Control parameter configuration method and device, electronic equipment, storage medium and computer product
By constructing a control parameter configuration model using deep reinforcement learning algorithms, the DRX parameters are dynamically adjusted, solving the problem of balancing the performance of vehicle-to-everything (V2X) users' services and energy-saving needs, and achieving improvements in terminal energy saving and service quality.
Patent Information
- Application Number
- CN202510564801.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-30
- Publication Date
- 2025-12-16
AI Technical Summary
In existing technologies, manually configured control parameters are difficult to balance between the service performance and energy-saving needs of vehicle-to-everything (V2X) users, which affects the development of V2X services and user experience.
A control parameter configuration model based on deep reinforcement learning algorithm is adopted. By acquiring information on the service type, latency requirements and network status of vehicle network users, the optimal control parameters are predicted. The DRX parameters are dynamically adjusted with service latency and vehicle energy consumption as optimization objectives.
This approach effectively reduces terminal energy consumption while ensuring service quality, adapts to the dynamic changes in vehicle networking services, and improves user experience and business development.
Smart Images
Figure CN121151998A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of computer, in particular to a control parameter configuration method and device, electronic equipment, storage medium and computer product. BACKGROUND
[0002] With the rapid development of 5th Generation (5G) private network, Internet of Vehicles has become an important part of private network business. However, since the 5G public network is optimized based on the perception of ordinary consumers, some network energy saving parameter configurations cannot meet the requirements of Internet of Vehicles users for latency. The existing network configuration or parameter setting mostly depends on manual analysis and adjustment, that is, a group of fixed Discontinuous Reception (DRX) parameters are configured as control parameters for terminals by gNodeB (5G New Radio Node B) through artificial means. In this case, when the connected mode DRX is enabled, the battery power consumption of the vehicle terminal will be reduced, but this will greatly increase the business delay of the vehicle terminal. If the DRX is closed, the terminal will transmit periodic channel state information or reference signals very frequently, which will greatly increase the power consumption and the resource overhead of the network.
[0003] Therefore, the current control parameters configured by artificial means cannot balance the needs of Internet of Vehicles users for vehicle terminal business performance and power saving, which affects the development and user perception of Internet of Vehicles business. SUMMARY
[0004] The present application aims to at least solve one of the technical problems in the related art. To this end, the present application provides a control parameter configuration method and device, electronic equipment, storage medium and computer product, which can solve the problem that it is difficult to balance the needs of Internet of Vehicles users for vehicle terminal business performance and power saving by manually configuring a group of fixed control parameters for terminals.
[0005] According to the control parameter configuration method of the first aspect of the present application, the method comprises: obtaining business type information, latency requirement information and network state information of Internet of Vehicles users; inputting the business type information, the latency requirement information and the network state information into a control parameter configuration model to obtain control parameter information output by the control parameter configuration model; wherein the control parameter configuration model is constructed based on a deep reinforcement learning algorithm and is a model for predicting optimal control parameters based on business type, latency requirement and network state; the control parameter configuration model takes business latency and vehicle terminal energy consumption as optimization objectives when predicting optimal control parameters; controlling the vehicle terminal of the Internet of Vehicles user based on the control parameter information.
[0006] According to an embodiment of the present application, the control parameter configuration model is obtained by the following way: obtaining historical service information of a sample vehicle networking user; the historical service information includes historical service type information, historical network state information, historical complaint information, historical control parameter information, historical service time delay information and historical vehicle machine energy consumption information; based on the historical service information, using a weight dynamic adjustment strategy to train a prediction model constructed based on a deep reinforcement learning algorithm to obtain a control parameter configuration model; wherein the complaint information is used as a veto condition for the corresponding control parameter.
[0007] According to an embodiment of the present application, the prediction model takes the physical channel occupancy rate, service weight, network state and complaint information as the state space, and takes the control parameter as the action space; and a multi-objective weighted sum designed based on service time delay, vehicle machine energy consumption and service matching degree is used as the reward function.
[0008] According to an embodiment of the present application, the control parameter is a combination of multiple discontinuous reception parameters; the multiple discontinuous reception parameters at least include listening time, continuous listening time after burst data, long cycle sleep time, short cycle sleep time and short cycle duration; wherein each continuous parameter in the multiple discontinuous reception parameters is discretized into multiple candidate values.
[0009] According to an embodiment of the present application, the network state information at least includes channel quality, user moving speed and network load.
[0010] According to an embodiment of the present application, the vehicle machine control of the vehicle networking user based on the control parameter information includes: downloading the control parameter information to a base station; the base station downloads the control parameter information to the vehicle networking user after receiving the control parameter information; and the vehicle networking user performs vehicle machine control based on the control parameter information.
[0011] The control parameter configuration device according to the second aspect of the present application includes: an obtaining module, configured to obtain service type information, time delay requirement information and network state information of a vehicle networking user; The configuration module is configured to input the service type information, the time delay requirement information and the network state information into a control parameter configuration model to obtain control parameter information output by the control parameter configuration model, wherein the control parameter configuration model is constructed based on a deep reinforcement learning algorithm and is a model for predicting optimal control parameters based on service type, time delay requirement and network state, and the control parameter configuration model takes service time delay and vehicle-machine energy consumption as optimization objectives when predicting the optimal control parameters. The control module is configured to perform vehicle-machine control of the vehicle Internet user based on the control parameter information.
[0012] The electronic device according to the third aspect of the present application includes a memory, a processor and a computer program stored in the memory and executable on the processor, and the processor implements the control parameter configuration method according to any of the above when executing the computer program.
[0013] The storage medium according to the fourth aspect of the present application is a non-transitory computer readable storage medium, and a computer program is stored on the storage medium, and the computer program is executed by a processor to implement the control parameter configuration method according to any of the above.
[0014] The computer program product according to the fifth aspect of the present application includes a computer program, and the computer program is executed by a processor to implement the control parameter configuration method according to any of the above.
[0015] The one or more technical solutions in the embodiments of the present application have at least the following technical effects: The control parameter configuration model is constructed based on a deep reinforcement learning algorithm and is a model for predicting optimal control parameters based on service type, time delay requirement and network state, and the control parameter configuration model takes service time delay and vehicle-machine energy consumption as optimization objectives when predicting the optimal control parameters. After obtaining the service type information, the time delay requirement information and the network state information of the vehicle Internet user, the control parameter configuration model is input with the service type information, the time delay requirement information and the network state information, and the control parameter information predicted and output by the control parameter configuration model with service time delay and vehicle-machine energy consumption as optimization objectives is obtained. When performing vehicle-machine control of the vehicle Internet user based on the control parameter information, the demand of the vehicle Internet user for service performance and power saving at the vehicle-machine end can be effectively balanced, which is conducive to the development of vehicle Internet services and the improvement of user perception.
[0016] Additional aspects and advantages of the present application will be in part apparent and in part pointed out hereinafter. BRIEF DESCRIPTION OF DRAWINGS
[0017] In order to more clearly illustrate the technical solutions of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative effort based on these drawings.
[0018] Figure 1 is a flowchart of a control parameter configuration method provided by an embodiment of the present application.
[0019] Figure 2 is a structural schematic diagram of an electronic device provided by the present application. DETAILED DESCRIPTION
[0020] The embodiments of the present application will be further described in detail below with reference to the drawings and embodiments. The following embodiments are used to illustrate the present application, but cannot be used to limit the scope of the present application.
[0021] In the description of the embodiments of the present application, it should be noted that the terms "center", "longitudinal", "transverse", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the embodiments of the present application and simplifying the description, and therefore cannot be understood as indicating or implying that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the embodiments of the present application. In addition, the terms "first", "second", "third" are only for the purpose of description, and cannot be understood as indicating or implying relative importance.
[0022] In the description of the embodiments of the present application, it should be noted that unless otherwise explicitly specified and limited, the terms "connected", "connected" should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium. For those skilled in the art, the specific meaning of the above terms in the embodiments of the present application can be understood according to the specific circumstances.
[0023] In the embodiments of the present application, unless specifically defined and limited otherwise, a first feature is "on", "under", "above" or "over" a second feature can mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, the first feature "above", "over" and "on" the second feature can mean that the first feature is directly above or obliquely above the second feature, or only means that the first feature is higher in horizontal height than the second feature. The first feature "below", "under" and "under" the second feature can mean that the first feature is directly below or obliquely below the second feature, or only means that the first feature is lower in horizontal height than the second feature.
[0024] In the description of the specification, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example" or "some examples" and the like means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In the specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in any one or more embodiments or examples. In addition, the person skilled in the art can combine and combine the different embodiments or examples described in the specification and the features of the different embodiments or examples without contradiction.
[0025] It should be noted that the existing related scheme mainly configures parameters according to the fixed mode of DRX, and cannot automatically balance between user perception delay and user terminal power saving. The traditional DRX parameter configuration method (manual adjustment) often cannot adapt to the rapidly changing network environment, resulting in that the balance between system performance and power consumption is difficult to reach the optimum. In addition, the existing DRX parameter configuration algorithm lacks sufficient flexibility and adaptive ability, and cannot dynamically adjust according to the real-time changes of network state and business demand.
[0026] Based on this, the present application provides a control parameter configuration method, device, electronic equipment, storage medium and computer product.
[0027] Figure 1 is one of the flowcharts of the control parameter configuration method provided by the embodiments of the present application, as shown in Figure 1 The control parameter configuration method comprises the following steps. Step 110, acquiring service type information, delay requirement information and network state information of a user of Internet of Vehicles.
[0028] In step 120, the service type information, the time delay requirement information and the network state information are input into a control parameter configuration model to obtain control parameter information output by the control parameter configuration model; the control parameter configuration model is constructed based on a deep reinforcement learning algorithm and is a model for predicting optimal control parameters based on service types, time delay requirements and network states; and the control parameter configuration model takes service time delay and vehicle-machine energy consumption as optimization objectives when predicting the optimal control parameters.
[0029] In step 130, vehicle-machine control of the vehicle networking user is performed based on the control parameter information.
[0030] It should be noted that the execution subject of the control parameter configuration method provided in the embodiments of the present application can be a computer device of an operator or an enterprise authorized by the operator, and the computer device can be, for example, a mobile phone, a tablet computer, a notebook computer, a palm computer, a vehicle-mounted electronic device, a wearable device, an Ultra-mobile Personal Computer (UMPC), a netbook or a Personal Digital Assistant (PDA), etc. It should be noted that the data required to be obtained in the present application are obtained through a regular channel after being authorized by a relevant user.
[0031] The control parameter configuration device can be arranged in or connected to the computer device of the present application, so that the control parameter configuration method of the present application can be executed by the control parameter configuration device.
[0032] Specifically, a certain number of vehicle networking users can be obtained in advance as sample vehicle networking users in the present application, and historical service type information, historical network state information, historical complaint information, historical control parameter information, historical service time delay information and historical vehicle-machine energy consumption information and other historical service information of each sample vehicle networking user are obtained. The control parameter is a combination of multiple discontinuous reception parameters; the multiple discontinuous reception parameters at least include an OnDurationTimer, a DrxInactivityTimer, a LongCycle, a drxShortCycleTimer and a shortDrxCycle.
[0033] In addition, the present application can construct a prediction model according to a Deep Reinforcement Learning (DRL) algorithm. The deep reinforcement learning algorithm is a density-based anomaly detection algorithm for detecting abnormal points in time series data. The principle is as follows: 1. Policy representation and optimization: Deep reinforcement learning uses neural networks to model the mapping between states and actions; These neural networks are optimized through reinforcement learning methods, allowing the agent to learn the optimal policy.
[0034] 2. Value function estimation: Deep reinforcement learning uses value functions to evaluate the value of states or state-action pairs; Value functions can help agents choose actions with higher expected cumulative rewards; Deep reinforcement learning solves high-dimensional, complex environment problems by constructing and training deep neural networks to approximate value functions.
[0035] 3. Combination of deep learning and reinforcement learning: Deep learning technology can automatically extract features from raw data, reducing the workload of feature engineering; Reinforcement learning technology learns strategies by interacting with the environment, allowing agents to adapt to different environments; The combination of deep learning and reinforcement learning represents and optimizes the behavior strategy of the agent through neural networks, and uses value functions to evaluate the value of states or state-action pairs.
[0036] Further, the prediction model can be trained by historical service information of sample V2X users, wherein the service delay and the vehicle terminal energy consumption are taken as optimization targets during the training process to predict the optimal control parameter.
[0037] Therefore, after the prediction model is trained, a model that can predict the optimal control parameter according to the input service type, delay requirement and network state information, taking the service delay and the vehicle terminal energy consumption as optimization targets, is obtained, and the model is used as a control parameter configuration model.
[0038] Therefore, for V2X users (specifically, vehicle terminal of vehicles in V2X) within the jurisdiction, multi-dimensional data collection can be performed. Among them, the service type information and the delay requirement information of the V2X user can be determined through the historical service information of the V2X user. At the same time, the network state information of the V2X user can also be obtained. The network state information in the present application at least includes channel quality, user moving speed and network load, and the network state information can be collected by the base station. Thus, information perception is realized.
[0039] Further, the service type information, the time delay requirement information and the network state information of the vehicle networking user can be input to the control parameter configuration model together, and the control parameter configuration model predicts the optimal control parameter based on the service type, the time delay requirement and the network state. After the control parameter configuration model is predicted, the control parameter output by the control parameter configuration model can be obtained. Thus, information decision is realized.
[0040] Further, the control parameter can be issued to the vehicle networking user, so that the vehicle networking user can perform vehicle control through the control parameter. Thus, configuration and execution of the control parameter are realized.
[0041] According to the control parameter configuration method of the embodiment of the present application, the control parameter configuration model for predicting the optimal control parameter based on the service type, the time delay requirement and the network state, and taking the service time delay and the vehicle energy consumption as the optimization target is constructed through the deep reinforcement learning algorithm. Then, after the service type information, the time delay requirement information and the network state information of the vehicle networking user are obtained, the service type information, the time delay requirement information and the network state information are input to the control parameter configuration model, and the control parameter information predicted and output by the control parameter configuration model taking the service time delay and the vehicle energy consumption as the optimization target can be obtained. When the vehicle control of the vehicle networking user is performed based on the control parameter information, the demand of the vehicle networking user for the service performance and the power saving of the vehicle can be effectively balanced, which is beneficial to the development of the vehicle networking service and the improvement of the user perception.
[0042] Based on the above embodiment, the control parameter configuration model is obtained in the following way: The historical service information of the sample vehicle networking user is obtained. The historical service information includes the historical service type information, the historical network state information, the historical complaint information, the historical control parameter information, the historical service time delay information and the historical vehicle energy consumption information. Based on the historical service information, the prediction model constructed based on the deep reinforcement learning algorithm is trained by using the strategy of dynamic adjustment of weights, and the control parameter configuration model is obtained. The complaint information is used as the veto condition of the corresponding control parameter.
[0043] Specifically, the present application can select the connected vehicle with free state mixed service (i.e. with different service types) and complaint in a certain city as the sample vehicle networking user.
[0044] Further, the historical service information of the sample vehicle networking user, such as the historical service type information, the historical network state information, the historical complaint information, the historical control parameter information, the historical service time delay information and the historical vehicle energy consumption information, is obtained, and the training data set is further constructed. The control parameter information is the DRX parameter combination. The complaint information is used as the veto condition of the corresponding control parameter. That is, when a certain group of control parameter information generates user complaint when executed, the control parameter information will be vetoed.
[0045] For example, part of the training data shown in Table 1 can be obtained as follows: Table 1 .
[0046] Wherein, RAT is a wireless access technology; XDR is a data record; OMC-R is a radio access network operation and maintenance center; ECT is used to identify the cell where the user is located; ID represents an identifier; APP-TYPE represents an application type; IMSI represents an international mobile subscriber identity; APP-SUBTYPE is used to further refine the type of application in the mobile network data or application classification, indicating a subcategory under a certain main application category (APP-TYPE); TCP is one of the core protocols on the Internet, belonging to the transport layer; MQTT is a widely used lightweight message transmission protocol; CELL represents a cell; CELL_ID represents a cell number.
[0047] Further, data cleaning can be performed on the data in the training data set: the CELLID (base station cell number) in the above data is the key value, and the data that cannot be associated or the field data that is empty is cleaned and deleted as redundant data.
[0048] Further, feature engineering can be performed on the cleaned data: the collected data is preprocessed and feature extracted, including: Vehicle terminal state: speed, terminal power; Network state: Physical Resource Block (PRB) physical channel occupancy rate, service delay threshold.
[0049] Service type: delay-sensitive (such as intersection collision warning), low-power (such as periodic state reporting).
[0050] Further, service classification and delay weight allocation can be performed: Classification model: in order to achieve the purpose of lightweight, on the basis of 5G Deep Packet Inspection (DPI) historical data, according to the sensitivity and preference of the service to the delay, the Internet of Vehicles service is divided into delay-sensitive service and low-power service. The following is the basis for division: Delay-sensitive service: data packet arrival delay (average), service type probability (delay-sensitive weight θ, low-power weight δ); Low-power service: high weight δ is allocated to extend the sleep cycle.
[0051] The weight is dynamically adjusted. When implementing a time delay high-sensitive service such as automatic driving, remote control, navigation, dynamic path planning, etc., the theta weight is automatically increased to reduce the risk of time delay; when performing vehicle entertainment, over-the-air (OTA) firmware upgrade and other services, the delta weight is automatically increased to reduce power consumption.
[0052] Further, based on historical service information, the prediction model constructed based on the deep reinforcement learning algorithm can be trained by using the strategy of dynamic weight adjustment, and the control parameter configuration model is obtained after the training is completed.
[0053] Specifically, the optimal strategy can be learned online by combining deep reinforcement learning with parameter mapping rules, and a hierarchical parameter mapping rule is introduced to reduce the computational complexity.
[0054] The prediction model constructed based on the deep reinforcement learning algorithm can be realized by the following process: (1) State space definition: including PRB physical channel occupation rate, service type weight, network state and historical complaint data.
[0055] (2) Action space definition: the DRX parameter combination is taken as the action space for dynamic adjustment. In order to reduce the output dimension, the continuous parameters are discretized into 3-5 candidate values.
[0056] Among them, OnDurationTimer: listening duration, positively correlated with business activity. Mapped to the [2 milliseconds, 20 milliseconds] interval (typical value of Internet of Vehicles).
[0057] DrxInactivityTimer: continuous listening time after burst data, shortened when high channel utilization. Mapped to the [1 millisecond, 10 millisecond] interval. Discrete action (candidate values 20 milliseconds, 40 milliseconds, 80 milliseconds).
[0058] LongCycle / ShortCycle: hierarchical mapping based on service delay threshold (such as LongCycle=80 milliseconds for low latency sensitive service, shortened to 10 milliseconds for high latency sensitive service). The two are linked together, and the proportion according to the 3rd Generation Partnership Project (3GPP) standard (such as LongCycle=80 milliseconds for low latency sensitive service, shortened to 20 milliseconds for high latency sensitive service).
[0059] (3) Reward function design: The Internet of Vehicles scenario needs to balance the time delay, energy consumption and service type, and the reward function is designed as a multi-objective weighted sum. Specifically as follows: ; Wherein, E saved represents energy saving, which is proportional to DRX sleep duration, and the maximum value is (terminal maximum power); D_actual represents actual latency, and the penalty is increased when the threshold D_threshold (such as 50 milliseconds) is exceeded; δ, θ and λ are weight coefficients for balancing the importance of latency and energy consumption; energy saving reward δ: E_saved = DRX start sleep duration * average unit time energy consumption; latency penalty θ: when the actual latency D_actual exceeds the threshold D_threshold, the penalty is increased linearly according to the overtime ratio, for example: the vehicle collision warning latency threshold D_threshold requires 50 milliseconds, and the actual latency D_actual is 60 milliseconds, and the penalty is calculated according to 0.2; service type matching reward λ: P_match = probability of latency sensitive service request, which needs to be aligned with the DRX parameters as much as possible.
[0060] (4) Dynamic adjustment of weights: When the service is a latency sensitive service: increase the weight of θ (latency sensitive) and reduce the weight of δ; During the emergency service (such as collision warning), ignore the energy saving reward (δ = 0) and only optimize the latency and service matching. If there are complaints from users of the network latency quality after the implementation of the configuration scheme, δ needs to be set to zero; When the network is congested (PRB physical channel occupancy rate > 65%): increase the weight of λ to prioritize key services.
[0061] Further, simulation training can be performed, and a control parameter configuration model is obtained after the training is completed.
[0062] The application can periodically update the training data set and the prediction model to adapt to changes in the network environment and business needs.
[0063] In addition, a user feedback mechanism can be introduced to adjust the reward function and model parameters according to the actual experience of complaining users and the complaint content, further improving the prediction accuracy.
[0064] The application can dynamically adjust the DRX parameters according to the user business latency demand and network state, realize terminal energy saving under the premise of ensuring business service quality, and has the following advantages: Strong adaptability: can dynamically adjust the DRX parameters according to the business demand and network state, adapt to the dynamic changes of vehicle networking business and user mobility.
[0065] Significant energy saving effect: by optimizing the DRX parameters, the terminal energy consumption is effectively reduced, and the battery endurance time is prolonged.
[0066] Ensure business service quality: energy saving optimization is carried out under the premise of ensuring business latency demand, avoiding affecting business service quality due to energy saving.
[0067] High degree of intelligence: using machine learning algorithms to realize intelligent prediction and optimization of DRX parameters, improve network resource utilization efficiency.
[0068] Based on the above embodiments, the vehicle machine control of the Internet of Vehicles user is based on the control parameter information, which includes: The control parameter information is sent to the base station, and the base station sends the control parameter information to the Internet of Vehicles user after receiving the control parameter information. The Internet of Vehicles user performs vehicle machine control based on the control parameter information.
[0069] Specifically, the application can send control parameter information to the base station in the area where the Internet of Vehicles user is located.
[0070] Further, the base station can receive the control parameter information, establish an index for the optimal DRX parameter configuration combination in the control parameter information and send it to the Internet of Vehicles user, so that the Internet of Vehicles user can control the vehicle machine terminal to execute the DRX mechanism according to the control parameter information.
[0071] The application can effectively balance the demand of the Internet of Vehicles user for the business performance and power saving of the vehicle machine terminal by performing vehicle machine control based on the control parameter information predicted by taking the business delay and vehicle machine energy consumption as the optimization target, which is conducive to the development of Internet of Vehicles business and improvement of user perception.
[0072] The control parameter configuration method provided by the application is a self-adaptive parameter configuration scheme for terminal power saving and business delay of Internet of Vehicles based on deep reinforcement learning algorithm, which finds the optimal solution of terminal delay and terminal power saving parameter configuration, realizes terminal energy saving on the premise of guaranteeing business service quality, and finally adjusts the DRX parameters of the system to the special website.
[0073] Through testing, it is found that when the scheme of the application is adopted, the terminal delay of Internet of Vehicles can meet the average 17ms of mixed business delay, and the power consumption is relatively low. When other schemes are adopted, the terminal power consumption is the lowest, but it cannot meet the delay requirement, or the delay meets the requirement but the terminal power consumption is high.
[0074] The control parameter configuration device provided by the application is described below, and the control parameter configuration device described below can be correspondingly referred to the control parameter configuration method described above.
[0075] Further, the application also provides a control parameter configuration device.
[0076] The control parameter configuration device includes: An acquisition module is configured to acquire business type information, delay demand information and network state information of an Internet of Vehicles user. The configuration module is configured to input the service type information, the time delay requirement information and the network state information into a control parameter configuration model to obtain control parameter information output by the control parameter configuration model; wherein the control parameter configuration model is constructed based on a deep reinforcement learning algorithm and is a model for predicting optimal control parameters based on service type, time delay requirement and network state; and the control parameter configuration model takes service time delay and vehicle-machine energy consumption as optimization objectives when predicting the optimal control parameters. The control module is configured to perform vehicle-machine control of the vehicle Internet user based on the control parameter information.
[0077] The control parameter configuration device of the application constructs a control parameter configuration model for predicting optimal control parameters based on service type, time delay requirement and network state and taking service time delay and vehicle-machine energy consumption as optimization objectives through a deep reinforcement learning algorithm, and then inputs service type information, time delay requirement information and network state information of a vehicle Internet user into the control parameter configuration model to obtain control parameter information predicted and output by the control parameter configuration model based on service time delay and vehicle-machine energy consumption as optimization objectives; and thus, when performing vehicle-machine control of the vehicle Internet user based on the control parameter information, the demand of the vehicle Internet user for service performance and power saving of the vehicle-machine end can be effectively balanced, which is conducive to the development of vehicle Internet services and improvement of user perception.
[0078] In one embodiment, the control module is specifically configured to: The control parameter information is sent to a base station; the base station sends the control parameter information to the vehicle Internet user after receiving the control parameter information; and the vehicle Internet user performs vehicle-machine control based on the control parameter information.
[0079] Figure 2 An example of an entity structure diagram of an electronic device is shown in Figure 2 As shown in the figure, the electronic device can include a processor 210, a communications interface 220, a memory 230 and a communications bus 240, wherein the processor 210, the communications interface 220 and the memory 230 complete mutual communication through the communications bus 240. The processor 210 can invoke logical instructions in the memory 230 to perform the following method: obtaining service type information, time delay requirement information and network state information of a vehicle Internet user; input the service type information, the time delay requirement information and the network state information into a control parameter configuration model to obtain control parameter information output by the control parameter configuration model; the control parameter configuration model is constructed based on a deep reinforcement learning algorithm and is a model for predicting optimal control parameters based on service type, time delay requirement and network state; when predicting the optimal control parameters, the control parameter configuration model takes service time delay and vehicle-machine energy consumption as optimization objectives; perform vehicle-machine control of the vehicle networking user based on the control parameter information.
[0080] In addition, the logical instructions in the memory 230 described above can be implemented in the form of a software function unit and sold or used as an independent product, and can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the parts that contribute to the related art or parts of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.
[0081] In yet another aspect, the embodiments of the present application also provide a non-transitory computer readable storage medium having a computer program stored thereon, which is executed by a processor to implement the method provided by the above-mentioned embodiments, for example, including: obtaining service type information, time delay requirement information and network state information of a vehicle networking user; input the service type information, the time delay requirement information and the network state information into a control parameter configuration model to obtain control parameter information output by the control parameter configuration model; the control parameter configuration model is constructed based on a deep reinforcement learning algorithm and is a model for predicting optimal control parameters based on service type, time delay requirement and network state; when predicting the optimal control parameters, the control parameter configuration model takes service time delay and vehicle-machine energy consumption as optimization objectives; perform vehicle-machine control of the vehicle networking user based on the control parameter information.
[0082] In yet another aspect, the embodiments of the present application also provide a computer program product having a computer program stored thereon, which is executed by a processor to implement the method provided by the above-mentioned embodiments, for example, including: obtaining service type information, time delay requirement information and network state information of a vehicle networking user; input the service type information, the time delay requirement information and the network state information into a control parameter configuration model to obtain control parameter information output by the control parameter configuration model; wherein the control parameter configuration model is constructed based on a deep reinforcement learning algorithm and is a model for predicting optimal control parameters based on service type, time delay requirement and network state; and the control parameter configuration model takes service time delay and vehicle energy consumption as optimization objectives when predicting the optimal control parameters; perform vehicle-machine control of the vehicle networking user based on the control parameter information.
[0083] The device embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separated, and the components displayed as units can or can not be physical units, i.e., can be located in one place or distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the embodiment scheme according to actual needs. Those skilled in the art can understand and implement it without creative labor.
[0084] Through the above description of the embodiments, those skilled in the art can clearly understand that the embodiments can be realized by means of software and necessary universal hardware platforms, and of course can also be realized by hardware. Based on such understanding, the above technical solutions can be embodied in the form of software products, and the computer software products can be stored in a computer readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and include a plurality of instructions for making a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiments.
[0085] Finally, it should be noted that the above embodiments are only used to illustrate the present application, and not to limit the present application. Although the present application is described in detail with reference to the embodiments, those skilled in the art should understand that various combinations, modifications or equivalent replacements of the technical solutions of the present application do not deviate from the spirit and scope of the present application.
Claims
1. A method of controlling parameter configuration, characterized by, The method comprises: obtaining service type information, time delay requirement information and network state information of a vehicle Internet user; inputting the service type information, the time delay requirement information and the network state information into a control parameter configuration model to obtain control parameter information output by the control parameter configuration model; wherein the control parameter configuration model is constructed based on a deep reinforcement learning algorithm and is a model for predicting optimal control parameters based on service type, time delay requirement and network state; when predicting the optimal control parameters, the control parameter configuration model takes service time delay and vehicle terminal energy consumption as optimization objectives; performing vehicle terminal control of the vehicle Internet user based on the control parameter information.
2. The control parameter configuration method according to claim 1, wherein The control parameter configuration model is obtained by: obtaining historical service information of a sample vehicle Internet user; the historical service information comprises historical service type information, historical network state information, historical complaint information, historical control parameter information, historical service time delay information and historical vehicle terminal energy consumption information; training a prediction model constructed based on a deep reinforcement learning algorithm based on the historical service information by using a strategy of dynamically adjusting weights to obtain a control parameter configuration model; wherein the complaint information is used as a veto condition for the corresponding control parameter.
3. The control parameter configuration method according to claim 2, wherein The prediction model takes physical channel occupancy rate, service weight, network state and complaint information as a state space and takes control parameters as an action space; a multi-objective weighted sum designed based on service time delay, vehicle terminal energy consumption and service matching degree is used as a reward function.
4. The control parameter configuration method according to claim 3, wherein The control parameters are combinations of multiple discontinuous reception parameters; the multiple discontinuous reception parameters at least comprise listening time, continuous listening time after burst data, long cycle sleep time, short cycle sleep time and short cycle duration; wherein each continuous parameter in the multiple discontinuous reception parameters is discretized into multiple candidate values.
5. The control parameter configuration method of claim 1, wherein, The network state information at least comprises channel quality, user moving speed and network load.
6. The control parameter configuration method of claim 1, wherein, The vehicle terminal control of the vehicle Internet user based on the control parameter information comprises: downloading the control parameter information to a base station; the base station downloads the control parameter information to the vehicle Internet user after receiving the control parameter information; and the vehicle Internet user performs vehicle terminal control based on the control parameter information.
7. A control parameter configuration apparatus characterized by comprising: The method comprises: obtaining service type information, time delay requirement information and network state information of a vehicle Internet user; inputting the service type information, the time delay requirement information and the network state information into a control parameter configuration model to obtain control parameter information output by the control parameter configuration model; wherein the control parameter configuration model is constructed based on a deep reinforcement learning algorithm and is a model for predicting optimal control parameters based on service type, time delay requirement and network state; when predicting the optimal control parameters, the control parameter configuration model takes service time delay and vehicle terminal energy consumption as optimization objectives; performing vehicle terminal control of the vehicle Internet user based on the control parameter information.
8. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the control parameter configuration method in any one of claims 1-6.
9. A storage medium, which is a non-transitory computer-readable storage medium, having stored thereon a computer program, characterized in that, The computer program, which is executed by a processor, implements the control parameter configuration method according to any one of claims 1-6.
10. A computer program product comprising a computer program, characterized in that, The computer program, which is executed by a processor, implements the control parameter configuration method according to any one of claims 1-6.