fifth parameter, which characterizes the total duration for which the electronic device maintains the wireless resource control connection with the network.

[0016] In one possible implementation, the wireless resource connection control method further includes:

[0017] Model configuration information is obtained from the first storage space of the electronic device; the model configuration information is used to describe the structure of the target model and the reasoning behavior of the target model;

[0018] Configure the target model based on the model configuration information.

[0019] In one possible implementation, the wireless resource connection control method further includes:

[0020] If the model configuration information is not obtained from the first storage space, historical parameters are obtained from the second storage space of the electronic device; the historical parameters represent the historical interaction status between the electronic device and the network.

[0021] The historical parameters are input into the target model, which processes the historical parameters to obtain a prediction result; the prediction result indicates whether the radio resource control connection corresponding to the historical parameters is abnormal.

[0022] Based on the loss value, the parameters of the target model are updated; the loss value represents the difference between the prediction result and the labeled result corresponding to the historical parameters; the labeled result represents the actual conclusion of whether the radio resource control connection corresponding to the historical parameters is abnormal; the status label represents the actual status of the radio resource control connection between the electronic device and the network.

[0023] In one possible implementation, before inputting the historical parameters into the target model and processing the historical parameters by the target model to obtain the prediction result, the method further includes:

[0024] Based on the historical parameters, a target structural description is selected from a variety of pre-stored structural description information; the target structural description information is used to describe the structure of the target model.

[0025] Based on the target structure description information, a target model is constructed.

[0026] In one possible implementation, configuring the target model based on the model configuration information includes:

[0027] Based on the model configuration information, the target model is configured in the baseband processor; the target model is used in the baseband processor to determine whether the radio resource control connection between the electronic device and the network is abnormal.

[0028] The control interruption of the wireless resource control connection between the electronic device and the network includes any of the following:

[0029] The control baseband processor locally releases the wireless resource control connection between the electronic device and the network;

[0030] The control baseband processor generates a measurement report and sends the measurement report to the network; the measurement report is used to trigger the network to initiate a cell handover process.

[0031] In one possible implementation, the preset termination condition includes at least one of the following:

[0032] The duration for which the screen of the electronic device remains off meets a set time length threshold.

[0033] The electronic device is in a non-call state.

[0034] Another aspect of this application provides a wireless resource connection control device, comprising:

[0035] The acquisition module is used to acquire target parameters in response to a target event; the target parameters characterize the interaction state between the electronic device and the network.

[0036] The processing module is used to process the target parameters based on the target model to obtain the target result; the target result characterizes whether the wireless resource control connection between the electronic device and the network is abnormal;

[0037] An interrupt module is used to control the interruption of the wireless resource control connection between the electronic device and the network if the target result is characterized as abnormal and a preset termination condition is met.

[0038] In a third aspect of this application, an electronic device is provided, comprising:

[0039] Memory, used to store model configuration information or historical parameters;

[0040] Baseband processor, used for:

[0041] In response to a target event, the model configuration information or the historical parameters are read from the memory;

[0042] The target model is obtained based on the model configuration information or the historical parameters;

[0043] Obtain target parameters; the target parameters characterize the interaction state between the electronic device and the network;

[0044] Based on the target model, the target parameters are processed to obtain the target result; the target result characterizes whether the wireless resource control connection between the electronic device and the network is abnormal;

[0045] If the target result is characterized as abnormal and a preset termination condition is met, the wireless resource control connection between the electronic device and the network is interrupted. Attached Figure Description

[0046] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and the originals and elements are not necessarily drawn to scale.

[0047] Figure 1 This is a flowchart illustrating a wireless resource connection control method provided in Embodiment 1 of this application;

[0048] Figure 2 This is a flowchart illustrating a wireless resource connection control method provided in Embodiment 4 of this application;

[0049] Figure 3 This is a flowchart illustrating a wireless resource connection control method provided in Embodiment 5 of this application;

[0050] Figure 4 This is a flowchart illustrating a wireless resource connection control method provided in Embodiment 6 of this application;

[0051] Figure 5 A schematic diagram of the structure of a wireless resource connection control device provided in this application;

[0052] Figure 6 This is a schematic diagram of the structure of an electronic device provided in this application. Detailed Implementation

[0053] The embodiments of this application are described below with reference to the accompanying drawings. The terminology used in the implementation section of this application is for explaining specific embodiments only and is not intended to limit the scope of this application.

[0054] The embodiments of this application will now be described with reference to the accompanying drawings. Those skilled in the art will recognize that, with technological advancements and the emergence of new scenarios, the technical solutions provided in the embodiments of this application are equally applicable to similar technical problems.

[0055] The terms "first," "second," etc., used in this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such terms can be used interchangeably where appropriate; this is merely a way of distinguishing objects with the same attributes in the embodiments of this application. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion, so that a process, method, system, product, or apparatus that comprises a series of units is not necessarily limited to those units, but may include other units not explicitly listed or inherent to those processes, methods, products, or apparatuses.

[0056] Reference Figure 1 This is a flowchart illustrating a wireless resource connection control method provided in Embodiment 1 of this application. Figure 1 As shown, the method may include, but is not limited to, the following steps:

[0057] Step S101: In response to the target event, obtain the target parameters; the target parameters characterize the interaction state between the electronic device and the network.

[0058] In this embodiment, the target event may include, but is not limited to, at least one of the following:

[0059] Electronic devices switch from screen-on state to screen-off state (detected by the device's power management module, such as the screen-off broadcast sent by the Android system's PowerManager service).

[0060] The LTE RRC connection is established for the first time after the screen is turned off.

[0061] The LTE RRC connection continues for more than a preset time (e.g., 1 minute) while the screen is off.

[0062] When the screen is on, there is no user touch operation for a preset duration (e.g., 5 minutes).

[0063] In today's network environment, the stable operation of network devices is a crucial foundation for ensuring normal communication and reasonable power consumption of electronic devices. However, in actual operation, network devices inevitably encounter various abnormal situations.

[0064] Base stations, as crucial nodes for electronic devices to access the network, require accurate configuration. Errors in base station configuration, such as improper parameter settings or incorrect frequency band allocation, directly interfere with the normal communication process between electronic devices and the base station. This interference prevents electronic devices from operating according to predetermined rules and logic when attempting to establish or maintain RRC (Radio Resource Control) connections.

[0065] As the core hub of the entire network communication, the core network is responsible for processing various signaling and service data. When a signaling processing failure occurs in the core network, such as signaling transmission delays or signaling parsing errors, it can lead to chaotic signaling interactions between electronic devices and the core network. The management of RRC connections is highly dependent on the signaling interactions between electronic devices and the network; chaotic signaling interactions can disrupt the established logic followed by RRC connection management.

[0066] Under normal circumstances, RRC connection management follows an operating mode that matches service demands with power consumption. When electronic devices have service demands, such as data downloading, voice calls, or video playback, the network establishes an RRC connection for the electronic device based on these demands. At this time, the electronic device operates in a high-power state to ensure smooth service transmission; this high power consumption is reasonable and necessary. When the service ends and the electronic device no longer has service demands, the RRC connection is released promptly, and the electronic device enters a low-power state to conserve power.

[0067] However, due to network equipment anomalies such as base station configuration errors and core network signaling processing failures, the RRC connection management logic is severely disrupted. Even when electronic devices no longer have actual service needs, the RRC connection cannot be released according to normal logic and remains in the RRC connection state. Maintaining this state requires the electronic devices to maintain a high power consumption level, which leads to abnormally high power consumption in electronic devices even without service needs, resulting in abnormal power consumption and seriously affecting the battery life and user experience of electronic devices.

[0068] To accurately identify abnormal RRC connections, electronic devices can determine whether the maintenance of the RRC connection state is supported by corresponding service requirements. In normal communication scenarios, the high power consumption of electronic devices in the RRC connection state is necessary to ensure the smooth transmission of various services such as data downloads and voice calls, and is a reasonable and necessary energy expenditure. However, in abnormal scenarios, the RRC connection state persists without corresponding service requirements; in this case, the high power consumption is ineffective energy loss, directly causing abnormal power consumption of the electronic device.

[0069] The target parameters can characterize the interaction state between electronic devices and the network, and can indirectly map the actual existence of business needs. For example, when there are business needs, the interaction state will exhibit regularity that matches the business transmission (such as continuous data interaction and timely signaling response); when there are no business needs, the interaction will gradually terminate in normal scenarios, while in abnormal scenarios it will exhibit the abnormal characteristic of continuous connection without data interaction.

[0070] Therefore, acquiring target parameters from electronic devices can provide crucial data support for determining RRC connection anomalies, ensuring that RRC connection anomalies caused by network device malfunctions can be specifically located, thus addressing the root cause of abnormal power consumption in electronic devices.

[0071] In this embodiment, the target parameters can be format adapted by converting them into floating-point format and mapping them to the [0,1] range through numerical normalization. This avoids the impact of numerical range differences on the recognition accuracy of the subsequent target model and ensures that the parameters can be effectively processed by the target model.

[0072] Step S102: Based on the target model, process the target parameters to obtain the target result; the target result characterizes whether the wireless resource control connection between the electronic device and the network is abnormal.

[0073] In this embodiment, during the training phase, data samples can be collected, which can cover various normal and abnormal scenarios in the interaction between electronic devices and the network. Normal scenario sample data can include various parameter information when electronic devices establish, maintain, and release RRC connections with the network under different service requirements (such as routine business operations like data downloading, voice calls, and video playback). These parameters reflect the dynamic changes in the RRC connection state driven by normal service requirements, including connection establishment duration, signaling interaction sequence and frequency, data transmission rate, and stability.

[0074] The sample data for abnormal scenarios can cover a variety of RRC connection anomalies caused by network equipment failures or abnormal conditions. For example, problems such as unreasonable parameter settings or frequency band allocation errors caused by base station configuration errors can cause electronic devices to malfunction when establishing or maintaining RRC connections, resulting in the inability to establish connections normally, frequent disconnections, or continuous maintenance in a non-service-required state. Signaling transmission delays and signaling parsing errors caused by core network signaling processing failures can disrupt the established logic followed by RRC connection management, causing the connection state to mismatch with service requirements, resulting in a high-power abnormal state where the connection is continuously maintained even when there is no service requirement.

[0075] By performing deep learning and analysis on a large amount of sample data from both normal and abnormal scenarios, the target model can extract and mine key distinguishing features between normal and abnormal RRC connection states. The target model can not only identify simple and obvious abnormal patterns, such as sudden connection interruptions or prolonged failures to establish a connection, but also accurately distinguish complex and more subtle anomalies. For example, for some seemingly normal connection states that are actually non-service-driven connection maintenance caused by network device malfunctions, the target model can accurately identify this anomaly by analyzing subtle changes and potential correlations in target parameters, and differentiate it from normal service-demand-driven state changes.

[0076] The fully trained target model possesses powerful pattern recognition and classification capabilities. When target parameters collected from real-world scenarios are input, it can quickly and accurately process these parameters, determine whether the RRC connection between electronic devices and networks is in an abnormal state based on the learned differential features and abnormal patterns, and output the corresponding target results.

[0077] Step S103: If the target result is characterized as abnormal and a preset termination condition is met, the wireless resource control connection between the electronic device and the network is interrupted.

[0078] During the interaction between electronic devices and the network, in the event of an abnormal scenario, even if the RRC connection is maintained, there may be no corresponding service demand to support it. At this time, the electronic device continues to operate in a high-power state. This ineffective high power consumption will rapidly and significantly shorten the device's battery life. For users, frequent charging needs and the inability to use the device normally due to insufficient power will greatly affect the user experience, reducing user satisfaction and trust in the device.

[0079] Disconnecting this abnormal RRC connection allows electronic devices to break free from high power consumption and re-enter a low-power state. In this state, the device's energy consumption is significantly reduced, effectively mitigating the abnormal power consumption caused by the abnormal RRC connection. This significantly extends the device's battery life, allowing users to use the device for longer periods without frequently searching for charging ports, thus improving user convenience and comfort.

[0080] In this embodiment, the preset termination conditions can be set according to various factors such as different application scenarios, device types, user needs, and network environments, and no restrictions are imposed in this application.

[0081] In this embodiment, target parameters representing the interaction status between electronic devices and the network are obtained by responding to target events. The target parameters are processed using a target model that can accurately distinguish between normal and abnormal RRC connection states to determine whether the connection is abnormal. If it is abnormal and meets the preset termination conditions, the connection is interrupted. This can avoid high power consumption and invalid energy loss caused by network device abnormalities, which leads to the continuous maintenance of RRC connections without service requirements. This solves the problem of abnormal power consumption of electronic devices from the root, extends battery life, and improves the convenience and comfort of users using the devices.

[0082] As another optional embodiment of this application, a wireless resource connection control method is provided in Embodiment 2 of this application. This embodiment is mainly an implementation method for obtaining target parameters in Embodiment 1, and may specifically include, but is not limited to, at least one of the following:

[0083] Step S11: Obtain a first type of parameter; the first type of parameter represents the correlation between user operation and the power consumption state of the electronic device.

[0084] In this embodiment, we can start from the actual usage scenarios on the electronic device terminal side to capture the linkage between business needs caused by user operations and device power consumption.

[0085] User operations are the core source of business needs, while the power consumption status of electronic devices is a direct external manifestation of the RRC connection status. The relationship between the two can directly reflect whether the current power consumption of the device is driven by real business needs triggered by user operations.

[0086] By obtaining the first type of parameters, it is possible to accurately determine whether the electronic device currently has a service demand that is actively or passively triggered by the user, and whether the power consumption corresponding to the service demand is reasonable.

[0087] Step S12: Obtain the second type of parameter; the second type of parameter is used to characterize the resource configuration change of the wireless resource control connection between the electronic device and the network.

[0088] In this embodiment, the resource configuration of the RRC connection is dynamically adjusted by the network side according to the service requirements of the electronic devices. Under normal circumstances, the changes in resource configuration are adapted to the strength and type of service requirements, and there is a clear service-driven logic. However, when network devices are abnormal, such as base station configuration errors or core network signaling processing failures, the resource scheduling logic of the network side will be disrupted, resulting in disordered and frequent changes in resource configuration without service drivers.

[0089] By acquiring the second type of parameters, the operating status of network-side resource configuration can be accurately captured, and it can be determined whether the resource interaction between electronic devices and the network conforms to normal business-driven logic. From the network side, it can be identified whether the abnormality of RRC connection management logic is caused by network device abnormality.

[0090] Step S13: Obtain the third type of parameter; the third type of parameter is used to characterize the dynamic attributes of the radio resource control connection.

[0091] In this embodiment, the dynamic attributes of the RRC connection are highly correlated with business requirements. Under normal circumstances, the connection duration and state switching change regularly with the generation and termination of business requirements. However, under abnormal circumstances, dynamic characteristics that are disconnected from business requirements will appear, that is, the connection is maintained continuously even when there are no business requirements.

[0092] By obtaining the third type of parameters, it is possible to directly capture whether there are abnormal characteristics in the operation of the RRC connection itself. This is a direct basis for judging whether the connection is in an abnormal state of having no service but continuing to maintain it.

[0093] In this embodiment, by acquiring at least one of the first type of parameters, the second type of parameters, and the third type of parameters, the interaction state between the electronic device and the network can be specifically characterized from three different core dimensions: terminal-side business demand-driven, network-side resource scheduling and management, and the operating state of the RRC connection itself. Each type of parameter has its own emphasis and can be mutually verified, which not only avoids the one-sidedness of a single-dimensional parameter in representing the interaction state, but also allows the collected target parameters to more accurately map the real characteristics of the RRC connection operation and the matching relationship with business needs, providing more targeted and comprehensive core data support for subsequent RRC connection anomaly judgment.

[0094] In this embodiment, the parameter types to be collected can be selected as needed, eliminating the need for indiscriminate full collection of all three types of parameters. This reduces unnecessary collection overhead while ensuring data validity, thereby improving the flexibility and efficiency of parameter collection.

[0095] As another optional embodiment of this application, a wireless resource connection control method is provided in Embodiment 3 of this application. This embodiment is mainly an implementation of the first type parameter, the second type parameter, and the third type parameter in Embodiment 2. The first type parameter may specifically include, but is not limited to, at least one of the following:

[0096] First parameter, second parameter, and second parameter.

[0097] The first parameter characterizes the length of time the screen of the electronic device remains in a closed state.

[0098] In this embodiment, the duration for which the screen of an electronic device remains off directly reflects whether the user has a need to actively operate the device, thereby helping to determine whether the device is currently in a high-power consumption anomaly risk scenario without active service needs. Generally, the longer the screen remains off, the higher the probability that the user has no active service needs. If the RRC connection is still maintained during this time, the probability of an anomaly also increases.

[0099] In this embodiment, the specific value of the continuous screen-off duration can be obtained by detecting the on / off state of the screen in real time and recording the state switching timestamp, and by calculating the time difference between when the screen switches from on to off to the current time.

[0100] The second parameter represents the total amount of data transmitted by the electronic device within a specified time period.

[0101] When there is actual data service demand, data traffic will increase continuously or in stages as service transmission occurs; when there is no data service demand, data traffic will remain at zero or extremely low values. The second parameter can be used to distinguish between normal connections with data transmission and abnormal connections without data transmission.

[0102] In this embodiment, the total data traffic within a specified time window (e.g., 1 minute, 5 minutes, which can be configured as needed) can be obtained by summing up the number of uplink and downlink data transmission bytes of the baseband processor's LTE protocol layer. This supports quantization in multiple units such as bytes and kilobytes.

[0103] The third parameter indicates whether the electronic device is in a call state.

[0104] In this embodiment, voice calls, as a basic communication service, require the RRC connection to remain continuously connected throughout the entire process. The network side allocates dedicated signaling and data bearers for voice calls. Even without additional data transmission, maintaining the RRC connection is normal logic driven by service needs, and the corresponding high power consumption is necessary to ensure continuous and smooth calls. The RRC connection state in this scenario is not abnormal. Conversely, if the electronic device is not in a call state but has no other data service needs and the RRC connection remains continuously maintained, then this connection state meets the prerequisites for anomaly detection.

[0105] Based on the third parameter, normal RRC connections in voice call scenarios can be avoided from being misjudged as abnormal. Defining the boundary for RRC connection anomaly judgment from the perspective of business scenarios is an important indicator to ensure the accuracy of anomaly judgment and eliminate interference from non-abnormal scenarios.

[0106] In this embodiment, the current call status (such as "calling" or "no call") of the device can be directly obtained by detecting the call status flag bit in the CS domain (circuit domain, traditional voice call) / IMS domain (IP multimedia subsystem, VoLTE / VoNR voice call) of the baseband processor. The flag bit status is updated in real time as the call is established, hung up, or the network is switched, ensuring the real-time performance and accuracy of the call status acquisition.

[0107] The second type of parameter may include, but is not limited to, a fourth parameter, which characterizes the number of times the carrier aggregation configuration changes per unit time.

[0108] In this embodiment, carrier aggregation (CA) is a core resource configuration strategy for improving data transmission rates in LTE (Long Term Evolution) networks. Under normal circumstances, the network side will dynamically add and delete CA carriers according to the actual data service needs of electronic devices (such as high-speed downloads, high-definition video playback, and other high bandwidth requirements). The frequency of its configuration changes is highly adapted to the strength and type of service needs, and there is a clear service-driven logic.

[0109] When network equipment malfunctions, such as incorrect base station configuration or core network signaling processing failure, the resource scheduling logic on the network side will be disrupted. This manifests as disordered and frequent changes in CA configuration without service drivers. Such invalid changes will directly lead to abnormal RRC connection management logic, which in turn will cause abnormal power consumption of the equipment.

[0110] The fourth parameter provides crucial network-side data support for identifying RRC connection anomalies caused by network device malfunctions.

[0111] In this embodiment, the total frequency of LTE CA addition and deletion can be obtained by counting the number of CA carrier addition and deletion instructions received by the baseband processor from the network side in CA configuration signaling (such as RRCConnectionReconfiguration) within a specified unit time (e.g., 1 minute), and then summing them up. This supports quantization of multiple time dimensions such as times / minute and times / 5 minutes.

[0112] The third type of parameter may include, but is not limited to, a fifth parameter, which characterizes the total duration for which the electronic device maintains the wireless resource control connection with the network.

[0113] In this embodiment, under normal circumstances, the total duration of maintaining the RRC connection is highly adapted to the duration of service requirements. When there are service requirements, the total duration of maintaining the RRC connection is extended as the service is carried out. After the service is terminated, the connection will be released in a timely manner. At this time, the value of the fifth parameter will stop increasing and be cleared to zero when the connection is released. Under abnormal circumstances, due to network equipment abnormalities such as base station configuration errors or core network signaling processing failures, the RRC connection cannot be released normally after the service is terminated. At this time, the value of the fifth parameter will continue to increase. Even if the electronic device has no service requirements, the value will continue to accumulate, showing the abnormal characteristic of continuous growth without service support.

[0114] The fifth parameter clearly indicates whether the RRC connection is in a long-term maintenance state. Combined with the first type of parameters (screen-off duration, data traffic, and call status), it is possible to accurately distinguish between long-term normal connections with service support and long-term abnormal connections without service support. For example, if the fifth parameter shows that the RRC connection duration exceeds 5 minutes, but the first type of parameters show 0 data traffic, screen-off duration exceeding 5 minutes, and no call status, then the "long-term maintenance" characteristic of the fifth parameter, combined with the "no service" characteristic of other parameters, can be used to jointly determine that the RRC connection is abnormal. Conversely, if the fifth parameter shows that the connection duration exceeds 5 minutes, but the data traffic continues to increase (data service is present) or the connection is in a call status (voice service is present), then the duration can be determined to be normal, avoiding misjudgment.

[0115] In this embodiment, the baseband processor can detect the state transitions of the RRC connection in real time (such as switching from the RRC_IDLE idle state to the RRC_CONNECTED connected state, and from the RRC_CONNECTED connected state to the released state), and accurately record the timestamp of the RRC connection establishment. By calculating the time difference between the current moment and the connection establishment timestamp in real time, the specific value of the LTE RRC duration is obtained. In this embodiment, the value of the fifth parameter is collected with second-level precision, and the value is updated in real time as the RRC connection continues. When the RRC connection is released, the value of this parameter is cleared to zero, and it starts accumulating again after the next connection is established.

[0116] In this embodiment, the first parameter (screen off duration), the second parameter (total data volume), and the third parameter (call status) accurately capture the actual existence of service needs from multiple scenarios on the terminal side, and can directly distinguish between scenarios with user operation / service needs and scenarios without user operation / service needs, providing a direct basis for determining whether the RRC connection has reasonable service support; the fourth parameter (LTE CA frequency addition / deletion) accurately captures the operating logic of resource configuration from the network side, and can effectively identify service-less driven resource configuration anomalies caused by network device anomalies, and locate the network-side root cause of abnormal RRC connections; the fifth parameter (LTE RRC duration) quantifies its lifecycle characteristics from the perspective of the connection itself, and can directly capture the core abnormal phenomenon of maintaining a connection without service.

[0117] The specific parameters can work together and complement each other for verification, accurately quantifying and organically combining the characteristics of three dimensions: terminal-side business needs, network-side resource scheduling, and the operation of the RRC connection itself. This comprehensively and accurately restores the real operating state of the RRC connection, clearly distinguishes between normal connections driven by business and abnormal connections caused by network device anomalies, and provides high-quality and highly targeted underlying data support for subsequent anomaly judgment by the target model, greatly improving the accuracy and robustness of model anomaly identification.

[0118] As another optional embodiment of this application, refer to Figure 2 This is a flowchart illustrating a wireless resource connection control method provided in Embodiment 4 of this application. Figure 2 As shown, the method may include, but is not limited to, the following steps:

[0119] Step S201: Obtain model configuration information from the first storage space of the electronic device; the model configuration information is used to describe the structure of the target model and the reasoning behavior of the target model.

[0120] In this embodiment, the first storage space is a storage area accessible to the electronic device, and is not limited to the local storage of the electronic device. It includes, but is not limited to, local storage areas such as the device's built-in flash memory, solid-state drive, and system partition storage, and may also include remote storage areas such as cloud storage servers and distributed storage nodes. The storage location can be flexibly selected according to the actual application scenario (such as device computing power and network stability). It can achieve the advantages of fast reading and low latency of local storage, as well as the advantages of flexible updates and multi-device synchronization of cloud storage, ensuring the stable acquisition and efficient retrieval of model configuration information, and adapting to the deployment needs of different terminal devices and network environments.

[0121] In this embodiment, the model configuration information may include, but is not limited to, the core reasoning behavior obtained after training the target model, namely, the RRC connection anomaly judgment logic and parameter feature matching rules mined during the training process, which are the core foundation for the model to achieve anomaly judgment.

[0122] The rule information for adapting to the deployment and inference of electronic devices (such as rule information related to model structure and rule information related to model inference behavior) can be used to clarify the specific implementation method, operation adaptation strategy and execution process of the trained core inference logic on the electronic device, define how the core inference logic is adapted to the hardware resources (such as processors and storage modules), system environment and computing power allocation mechanism of the electronic device, and how it is transformed into running instructions that the electronic device can recognize and execute. This ensures that the trained core inference logic can be executed stably and efficiently on the electronic device and is adapted to the hardware computing power level of the electronic device and the actual detection application requirements.

[0123] The model structure-related rule information may include, but is not limited to: the network type of the target model (such as BP neural network, lightweight CNN, random forest, etc.), the number of network layers, the number of neurons, the feature extraction dimension, the definition of input and output nodes, the connection method of each layer of the network, etc. At the same time, it is associated with the structure adaptation logic determined after training, which is used to accurately define the core architecture of the target model. This ensures that the electronic device can accurately restore the trained target model structure and the structure association rules determined after training based on the configuration information, so as to avoid model deployment failure, inference logic distortion or judgment bias due to ambiguity in structure definition or deviation in restoration of training results.

[0124] The rules related to model inference behavior may include, but are not limited to: parameter input format (e.g., whether floating-point format is required, whether normalization to the [0,1] range is required), result output format (e.g., binary identifier, quantized value), and inference computing power allocation strategy (e.g., allocation to application processor or baseband processor), etc., to standardize the target model's processing flow for target parameters, and ensure that when the target model performs inference on the electronic device, it can adapt to the terminal hardware computing power, efficiently process target parameters, and output accurate RRC connection anomaly judgment results.

[0125] Step S202: Configure the target model based on the model configuration information.

[0126] In this embodiment, based on the model structure-related rule information in the model configuration information, a network architecture that perfectly matches the trained target model can be built in the electronic device. This accurately restores the core structural features of the model, such as network type, number of layers, number of neurons, and input / output nodes. At the same time, the structure adaptation logic determined after training is loaded to ensure the accurate reproduction of the model structure on the hardware side. This provides a basic framework for the execution of the core inference logic and avoids distortion of the inference logic due to structural deviations.

[0127] Furthermore, based on the model inference behavior-related rule information in the model configuration information, the inference process and operation strategy of the completed model architecture are standardized and configured. This includes defining the input format of the target parameters according to preset requirements, embedding the core inference logic of RRC connection anomaly judgment obtained after training, setting the output format of the results, allocating dedicated inference computing power resources according to the hardware computing power of electronic devices, and deploying operation optimization rules adapted to the terminal. This ensures that the model's inference behavior strictly follows the core judgment logic after training and can adapt to the hardware environment of electronic devices, achieving efficient and low-power inference operations.

[0128] Step S203: In response to the target event, obtain the target parameters; the target parameters characterize the interaction state between the electronic device and the network.

[0129] Step S204: Based on the target model, process the target parameters to obtain the target result; the target result characterizes whether the wireless resource control connection between the electronic device and the network is abnormal.

[0130] Step S205: If the target result is characterized as abnormal and a preset termination condition is met, the wireless resource control connection between the electronic device and the network is interrupted.

[0131] For a detailed description of steps S203-S205, please refer to the relevant description of steps S101-S103 in Example 1, which will not be repeated here.

[0132] In this embodiment, the dynamic configuration method based on model configuration information can flexibly adjust the model architecture, inference logic, and operation strategy directly according to the content of the configuration information. When it is necessary to update the model inference logic, switch the target model type (such as switching from a BP neural network to a lightweight CNN), or adapt to new terminal hardware or detection scenarios, only the model configuration information in the first storage space needs to be updated. The electronic device can then reconfigure the model based on the new configuration information without modifying any underlying code or upgrading the system. This improves the deployment efficiency, iteration efficiency, and maintainability of the target model on the electronic device.

[0133] Meanwhile, this configuration method gives the target model a strong ability to adapt to different scenarios. The same wireless resource connection control scheme can be used to deploy adaptable models in electronic devices with different computing power levels and in different network environments by changing different model configuration information. Moreover, the model's inference logic can be iterated in real time as the configuration information is updated, ensuring that the model can always accurately match the latest RRC connection anomaly detection requirements. This allows the trained model results to be applied quickly and efficiently on the terminal side, laying a solid foundation for the accurate determination of RRC connection anomalies based on the target model in the future.

[0134] As another optional embodiment of this application, refer to Figure 3 This is a flowchart illustrating a wireless resource connection control method provided in Embodiment 5 of this application. Figure 3 As shown, the method may include, but is not limited to, the following steps:

[0135] Step S301: Obtain model configuration information from the first storage space of the electronic device; the model configuration information is used to describe the structure of the target model and the reasoning behavior of the target model;

[0136] Step S302: Configure the target model based on the model configuration information.

[0137] For a detailed description of steps S301-S302, please refer to the relevant description of steps S201-S202 in Example 4, which will not be repeated here.

[0138] Step S303: Determine that the model configuration information has not been obtained from the first storage space, and obtain historical parameters from the second storage space of the electronic device; the historical parameters represent the historical interaction status between the electronic device and the network.

[0139] In this embodiment, the second storage space may include, but is not limited to, at least one of the following: locally accessible storage of the electronic device (such as a designated partition of the built-in flash memory, a data cache area, a system underlying communication log storage area, etc.); remote storage areas such as cloud storage servers and distributed storage nodes, and the storage location can be flexibly selected according to the actual operating scenario of the electronic device (such as local storage capacity and network stability).

[0140] Locally accessible storage enables fast reading and low-latency retrieval of historical parameters without relying on external networks; cloud storage enables large-capacity storage of historical parameters and synchronization across multiple devices, preventing data loss due to local storage failure. The combination of the two ensures stable acquisition and efficient retrieval of historical parameters, further enhancing the robustness of the process.

[0141] In this embodiment, historical parameters can be continuously collected and stored during the network communication process of the electronic device, according to the collection rules, collection frequency, and quantization standards of the first to fifth parameters specified in Embodiments 2 and 3. The historical parameters can cover parameter data of two core scenarios: one is the normal network interaction scenario (such as user-initiated operation triggering service requirements, and RRC connection being established and released normally with service requirements), and the other is the abnormal interaction scenario of RRC connection caused by network device anomalies (such as base station configuration errors, core network signaling failures causing RRC connection to be maintained without service requirements).

[0142] Each set of historical parameters uniquely corresponds to a labeling result. The labeling result is used to clearly characterize whether the wireless resource control connection between the electronic device and the network was actually abnormal when the set of historical parameters was collected. The labeling results can be divided into two categories: one is normal labels (corresponding to normal scenarios of RRC connection supported by service needs, such as RRC connection during a call or RRC connection during data download), and the other is abnormal labels (corresponding to abnormal scenarios of RRC connection being maintained without service needs, such as maintaining RRC connection even when the screen is off and there is no data transmission).

[0143] In this embodiment, by retrieving the historical parameters with the above-mentioned labeled results from the second storage space (local or cloud), standardized and labeled basic training samples are provided for the local update of the target model, which can complete iterative adaptation and ensure that the target model has the ability to determine RRC connection anomalies adapted to the local network environment.

[0144] Step S304: Input the historical parameters into the target model, and the target model processes the historical parameters to obtain a prediction result; the prediction result indicates whether the radio resource control connection corresponding to the historical parameters is abnormal.

[0145] In this embodiment, the target model can be flexibly selected according to the actual application scenario, the computing power level of electronic devices, and the accuracy requirements of anomaly detection. For example, the target model may include, but is not limited to, BP (Back Propagation) neural network, lightweight convolutional neural network (CNN), and decision tree.

[0146] Step S305: Update the parameters of the target model based on the loss value; the loss value represents the difference between the prediction result and the labeled result corresponding to the historical parameters; the labeled result represents the actual conclusion of whether the radio resource control connection corresponding to the historical parameters is abnormal.

[0147] The higher the loss value, the lower the fit between the current parameters of the target model and the local scene, and the worse the accuracy of the inference logic; conversely, the lower the loss value, the higher the fit and accuracy of the model.

[0148] In this embodiment, after updating the parameters of the target model, the step of obtaining historical parameters from the second storage space of the electronic device can continue until the loss value converges and the parameter update stops. At this time, the target model completes the local update and has the ability to determine RRC connection anomalies adapted to the local network scenario.

[0149] Step S306: In response to the target event, obtain the target parameters; the target parameters characterize the interaction state between the electronic device and the network.

[0150] Step S307: Based on the target model, process the target parameters to obtain the target result; the target result characterizes whether the wireless resource control connection between the electronic device and the network is abnormal.

[0151] Step S308: If the target result is characterized as abnormal and the preset termination condition is met, the wireless resource control connection between the electronic device and the network is interrupted.

[0152] For a detailed description of steps S306-S308, please refer to the relevant description of steps S101-S103 in Example 1, which will not be repeated here.

[0153] In this embodiment, when the scheme cannot be executed normally due to missing, damaged, or unretrievable model configuration information, the target model can be autonomously iterated and optimized using the device's own historical interaction data. This enables the target model to accurately adapt to the actual network interaction characteristics of the electronic device and has the ability to determine RRC connection anomalies in a local context. This avoids situations where abnormal power consumption problems cannot be resolved and user experience is compromised due to the lack of preset configuration information. It also improves the accuracy and specificity of the target model's anomaly determination, enhances the robustness, scenario adaptability, and terminal-side autonomy of the entire wireless resource connection control scheme, and steadily advances the RRC connection anomaly determination process, providing reliable model support for subsequent interruption of abnormal RRC connections and resolution of abnormal power consumption problems.

[0154] As another optional embodiment of this application, refer to Figure 4 This is a flowchart illustrating a wireless resource connection control method provided in Embodiment 6 of this application. Figure 4 As shown, the method may include, but is not limited to, the following steps:

[0155] Step S401: Obtain model configuration information from the first storage space of the electronic device; the model configuration information is used to describe the structure of the target model and the reasoning behavior of the target model.

[0156] Step S402: Configure the target model based on the model configuration information.

[0157] Step S403: Determine that the model configuration information has not been obtained from the first storage space, and obtain historical parameters from the second storage space of the electronic device; the historical parameters represent the historical interaction status between the electronic device and the network.

[0158] For a detailed description of steps S401-S403, please refer to the relevant description of steps S301-S303 in Example 5, which will not be repeated here.

[0159] Step S404: Based on the historical parameters, select a target structural description information from a variety of pre-stored structural description information; the target structural description information is used to describe the structure of the target model.

[0160] In this embodiment, the pre-stored multiple structural description information can be adapted to different network scenarios and different device computing power. Each structural description information clearly defines the core architecture of the model (such as the number of network layers, number of neurons, feature extraction dimension, and inference operation rules), and corresponds to different computing power consumption and anomaly detection accuracy (such as complex structures: BP neural network, lightweight CNN, high computing power consumption, high detection accuracy; simple structures: decision tree, low computing power consumption, detection accuracy adapted to simple scenarios).

[0161] If the historical parameters show that the abnormal scenarios are mainly based on simple patterns (such as most abnormalities being screen-off with no data and long-term RRC connections, with simple parameter correlations), and the electronic device is a low-computing-power device (such as an entry-level mobile phone or IoT terminal), then a simple structure can be chosen to describe the information (such as a decision tree structure) to avoid wasting the computing power of the baseband chip on complex structures and to ensure that the model inference is efficient and low-power.

[0162] If the historical parameters contain a variety of complex anomalies (such as multiple anomalies caused by base station configuration errors or core network signaling failures, with complex parameter relationships), and the electronic devices are high-computing devices (such as mid-to-high-end mobile phones and tablets), then complex structural description information (such as lightweight CNNs or BP neural networks) can be selected to ensure that the model can accurately capture complex anomaly patterns and improve the accuracy of anomaly judgment.

[0163] Step S405: Construct a target model based on the target structure description information.

[0164] In this embodiment, a complete target model architecture can be built in an electronic device based on the core parameters defined in the target structure description information, such as the model network type, number of layers, number of neurons, feature extraction dimension, and connection method of each layer. This ensures that the constructed model structure is completely consistent with the target structure description information and is compatible with the hardware resources of the electronic device (such as baseband processor and storage module), thus avoiding deployment failures caused by incompatibility between the structure and hardware.

[0165] Step S406: Input the historical parameters into the target model, and the target model processes the historical parameters to obtain a prediction result; the prediction result indicates whether the radio resource control connection corresponding to the historical parameters is abnormal.

[0166] Step S407: Update the parameters of the target model based on the loss value; the loss value represents the difference between the prediction result and the labeled result corresponding to the historical parameters; the labeled result represents the actual conclusion of whether the radio resource control connection corresponding to the historical parameters is abnormal.

[0167] Step S408: In response to the target event, obtain the target parameters; the target parameters characterize the interaction state between the electronic device and the network.

[0168] Step S409: Based on the target model, process the target parameters to obtain the target result; the target result characterizes whether the wireless resource control connection between the electronic device and the network is abnormal.

[0169] Step S410: If the target result is characterized as abnormal and a preset termination condition is met, the wireless resource control connection between the electronic device and the network is interrupted.

[0170] For a detailed description of steps S406-S410, please refer to the relevant description of steps S304-S308 in Example 5, which will not be repeated here.

[0171] In this embodiment, based on the historical parameters, a target structural description is selected from a variety of pre-stored structural descriptions to ensure that the selected target structural description fits the local application scenario. Based on this, a target model is constructed using the target structural description. After updating the parameters of the target model, it is ensured that the target model can adapt to complex and abnormal local scenarios to guarantee accuracy, while also adapting to the terminal's computing power level to avoid unnecessary computing power consumption and ensure efficient inference. This enhances the terminal adaptability and robustness of the entire wireless resource connection control scheme, ensuring that the model can stably and efficiently determine RRC connection anomalies. This provides reliable model support in terms of structural adaptation and hardware compatibility for subsequently interrupting abnormal RRC connections and addressing the root cause of abnormal power consumption in terminal devices.

[0172] As another optional embodiment of this application, a wireless resource connection control method provided in Embodiment 7 of this application is mainly an implementation of step S202 in Embodiment 4, which may include, but is not limited to, the following steps:

[0173] Step S21: Based on the model configuration information, configure the target model in the baseband processor; the target model is used in the baseband processor to determine whether the radio resource control connection between the electronic device and the network is abnormal.

[0174] The baseband processor can be used to undertake core functions such as RRC connection establishment, signaling interaction, data transmission, and radio resource scheduling. It can directly communicate and interact with the network side (base station, core network) and obtain RRC connection-related parameters, signaling and other information in real time.

[0175] The core function of the target model is to determine whether an RRC connection is abnormal based on target parameters. These target parameters (such as RRC connection duration, total data transmission volume, and number of CA configuration changes) can be collected and stored in real time by the baseband processor. By configuring the target model directly in the baseband processor, there is no need to transmit parameters from the baseband processor to other processors (such as the application processor, AP), avoiding delays during parameter transmission. This enables localized linkage between parameter acquisition and model inference, allowing for rapid output of RRC connection anomaly determination results, and providing assurance for timely termination of abnormal connections and reduction of unnecessary power consumption.

[0176] If the target model is configured in the application processor, the application processor needs to be started separately for model inference, and computing power is also consumed for parameter transmission, which increases the overall power consumption of the device. However, if the model is configured in the baseband processor, no other hardware modules need to be started. The inference operation can be completed by relying on the existing computing power of the baseband processor, which reduces the overall power consumption of the device.

[0177] The baseband processor directly participates in the management and signaling interaction of the RRC connection, enabling it to accurately and in real-time acquire the raw parameters and signaling information related to the RRC connection. Configuring the target model on this processor ensures that the model directly acquires the raw data without relaying, avoiding loss or distortion during parameter relay and improving the accuracy of anomaly detection in the target model. At the same time, configuring the model on the baseband processor allows for better adaptation to the hardware logic related to wireless communication, avoiding inference failures and judgment biases caused by model-hardware incompatibility.

[0178] In this embodiment, the method of configuring the target model in the baseband processor is similar to the principle of configuring the target model in Embodiment 4. For the specific configuration process, please refer to the relevant description in Embodiment 4, which will not be repeated here.

[0179] In this embodiment, the control interruption of the radio resource control connection between the electronic device and the network may include, but is not limited to, any of the following:

[0180] Step S31: Control the baseband processor to locally release the wireless resource control connection between the electronic device and the network.

[0181] When the target model in the baseband processor outputs an RRC connection anomaly determination result and meets the preset termination conditions, the baseband processor actively triggers a local RRC connection release command. In accordance with the specifications of the wireless communication protocol (such as the LTE protocol), the RRC connection resources between the electronic device and the network are released, allowing the electronic device to switch from the RRC connection state to the idle state.

[0182] In this embodiment, without relying on signaling responses from the network side, it can quickly release abnormal RRC connections in scenarios where network devices are abnormal (such as signaling transmission delays or parsing errors), promptly cut off invalid high-power consumption without service requirements, effectively solve the problem of abnormal power consumption caused by network anomalies preventing the normal release of RRC connections, and improve adaptability and stable execution capabilities in complex scenarios such as network device anomalies and abnormal signaling interactions.

[0183] Step S32: Control the baseband processor to generate a measurement report and send the measurement report to the network; the measurement report is used to trigger the network to initiate a cell handover process.

[0184] In this embodiment, when the target model in the baseband processor determines that the RRC connection is abnormal and the preset termination condition is met, the baseband processor generates a standardized measurement report. This report contains key information such as the signal strength, interference situation, and RRC connection status of the current cell, and also implicitly contains relevant characteristics of the current cell's RRC connection abnormality (such as continuous connection without data transmission, abnormal signaling interaction, etc.).

[0185] The baseband processor sends the measurement report to the network side (base station). After receiving it, the base station will determine the current cell connection is abnormal based on the information in the report, and then initiate a cell handover process to switch the electronic device to another cell with good signal and no abnormalities.

[0186] In this embodiment, by controlling the baseband processor to generate a measurement report and sending the measurement report to the network, a cell handover on the network side is triggered, which indirectly interrupts the current abnormal RRC connection and switches the device to a normal cell. This not only solves the problem of abnormal power consumption, but also ensures the continuity of subsequent communication for users (after switching to a normal cell, a new RRC connection can be established normally without affecting service use).

[0187] As another optional embodiment of this application, a wireless resource connection control method is provided in Embodiment 8 of this application. This embodiment is mainly an implementation of the preset termination condition in Embodiment 1, and may include, but is not limited to, at least one of the following:

[0188] First termination condition: The duration for which the screen of the electronic device remains off meets a set time length threshold.

[0189] In this embodiment, the duration for which the screen of the electronic device remains in the off state can be found in the description of the first parameter of the first type parameter in Embodiment 3, and will not be repeated here.

[0190] The time threshold can be flexibly adjusted according to the terminal type or user habits, and the specific value is not limited in this application. For example, a mobile terminal can be configured to 1 minute, and an IoT terminal can be configured to 5 minutes.

[0191] By using the first termination condition, accidental interruptions can be avoided in scenarios where the user's screen is off for a short period of time (such as 10 seconds or 30 seconds). For example, if the user temporarily locks the screen or switches devices but there is still normal data transmission in the background (such as background downloads or message pushes), the continuous screen-off time does not reach the threshold and will not trigger an interruption, thus ensuring the continuity of the user's normal business.

[0192] Second termination condition: The electronic device is in a non-call state.

[0193] When an electronic device is in a non-calling state, please refer to the relevant description of the third parameter of the first type of parameter in Example 3, which will not be repeated here.

[0194] In this embodiment, the call status flag can be read in real time. If the flag is 0 (not in call state), the termination condition is met. If the flag is 1 (in call state), the termination condition is not met. In this case, even if the target model determines that the RRC connection is abnormal, the interruption operation will not be triggered to avoid affecting the continuity of voice calls.

[0195] In voice call scenarios, the continuous maintenance of RRC connections is driven by normal business needs. Even without additional data transmission, its high power consumption is necessary. This termination condition can accurately identify this scenario, avoid misjudging RRC connections in normal call scenarios as abnormal and interrupting them, and ensure the user's voice call experience.

[0196] The wireless resource connection control device provided in this application will be described below. The wireless resource connection control device described below can be referred to in correspondence with the wireless resource connection control method described above.

[0197] Reference Figure 5 A wireless resource connection control device may include:

[0198] The module 100 is used to obtain target parameters in response to a target event; the target parameters characterize the interaction state between the electronic device and the network.

[0199] The processing module 200 is used to process the target parameters based on the target model to obtain the target result; the target result indicates whether the wireless resource control connection between the electronic device and the network is abnormal.

[0200] The interrupt module 300 is used to control the interruption of the wireless resource control connection between the electronic device and the network if the target result is characterized as abnormal and a preset termination condition is met.

[0201] In this embodiment, the target parameters obtained by the obtaining module 100 may include at least one of the following:

[0202] Obtain a first type of parameter; the first type of parameter represents the correlation between user operation and the power consumption state of the electronic device;

[0203] Obtain a second type of parameter; the second type of parameter is used to characterize changes in the resource configuration of the wireless resource control connection between the electronic device and the network.

[0204] Obtain a third type of parameter; the third type of parameter is used to characterize the dynamic attributes of the radio resource control connection.

[0205] In this embodiment, the first type parameter may include at least one of the following:

[0206] The first parameter, the second parameter, and the third parameter; the first parameter represents the continuous duration for which the screen of the electronic device remains off; the second parameter represents the total amount of data transmitted by the electronic device within a specified time period; and the third parameter represents whether the electronic device is in a call state.

[0207] The second type of parameter may include: a fourth parameter, which characterizes the number of times the carrier aggregation configuration changes per unit time.

[0208] The third type of parameter may include: a fifth parameter, used to characterize the total duration for which the electronic device maintains the wireless resource control connection with the network.

[0209] In this embodiment, the wireless resource connection control device may further include:

[0210] Configuration module, used for:

[0211] Model configuration information is obtained from the first storage space of the electronic device; the model configuration information is used to describe the structure of the target model and the reasoning behavior of the target model;

[0212] Configure the target model based on the model configuration information.

[0213] In this embodiment, the wireless resource connection control device may further include:

[0214] The training module is used for:

[0215] If the model configuration information is not obtained from the first storage space, historical parameters are obtained from the second storage space of the electronic device; the historical parameters represent the historical interaction status between the electronic device and the network.

[0216] The historical parameters are input into the target model, which processes the historical parameters to obtain a prediction result; the prediction result indicates whether the radio resource control connection corresponding to the historical parameters is abnormal.

[0217] Based on the loss value, the parameters of the target model are updated; the loss value represents the difference between the prediction result and the labeled result corresponding to the historical parameters; the labeled result represents the actual conclusion of whether the radio resource control connection corresponding to the historical parameters is abnormal; the status label represents the actual status of the radio resource control connection between the electronic device and the network.

[0218] The training module can also be used for:

[0219] Based on the historical parameters, a target structural description is selected from a variety of pre-stored structural description information; the target structural description information is used to describe the structure of the target model.

[0220] Based on the target structure description information, a target model is constructed.

[0221] In this embodiment, the configuration module configures the target model based on the model configuration information, which may specifically include:

[0222] Based on the model configuration information, the target model is configured in the baseband processor; the target model is used in the baseband processor to determine whether the radio resource control connection between the electronic device and the network is abnormal.

[0223] The interrupt module 300 controls the interruption of the wireless resource control connection between the electronic device and the network, and may include any of the following:

[0224] The control baseband processor locally releases the wireless resource control connection between the electronic device and the network;

[0225] The control baseband processor generates a measurement report and sends the measurement report to the network; the measurement report is used to trigger the network to initiate a cell handover process.

[0226] In this embodiment, satisfying the preset termination condition may include, but is not limited to, at least one of the following:

[0227] The duration for which the screen of the electronic device remains off meets a set time length threshold.

[0228] The electronic device is in a non-call state.

[0229] In another embodiment of this application, an electronic device is provided.

[0230] Reference Figure 6 Electronic devices may include:

[0231] Memory 10 is used to store model configuration information or historical parameters.

[0232] Baseband processor 20, used for:

[0233] In response to a target event, the model configuration information or the historical parameters are read from the memory;

[0234] The target model is obtained based on the model configuration information or the historical parameters;

[0235] Obtain target parameters; the target parameters characterize the interaction state between the electronic device and the network;

[0236] Based on the target model, the target parameters are processed to obtain the target result; the target result characterizes whether the wireless resource control connection between the electronic device and the network is abnormal;

[0237] If the target result is characterized as abnormal and a preset termination condition is met, the wireless resource control connection between the electronic device and the network is interrupted.

[0238] It should also be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. In addition, in the device embodiment drawings provided in this application, the connection relationship between modules indicates that they have a communication connection, which can be implemented as one or more communication buses or signal lines.

[0239] Through the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware, or it can be implemented by special-purpose hardware including application-specific integrated circuits, special-purpose CPUs, special-purpose memory, special-purpose components, etc. Generally, any function performed by a computer program can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be diverse, such as analog circuits, digital circuits, or special-purpose circuits. However, for this application, software program implementation is more often the preferred implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a readable storage medium, such as a computer floppy disk, USB flash drive, mobile hard disk, ROM, RAM, magnetic disk, or optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, training equipment, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0240] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product.

[0241] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, training device, or data center to another website, computer, training device, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a training device or data center that integrates one or more available media. The available media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state drives (SSDs)).

Claims

1. A wireless resource connection control method, comprising: Responding to the target event, obtain the target parameters; The target parameter characterizes the interaction state between the electronic device and the network; Based on the target model, the target parameters are processed to obtain the target result; The target result indicates whether the wireless resource control connection between the electronic device and the network is abnormal; If the target result is characterized as abnormal and a preset termination condition is met, the wireless resource control connection between the electronic device and the network is interrupted.

2. The wireless resource connection control method according to claim 1, wherein obtaining the target parameters includes at least one of the following: Obtain a first type of parameter; the first type of parameter represents the correlation between user operation and the power consumption state of the electronic device; Obtain a second type of parameter; the second type of parameter is used to characterize changes in the resource configuration of the wireless resource control connection between the electronic device and the network. Obtain a third type of parameter; the third type of parameter is used to characterize the dynamic attributes of the radio resource control connection.

3. The wireless resource connection control method according to claim 2, wherein the first type of parameter includes at least one of the following: The first parameter, the second parameter, and the third parameter; the first parameter represents the continuous duration for which the screen of the electronic device remains off; the second parameter represents the total amount of data transmitted by the electronic device within a specified time period. The third parameter indicates whether the electronic device is in a call state; The second type of parameter includes: a fourth parameter, which characterizes the number of times the carrier aggregation configuration changes per unit time; The third type of parameter includes: a fifth parameter, which characterizes the total duration for which the electronic device maintains the wireless resource control connection with the network.

4. The wireless resource connection control method according to claim 1, further comprising: Obtain model configuration information from the first storage space of the electronic device; The model configuration information is used to describe the structure of the target model and the reasoning behavior of the target model; Configure the target model based on the model configuration information.

5. The wireless resource connection control method according to claim 4, further comprising: If it is determined that the model configuration information was not obtained from the first storage space, historical parameters are obtained from the second storage space of the electronic device; The historical parameters characterize the historical interaction status between the electronic device and the network; The historical parameters are input into the target model, which processes the historical parameters to obtain a prediction result; the prediction result indicates whether the radio resource control connection corresponding to the historical parameters is abnormal. The parameters of the target model are updated based on the loss value; the loss value represents the difference between the prediction result and the labeled result corresponding to the historical parameters. The annotation result represents the actual conclusion of whether the radio resource control connection corresponding to the historical parameters is abnormal. The status label represents the actual status of the radio resource control connection between the electronic device and the network.

6. The wireless resource connection control method according to claim 5, before inputting the historical parameters into the target model and having the target model process the historical parameters to obtain the prediction result, further comprising: Based on the historical parameters, a target structural description is selected from a variety of pre-stored structural description information. The target structure description information is used to describe the structure of the target model; Based on the target structure description information, a target model is constructed.

7. The wireless resource connection control method according to claim 4, wherein configuring the target model based on the model configuration information includes: Based on the model configuration information, the target model is configured in the baseband processor; The target model is used in the baseband processor to determine whether the radio resource control connection between the electronic device and the network is abnormal. The control interruption of the wireless resource control connection between the electronic device and the network includes any of the following: The control baseband processor locally releases the wireless resource control connection between the electronic device and the network; The control baseband processor generates a measurement report and sends the measurement report to the network; the measurement report is used to trigger the network to initiate a cell handover process.

8. The wireless resource connection control method according to claim 1, wherein satisfying the preset termination condition includes at least one of the following: The duration for which the screen of the electronic device remains off meets a set time length threshold. The electronic device is in a non-call state.

9. A wireless resource connection control device, comprising: The acquisition module is used to obtain target parameters in response to target events; The target parameter characterizes the interaction state between the electronic device and the network; The processing module is used to process the target parameters based on the target model to obtain the target result; The target result indicates whether the wireless resource control connection between the electronic device and the network is abnormal; An interrupt module is used to control the interruption of the wireless resource control connection between the electronic device and the network if the target result is characterized as abnormal and a preset termination condition is met.

10. An electronic device, comprising: Memory, used to store model configuration information or historical parameters; Baseband processor, used for: In response to a target event, the model configuration information or the historical parameters are read from the memory; The target model is obtained based on the model configuration information or the historical parameters; Obtain the target parameters; The target parameter characterizes the interaction state between the electronic device and the network; Based on the target model, the target parameters are processed to obtain the target result; The target result indicates whether the wireless resource control connection between the electronic device and the network is abnormal; If the target result is characterized as abnormal and a preset termination condition is met, the wireless resource control connection between the electronic device and the network is interrupted.

Images