Resource scheduling method, device and system and storage medium

Abstract

The invention discloses a resource scheduling method, the method is applied to a resource scheduling node, and the method comprises the following steps: receiving first terminal application data based on a target service sent by a terminal device; the first terminal application data at least comprises a use experience evaluation value of the target service; acquiring current first network environment data; determining a trained resource scheduling model; the resource scheduling model is obtained by performing model training on a target cell where the terminal equipment is located at the management node; inputting the identification information of the target service, the first terminal application data and the first network environment data to a resource scheduling model for prediction to obtain a target resource scheduling parameter; the target resource scheduling parameter corresponds to a target resource of a target service of the terminal device; and executing a network scheduling operation for the terminal equipment based on the target resource scheduling parameter. The invention further discloses a first resource scheduling device, a second resource scheduling device, a third resource scheduling device, a system and a storage medium.

Description

Technical Field

[0001] This application relates to the field of wireless technologies, and in particular, to a resource scheduling method, apparatus, system, and storage medium. Background Art

[0002] In the process of network application, with the development of network technologies, the requirements for the network have gradually shifted from simply pursuing network performance to the user experience of using the network. However, in the current network resource scheduling process, operations are still implemented based on fixed parameters and preset scheduling rules, and there is currently no effective and reliable method for dynamically scheduling network resources according to real-time user feedback and user experience. Therefore, there is an urgent need to propose a resource scheduling method that can improve the service efficiency of the network system while meeting the user experience.

[0003] Content of the Application

[0004] To solve the above technical problems, this application aims to provide a resource scheduling method, apparatus, system, and storage medium, which solve the problem that the user experience is not considered in the current resource scheduling process, and propose a method for dynamically scheduling network resources according to real-time user feedback and user experience, fully considering the user experience while ensuring the service efficiency of the network system.

[0005] The technical solution of this application is implemented as follows:

[0006] This application provides a resource scheduling method, which is applied to a resource scheduling node, and the method includes:

[0007] Receiving first terminal application data based on a target service sent by a terminal device; wherein, the first terminal application data at least includes a usage experience evaluation value of the target service;

[0008] Obtaining current first network environment data;

[0009] Determining a trained resource scheduling model; wherein, the resource scheduling model is obtained by training a model at a management node for a target cell where the terminal device is located;

[0010] Inputting identification information of the target service, the first terminal application data, and the first network environment data into the resource scheduling model for prediction to obtain target resource scheduling parameters; wherein, the target resource scheduling parameters correspond to target resources of the target service of the terminal device;

[0011] Performing a network scheduling operation for the terminal device based on the target resource scheduling parameters.

[0012] In the above solution, performing a network scheduling operation for the terminal device based on the target resource scheduling parameter includes:

[0013] Allocating resources for the target cell according to the target resource scheduling parameter;

[0014] Determining a first weight coefficient of a priority determination parameter for determining a resource scheduling priority; wherein, the priority determination parameter includes an evaluation parameter corresponding to the usage experience evaluation value;

[0015] Based on the priority determination parameter and the first weight coefficient, determining the scheduling priority of the target service.

[0016] In the above solution, determining the first weight coefficient of the priority determination parameter for determining a resource scheduling priority includes:

[0017] If the usage experience evaluation value is less than or equal to a preset threshold, increasing the current first weight coefficient of the corresponding evaluation parameter by a preset adjustment step.

[0018] In the above solution, the method further includes:

[0019] Sending a downlink control message to at least one access device currently accessing in the target cell; wherein, at least one of the access devices includes the terminal device, and the downlink control message is used to instruct each access device to report second terminal application data;

[0020] Receiving the second terminal application data reported by each access device;

[0021] Determining second network environment data when receiving the second terminal application data of each;

[0022] Storing the second terminal application data of each and the corresponding second network environment data.

[0023] In the above solution, the method further includes:

[0024] Sending the stored terminal application data set and network environment data set corresponding to the target cell to the management node according to a preset period.

[0025] In the above solution, determining the trained resource scheduling model includes:

[0026] Receiving a reference scheduling model sent by the management node;

[0027] Updating the resource scheduling model to the reference scheduling model.

[0028] This application provides a resource scheduling method, which is applied to a management node. The method includes:

[0029] If an update instruction for updating the model is detected, the corresponding resource scheduling model is updated by model training using the set of terminal application data and the set of network environment parameters of the target cell sent by the resource scheduling node, and a reference scheduling model is obtained. Among them, the set of terminal application data at least includes the values of evaluation parameters of at least one provided network service.

[0030] Send the reference scheduling model to the resource scheduling node. Among them, the reference scheduling model is used to predict the target resource scheduling parameters for the resource scheduling node to perform network scheduling on the target cell.

[0031] In the above solution, the method further includes:

[0032] Obtain the data of the sample to be trained. Among them, the data of the sample to be trained includes historical terminal application data and historical network environment data in units of cells, and the historical terminal application data at least includes historical usage experience evaluation values for network services.

[0033] Use the data of the sample to be trained to perform model training on the model to be trained, and obtain the resource scheduling model. Among them, the model to be trained is used to determine the optimal network environment parameters corresponding to the highest usage experience evaluation value in each category after classifying the data according to features.

[0034] This application provides a resource scheduling method, which is applied to a terminal device. The method includes:

[0035] Obtain the key network performance indicators corresponding to the target service currently in use.

[0036] Based on the key network performance indicators, determine the first terminal application data.

[0037] Send the first terminal application data to the resource scheduling node, so that the resource scheduling node predicts the network resources of the target cell currently accessed based on the first terminal application data.

[0038] In the above solution, the determining the first terminal application data based on the key network performance indicators includes:

[0039] Determine the score corresponding to each indicator in the key network performance indicators.

[0040] Determine the second weight coefficient corresponding to each indicator in the key network performance indicators.

[0041] Perform weighted calculation on the said score corresponding to each index and the corresponding second weight coefficient to obtain the usage experience evaluation value of the said target service;

[0042] Determine that the said first terminal application data includes the usage experience evaluation value and the key network performance indicators.

[0043] This application provides a first resource scheduling device, which is applied to a resource scheduling node. The device includes: a receiving unit, a first obtaining unit, a first determining unit, a predicting unit, and an executing unit; wherein:

[0044] The receiving unit is configured to receive first terminal application data sent by a terminal device based on a target service; wherein, the first terminal application data at least includes the usage experience evaluation value of the said target service;

[0045] The first obtaining unit is configured to obtain current first network environment data;

[0046] The first determining unit is configured to determine a trained resource scheduling model; wherein, the resource scheduling model is obtained by model training at a management node for a target cell where the terminal device is located;

[0047] The predicting unit is configured to input the identification information of the said target service, the first terminal application data, and the first network environment data into the first resource scheduling model for prediction to obtain target resource scheduling parameters; wherein, the target resource scheduling parameters correspond to the target resources of the said target service of the terminal device;

[0048] The executing unit is configured to perform a network scheduling operation for the terminal device based on the target resource scheduling parameters.

[0049] This application provides a second resource scheduling device, which is applied to a management node. The device includes: an updating unit and a first sending unit; wherein:

[0050] The updating unit is configured to, if an update instruction for updating the model is detected, use a set of terminal application data and a set of network environment parameters of a target cell sent by a resource scheduling node to perform model training and updating on the corresponding resource scheduling model to obtain a reference scheduling model; wherein, the set of terminal application data at least includes the values of evaluation parameters of at least one provided network service;

[0051] The first sending unit is configured to send the reference scheduling model to the resource scheduling node; wherein, the reference scheduling model is used to enable the resource scheduling node to predict the target resource scheduling parameters for network scheduling of the target cell.

[0052] The present application provides a third resource scheduling device, which is applied to a terminal device and includes a second acquisition unit, a second determination unit, and a second transmission unit; where:

[0053] The second acquisition unit is configured to acquire network performance key indicators corresponding to a target service currently in use.

[0054] The second determination unit is configured to determine first terminal application data based on the network performance key indicators.

[0055] The second transmission unit is configured to send the first terminal application data to a resource scheduling node, so that the resource scheduling node predicts network resources of a target cell currently accessed based on the first terminal application data.

[0056] The present application provides a resource scheduling system, which includes a management node, at least one resource scheduling node, and at least one access device including a terminal device; where:

[0057] The management node is configured to implement the steps of the resource scheduling method described in any one of the above.

[0058] The resource scheduling node is configured to implement the steps of the resource scheduling method described in any one of the above.

[0059] The terminal device is configured to implement the steps of the resource scheduling method described in any one of the above.

[0060] The present application provides a storage medium, which is characterized in that a resource scheduling program is stored on the storage medium, and when the resource scheduling program is executed by a processor, the steps of the resource scheduling method described in any one of the above are implemented.

[0061] The embodiments of the present application provide a resource scheduling method, apparatus, system and storage medium. After the terminal device obtains the key network performance indicators corresponding to the target service currently in use, based on the key network performance indicators, the first terminal application data is determined and sent to the resource scheduling node. After receiving the first terminal application data based on the target service sent by the terminal device, the resource scheduling node obtains the current first network environment data and determines the trained resource scheduling model. The identification information of the target service, the first terminal application data and the first network environment data are input into the resource scheduling model for prediction to obtain the target resource scheduling parameters. Finally, based on the target resource scheduling parameters, a network scheduling operation for the terminal device is executed. In this way, the resource scheduling node calls the resource scheduling model to perform model prediction on the real-time feedback data of the terminal device, that is, the first terminal application data regarding the target service and the data of the first network environment where the resource scheduling node is located, and obtains the prediction result, so as to adjust the network scheduling operation for the terminal device according to the prediction result, solving the problems such as not considering the user experience in the current resource scheduling process, and proposing a method for dynamically scheduling network resources according to the user's real-time feedback and user experience. While fully considering the user experience, the service efficiency of the network system is also ensured. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 is a schematic flowchart of the resource scheduling method provided by the embodiment of the present application Figure 1 ;

[0063] Figure 2 is a schematic flowchart of the resource scheduling method provided by the embodiment of the present application Figure 2 ;

[0064] Figure 3 is a schematic flowchart of the resource scheduling method provided by the embodiment of the present application Figure 3 ;

[0065] Figure 4 is a schematic flowchart of the resource scheduling method provided by the embodiment of the present application Figure 4 ;

[0066] Figure 5 is a schematic flowchart of the resource scheduling method provided by the embodiment of the present application Figure 5 ;

[0067] Figure 6 is a schematic flowchart of the resource scheduling method provided by the embodiment of the present application Figure 6 ;

[0068] Figure 7 is a schematic flowchart of the resource scheduling method provided by the embodiment of the present application Figure 7 ;

[0069] Figure 8 Flow schematic of the resource scheduling method provided by the embodiment of the present application Figure 8 ;

[0070] Figure 9 Schematic diagram of a scoring magnitude provided by the embodiment of the present application;

[0071] Figure 10 Schematic diagram of the process of model construction by a network management platform provided by the embodiment of the present application;

[0072] Figure 11 Schematic diagram of the structure of a prediction model provided by the embodiment of the present application;

[0073] Figure 12 Schematic diagram of the implementation of an application embodiment of the resource scheduling method provided by the embodiment of the present application;

[0074] Figure 13 Schematic diagram of the structure of a first resource scheduling device provided by the embodiment of the present application;

[0075] Figure 14 Schematic diagram of the structure of a second resource scheduling device provided by the embodiment of the present application;

[0076] Figure 15 Schematic diagram of the structure of a third resource scheduling device provided by the embodiment of the present application;

[0077] Figure 16 Schematic diagram of the structure of a resource scheduling system provided by the embodiment of the present application. Detailed implementation manners

[0078] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application.

[0079] An embodiment of the present application provides a resource scheduling method. Referring to Figure 1 as shown, the method is applied to a resource scheduling node, and the method includes the following steps:

[0080] Step 101, receive the first terminal application data based on the target service sent by the terminal device.

[0081] Among them, the first terminal application data includes at least the evaluation value of the usage experience of the target service.

[0082] In the embodiments of the present application, the resource scheduling node may be a node in a communication system that provides communication network resource services for a terminal device and manages the communication network resources of the terminal device. For example, it may be a base station. When the terminal device uses a target service, it evaluates the target service to obtain a usage experience evaluation value for the target service, and obtains first terminal application data based at least on the usage experience evaluation value, and sends the first terminal application data to the resource scheduling node.

[0083] Step 102: Obtain the current first network environment data.

[0084] In the embodiments of the present application, the first network environment data is the network feature distribution of the current network environment of the resource scheduling node.

[0085] Step 103: Determine the trained resource scheduling model.

[0086] Among them, the resource scheduling model is obtained by training a model at the management node for the target cell where the terminal device is located.

[0087] In the embodiments of the present application, the resource scheduling model may be a resource prediction model currently stored in the resource scheduling node, or a resource prediction model obtained from the management node.

[0088] Step 104: Input the identification information of the target service, the first terminal application data, and the first network environment data into the resource scheduling model for prediction to obtain the target resource scheduling parameter.

[0089] Among them, the target resource scheduling parameter corresponds to the target resource of the target service of the terminal device.

[0090] In the embodiments of the present application, the identification information of the target service is the identity identification information used to uniquely identify the target service. The resource scheduling node inputs the obtained identification information of the target service, the received first terminal application data, and its own first network environment data into the input layer of the resource scheduling model, and after the prediction processing of the resource scheduling model, outputs the target resource scheduling parameter.

[0091] Step 105: Perform a network scheduling operation for the terminal device based on the target resource scheduling parameter.

[0092] In the embodiments of the present application, the resource scheduling node performs a network scheduling operation processing on the network corresponding to the terminal device according to the target resource scheduling parameter. In this way, the resource scheduling node performs network scheduling on the network of the target service corresponding to the terminal device according to the real-time feedback of the user and in combination with its own network environment, thereby improving the network performance of the target service and improving the user experience effect for the target service.

[0093] The resource scheduling method provided by the embodiment of the present application, after receiving the first terminal application data based on the target service sent by the terminal device through the resource scheduling node, obtains the current first network environment data, determines the trained resource scheduling model, inputs the identification information of the target service, the first terminal application data, and the first network environment data into the resource scheduling model for prediction, obtains the target resource scheduling parameters, and finally, based on the target resource scheduling parameters, executes the network scheduling operation for the terminal device. In this way, the resource scheduling node calls the resource scheduling model to perform model prediction on the real-time feedback data of the terminal device, that is, the first terminal application data regarding the target service and the first network environment and data where the resource scheduling node is located, and obtains the prediction result, so as to adjust the network scheduling operation for the terminal device according to the prediction result, solves the problems such as not considering the user experience in the current resource scheduling process, and proposes a method for dynamically scheduling network resources according to the user's real-time feedback and user experience. While fully considering the user experience, it also ensures the service efficiency of the network system.

[0094] Based on the foregoing embodiments, an embodiment of the present application provides a resource scheduling method. Referring to Figure 2 as shown, the method is applied to a management node, and the method includes the following steps:

[0095] Step 201, if an update instruction for updating the model is detected, use the set of terminal application data and the set of network environment parameters of the target cell sent by the resource scheduling node to perform model training and update on the corresponding resource scheduling model to obtain a reference scheduling model.

[0096] Among them, the set of terminal application data includes at least the values of the evaluation parameters of at least one provided network service.

[0097] In the embodiment of the present application, the management node can be a network management device or the like. The update instruction can be generated after the management node detects a trigger condition for updating a certain model, or can be an instruction sent by other devices to the management node for instructing the resource scheduling model to perform model update. When the management node detects an update instruction for indicating model update, the management node performs model training and update on the resource scheduling model corresponding to the target cell using the sample data set sent by the resource scheduling node, that is, the set of terminal application data and the corresponding set of network environment parameters, to obtain a reference scheduling model.

[0098] Among them, the resource scheduling models corresponding to different cells can be the same or different. Generally, different cells correspond to different resource scheduling models, that is, the resource scheduling models corresponding to different cells have their respective corresponding cell characteristics.

[0099] Step 202, send the reference scheduling model to the resource scheduling node.

[0100] Among them, the reference scheduling model is used to predict the target resource scheduling parameters for the resource scheduling node to perform network scheduling on the target cell.

[0101] In the embodiment of the present application, the management node sends the updated and trained reference scheduling model to the resource scheduling node to ensure that the prediction model at the resource scheduling node is the latest model, thereby ensuring that the prediction results at the resource scheduling node are more reliable, more real-time, and conform to the actual situation, effectively guaranteeing the user experience effect.

[0102] For the resource scheduling method provided in the embodiment of the present application, if an update instruction for updating the model is detected, the management node uses the set of terminal application data and network environment parameters of the target cell sent by the resource scheduling node to perform model training and update on the corresponding resource scheduling model, obtains the reference scheduling model, and sends the reference scheduling model to the resource scheduling node, so that the resource scheduling node calls the reference scheduling model to predict the target scheduling parameters for network scheduling of the target cell and perform corresponding network scheduling operations. In this way, the resource scheduling node calls the resource scheduling model to perform model prediction on the real-time feedback data of the terminal device, that is, the first terminal application data regarding the target service and the first network environment and data where the resource scheduling node is located, obtains the prediction result, and adjusts the network scheduling operation for the terminal device according to the prediction result, solving the problems such as not considering the user experience in the current resource scheduling process, and proposing a method for dynamically scheduling network resources according to the user's real-time feedback and user experience. While fully considering the user experience, it also ensures the service efficiency of the network system.

[0103] Based on the foregoing embodiments, the embodiment of the present application provides a resource scheduling method. Referring to Figure 3 as shown, this method is applied to a terminal device, and the method includes the following steps:

[0104] Step 301, obtain the key network performance indicators corresponding to the target service currently in use.

[0105] In the embodiment of the present application, the target service may be, for example, a video communication service, a voice communication service, an Internet communication service, etc. For the services of different services, the network performance indicators for ensuring the service quality of the service are different. Therefore, when the terminal device currently provides the target service, the key network performance indicators corresponding to the target service can be determined. The key network performance indicators corresponding to the target service include at least one indicator.

[0106] Step 302, determine the first terminal application data based on the key network performance indicators.

[0107] In an embodiment of the present application, the key network performance indicators corresponding to the target service are obtained, the corresponding indicator data is obtained, and the corresponding indicator data is analyzed to obtain the first terminal application data.

[0108] Step 303: Send the first terminal application data to the resource scheduling node.

[0109] Among them, the first terminal application data is sent to the resource scheduling node, so that the resource scheduling node predicts the network resources of the target cell currently accessed based on the first terminal application data.

[0110] In an embodiment of the present application, the terminal device sends the currently collected first terminal application data to the resource scheduling node. In this way, the resource scheduling node can make a prediction based on the real-time first terminal application data fed back by the terminal device, and then adjust the network of the target cell corresponding to the terminal device to improve the user experience effect.

[0111] The resource scheduling method provided by the embodiment of the present application obtains the key network performance indicators corresponding to the target service currently used by the terminal device, determines the first terminal application data based on the key network performance indicators, and sends the first terminal application data to the resource scheduling node. In this way, the resource scheduling node calls the resource scheduling model to perform model prediction on the real-time feedback data of the terminal device, that is, the first terminal application data regarding the target service, the first network environment where the resource scheduling node is located, and the data, and obtains the prediction result, so as to adjust the network scheduling operation for the terminal device according to the prediction result, solving the problems such as not considering the user experience in the current resource scheduling process, and proposing a method for dynamically scheduling network resources according to the user's real-time feedback and user experience. While fully considering the user experience, it also ensures the service efficiency of the network system.

[0112] Based on the foregoing embodiments, an embodiment of the present application provides a resource scheduling method. Referring to Figure 4 as shown, the method further includes the following steps:

[0113] Step 401: The terminal device obtains the key network performance indicators corresponding to the target service currently used.

[0114] In an embodiment of the present application, taking the terminal device as an intelligent mobile phone as an example for illustration, the intelligent mobile phone determines the target service currently used by the current user, and obtains the key network performance indicators corresponding to the target service currently used, such as the transmission rate of the currently used service, the service response delay, the bit error rate, etc.

[0115] Step 402: The terminal device determines the score corresponding to each indicator in the key network performance indicators.

[0116] In the embodiments of the present application, the intelligent mobile phone determines the score corresponding to each indicator according to the obtained key network performance indicators, and the scores corresponding to different indicator values of each indicator are different.

[0117] Step 403: The terminal device determines the second weight coefficient corresponding to each indicator in the key network performance indicators.

[0118] In the embodiments of the present application, for the impact of the key network performance indicators on the target service, the weight coefficient corresponding to each indicator is preset. The second weight coefficient corresponding to each indicator can be an empirical value obtained based on a large number of experiments in advance and can be corrected and adjusted according to the actual situation in the actual application scenario.

[0119] Step 404: The terminal device performs weighted calculation on the score corresponding to each indicator and the corresponding second weight coefficient to obtain the usage experience evaluation value of the target service.

[0120] In the embodiments of the present application, the intelligent mobile phone uses the weighted calculation method to calculate the score corresponding to each indicator and the corresponding second weight coefficient to obtain the total usage experience evaluation value of the intelligent mobile phone for the target service.

[0121] Step 405: The terminal device determines that the first terminal application data includes the usage experience evaluation value and the key network performance indicators.

[0122] In the embodiments of the present application, the intelligent mobile phone performs packaging processing at least based on the determined key network performance indicators and usage experience evaluation value to obtain the first terminal application data.

[0123] Step 406: The terminal device sends the first terminal application data to the resource scheduling node.

[0124] Wherein, sending the first terminal application data to the resource scheduling node enables the resource scheduling node to predict the network resources of the target cell currently accessed based on the first terminal application data.

[0125] Step 407: The resource scheduling node receives the first terminal application data sent by the terminal device based on the target service.

[0126] Wherein, the first terminal application data at least includes the usage experience evaluation value of the target service.

[0127] Step 408: The resource scheduling node obtains the current first network environment data.

[0128] In the embodiments of the present application, taking the resource scheduling node as the base station as an example for illustration, the base station obtains its current network state, such as network scenario, network load distribution and other parameters, to obtain the first network environment data.

[0129] Step 409: The resource scheduling node determines the trained resource scheduling model.

[0130] The resource scheduling model is obtained by training a model at the management node for the target cell where the terminal device is located.

[0131] In an embodiment of the present application, the base station obtains the currently stored resource scheduling model.

[0132] Step 410: The resource scheduling node inputs the identification information of the target service, the first terminal application data, and the first network environment data into the resource scheduling model for prediction to obtain the target resource scheduling parameters.

[0133] The target resource scheduling parameters correspond to the target resources of the target service of the terminal device.

[0134] In an embodiment of the present application, the base station inputs the identification information of the target service, the first terminal application data, and the first network environment output received from the terminal device in the target cell from the input layer of the resource scheduling model, and the output layer of the resource scheduling model outputs the target resource scheduling parameters.

[0135] Step 411: The resource scheduling node performs a network scheduling operation for the terminal device based on the target resource scheduling parameters.

[0136] In an embodiment of the present application, the base station performs a network scheduling operation on the network resources of the target cell according to the target resource scheduling parameters, so that the base station provides target service for the target cell according to the target resource scheduling parameters, ensuring the user experience effect.

[0137] Based on the foregoing embodiments, in other embodiments of the present application, step 411 may be implemented by steps 411a to 411c:

[0138] Step 411a: The resource scheduling node allocates resources for the target cell according to the target resource scheduling parameters.

[0139] In an embodiment of the present application, when the resource scheduling node performs network scheduling according to the target resource scheduling parameters, resource scheduling allocation is first performed, that is, the network resource parameters corresponding to the target cell are adjusted to the target resource scheduling parameters.

[0140] Step 411b: The resource scheduling node determines the first weight coefficient of the priority determination parameter for determining the resource scheduling priority.

[0141] The priority determination parameter includes an evaluation parameter corresponding to the user experience evaluation value.

[0142] In the embodiment of the present application, when the resource scheduling node performs network scheduling according to the target resource scheduling parameter, the scheduling user priority is also adjusted. When adjusting the scheduling user priority, the priority determination parameter is first determined. The priority determination parameter at least includes the evaluation parameter corresponding to the usage experience evaluation value, and further may also include the relevant parameters of the service quality provided by the resource scheduling node. The first weight coefficient can be an empirical value obtained based on a large number of experiments. The value of the first weight coefficient is usually less than 1.

[0143] Step 411c: The resource scheduling node determines the scheduling priority of the target service based on the priority determination parameter and the first weight coefficient.

[0144] In the embodiment of the present application, the resource scheduling node performs a weighted summation process on the priority determination parameter and the first weight coefficient to obtain the scheduling priority corresponding to the target service.

[0145] Based on the foregoing embodiment, in other embodiments of the present application, step 411b may be implemented by the following steps: If the usage experience evaluation value is less than or equal to the preset threshold, the current first weight coefficient of the corresponding evaluation parameter is increased according to the preset adjustment step.

[0146] In the embodiment of the present application, the preset threshold is the average value of the corresponding experience evaluation of the target service. When the usage experience evaluation value fed back by the terminal device is less than or equal to the preset threshold, the current first weight coefficient of the evaluation parameter corresponding to the usage experience evaluation value is directly increased according to the preset adjustment step to obtain the updated first weight coefficient, so as to ensure the user's usage experience by using the updated first weight coefficient.

[0147] Based on the foregoing embodiment, in other embodiments of the present application, referring to Figure 5 as shown, the resource scheduling node is further configured to execute steps 412 to 415:

[0148] Step 412: The resource scheduling node sends a downlink control message to at least one access device currently accessing within the target cell.

[0149] Among them, at least one access device includes a terminal device, and the downlink control message is used to instruct each access device to report second terminal application data.

[0150] In the embodiments of the present application, whether the resource scheduling node adjusts network scheduling according to user feedback can be controlled by the resource scheduling node itself. When the resource scheduling node is configured to adjust network scheduling for a target cell according to user feedback, the resource scheduling node sends a downlink control message to at least one access device in the target cell. In this way, after each access device receives the downlink control message sent by the resource scheduling node, it responds to the downlink control message, processes the target service such as scoring, generates corresponding second terminal application data, and sends the generated second terminal application data to the resource scheduling node.

[0151] Step 413: The resource scheduling node receives the second terminal application data reported by each access device.

[0152] Step 414: The resource scheduling node determines the second network environment data when receiving each second terminal application data.

[0153] In the embodiments of the present application, when receiving each second terminal application data, the resource scheduling node also obtains the network environment data at this time, and obtains the second network environment data corresponding to each second terminal application data.

[0154] Step 415: The resource scheduling node stores each second terminal application data and the corresponding second network environment data.

[0155] In the embodiments of the present application, the resource scheduling node stores each received second terminal application data and the corresponding second network environment data as sample data for subsequent model training or model updating.

[0156] Based on the foregoing embodiments, in other embodiments of the present application, as shown in Figure 6 After step 415 of the resource scheduling node, the resource scheduling node can also be used to execute step 416:

[0157] Step 416: The resource scheduling node sends the stored set of terminal application data and the set of network environment data corresponding to the target cell to the management node at a preset period.

[0158] In the embodiments of the present application, the preset period is a pre-configured time experience interval. The resource scheduling node sends the collected set of terminal application data and the corresponding set of network environment data about the target cell to the management node at the preset period, so that the management node can perform model training and updating based on the set of terminal application data and the corresponding set of network environment data sent by the resource scheduling node.

[0159] Based on the foregoing embodiments, in other embodiments of the present application, as shown in Figure 7 The management node can also be used to execute steps 417 to 418:

[0160] Step 417: If an update instruction for updating the model is detected, the management node uses the terminal application data set and the network environment parameter set of the target cell sent by the resource scheduling node to perform model training and update on the corresponding resource scheduling model, and obtains a reference scheduling model.

[0161] Among them, the terminal application data set includes at least the values of the evaluation parameters of at least one provided network service.

[0162] In the embodiment of the present application, the update instruction detected by the management node may be generated when the management node receives that the number of samples for the target cell exceeds a certain number, or may be generated when the management node reaches a preset time according to a certain time setting, or may also be sent to the management node by other devices for controlling the management node. Specifically, it can be determined according to the actual situation and is not limited here.

[0163] When the management node detects an update instruction, the management node uses the method of model training, and uses the data in the terminal application data set and the data in the network environment parameter set sent by the resource scheduling node to perform model training on the resource scheduling model until a trained reference scheduling model is obtained. In this way, model training is performed at the management node, effectively reducing the consumption pressure of the computing resources of the resource scheduling node.

[0164] Step 418: The management node sends the reference scheduling model to the resource scheduling node.

[0165] Among them, the reference scheduling model is used to predict the target resource scheduling parameters for the resource scheduling node to perform network scheduling on the target cell.

[0166] In the embodiment of the present application, the management node sends the trained reference scheduling model to the resource scheduling node, so that the resource scheduling node can perform prediction according to the reference scheduling model obtained from the latest user feedback, ensuring that the predicted target resource scheduling parameters closely follow the actual usage experience of the user and improving the usage experience effect of the user.

[0167] Correspondingly, step 409 can be implemented by steps 409a to 409b:

[0168] Step 409a: The resource scheduling node receives the reference scheduling model sent by the management node.

[0169] Step 409b: The resource scheduling node updates the resource scheduling model to the reference scheduling model.

[0170] Based on the foregoing embodiments, in other embodiments of the present application, as shown in Figure 8 The management node is further configured to execute steps 419 to 420:

[0171] Step 419: The management node obtains the sample data to be trained.

[0172] Among them, the sample data to be trained includes historical terminal application data and historical network environment data in units of cells. The historical terminal application data includes at least the historical usage experience evaluation value for network services.

[0173] Step 420: The management node uses the sample data to be trained to train the model to be trained, and obtains a resource scheduling model.

[0174] Among them, the model to be trained is used to determine the optimal network environment parameters corresponding to the highest usage experience evaluation value in each category after classifying the data according to features.

[0175] Based on the foregoing embodiments, the embodiments of the present application provide a resource scheduling method. From the perspective of user perception, by defining a user service experience evaluation method, poor-quality user identification is performed, and a mechanism for real-time feedback based on the evaluation results and an adjustment method for network policies are established. It can be adjusted in real time according to changes in service experience, ensure consistent service experience for various services in multiple scenarios, reduce the proportion of poor-quality users, and improve network performance.

[0176] Among them, the method for the user side to evaluate the service experience can be: The terminal extracts key index features such as the current service transmission rate, service response delay, and bit error rate, and uses the scoring method corresponding to each key index feature to score each key index feature. Among them, the scoring method corresponding to each key index feature can be implemented using a rating scale as shown in Figure 9 . If the score corresponding to each key index feature is higher, it means that the experience of this index is better; then, the scores of each key index are weighted and averaged to obtain the comprehensive user experience score. Among them, the weight of each index can be distinguished according to its importance in this type of service. Finally, the satisfaction score of the current user experience for this type of service is obtained.

[0177] The scoring principle followed during the scoring process can be: Initial experience indicators under different performance manifestations are formulated for each type of service. The terminal scores by comparing the key indicators of the current network performance with the initial indicators. The scoring level represents the difference from the fixed initial indicators, and the greater the difference, the worse the experience. In the process of realizing the comprehensive user experience score, the weights of each key index and the real-time scores of each key index are weighted and averaged. Different weight factors can be configured according to the service type during weighting. For example, video services have high requirements for the key index of service transmission efficiency. Therefore, when scoring video services, the weight of the key index of service transmission rate can be configured to be greater than the weights of other key indicators. For game services, which have high requirements for the key index of service response delay, the weight of game services can be configured to be greater than the weights of other key indicators.

[0178] When the network management platform obtains sample data during training, the process can be as follows: The terminal feeds back terminal-related content to the base station, and the base station periodically reports to the network management platform the terminal-related content and network environment-related content fed back by the terminal in units of cells. Among them, the terminal-related content can at least include, for example: service type, transmission rate, service response delay, key indicators such as bit error rate, and user experience evaluation score; among them, the user experience score can be calculated based on key indicators such as transmission rate, service response delay, and bit error rate. The network environment-related content can at least include, for example: the network scenario corresponding to the base station and the network load distribution of the base station.

[0179] The process of the network management platform constructing a model can refer to Figure 10 as shown, including:

[0180] Step b11: Construct input eigenvalue.

[0181] Among them, the network management platform constructs sample data. Taking the cell as the dimension, it associates the network environment-related content with the terminal-related content according to the data collection rule, such as the same collection time period, and then constructs a set of eigenvalues with the associated network environment-related content and terminal-related content in terms of the terminal, including network environment information, network compliance level, user experience score, service performance, and key performance indicators.

[0182] Step b12: Use Artificial Intelligence (AI) to train to obtain an association model and an association function.

[0183] Among them, the network management platform selects a suitable AI algorithm for training to obtain an association model and an association function between the eigenvalues corresponding to a certain service of the input. Through the association model and the association function, data with the same characteristics are divided into one grid in the form of constructing a virtual grid, and the key performance indicator value with the highest user experience score is selected and recorded in the virtual grid to obtain the key performance indicator value with the optimal user experience under different network scenarios, different network load distributions, and different user experience feedbacks, such as including transmission rate, service response delay, bit error rate, etc.

[0184] Step b13: Output the predicted optimal key performance indicator.

[0185] Among them, when predicting for a user, according to the network environment parameters (including network scenario and network load distribution) corresponding to the user's current base station and the terminal data (that is, the user's usage experience feedback and the key indicator values of the service), quickly match the belonging grid, and use the key performance indicator value of the belonging grid obtained by the match to adjust the scheduling strategy, so as to improve the user experience. The network management platform corresponds to the aforementioned management node.

[0186] Exemplarily, as Figure 11 shown, it is a schematic structural diagram of a prediction model provided by an embodiment of the present application, including an input layer, a hidden layer, and an output layer. Correspondingly, a certain number of cell opening data collections are selected within a certain area to obtain a training sample data set. Each training sample data corresponds to the foregoing feature values. The feature values input by the input layer can be denoted as d1, d2, d3, ……, dn, and the optimal key indicators output by the output layer include indicator 1, indicator 2, ……, indicator k. The specific values of n and k are determined by the actual situation. By Figure 11 training the model shown, continuously adjusting the weight coefficients of the functions in the hidden layer, training to obtain a fitting function between the input and the output, and obtaining the adjusted weight matrix v1 and weight matrix v2 corresponding to the hidden layer. In this way, a virtual grid based on user experience evaluation is constructed, and the key indicators corresponding to the optimal service experience under the current network environment and load level are obtained, thereby adjusting the scheduling strategy.

[0187] The implementation process of adjusting the scheduling strategy based on user experience feedback can refer to Figure 12 shown, including the following steps:

[0188] Step c11: The base station configures the opening mode of data collection for collecting user experience feedback.

[0189] Among them, the base station can inform the user terminal device (User Equipment, UE) to start user experience evaluation and information reporting through downlink control information, and can also specify the reporting location and information format through downlink indication information.

[0190] Step c12: The base station receives the user experience score and key performance indicators reported by the UE for the target service to obtain the terminal-side data.

[0191] Among them, the reporting information format of the UE may include: Message Type for indicating the message type, UE type of service indicating the UE service type, and User Experience Score indicating the current user experience score.

[0192] Step c13: The base station collects the network environment information of the base station.

[0193] Among them, the network environment information includes at least the network scenario and the network load distribution

[0194] Step c14: The base station reports the network environment information and the terminal-side data to the Operation Maintenance Center (OMC).

[0195] The OMC corresponds to the aforementioned network management platform and management node. After the network side receives the network side environment information and terminal side data reported by the base station, data cleaning and storage can also be performed.

[0196] Step c15: The OMC performs model training to obtain a resource scheduling model.

[0197] Step c16: The OMC configures and distributes the resource scheduling model.

[0198] Among them, in order to achieve intercommunication between abnormal manufacturers, a new transmission interface can be defined between the OMC and the base station, and new fields such as Experience Evaluation Model can be defined for transmitting relevant models. The content to be distributed includes: Message Type indicating the message type, Target Cell ID indicating the identity identifier of the target cell to be distributed, and Experience Evaluation Model Information indicating the user experience evaluation model information of this cell.

[0199] Step c17: The base station receives the real-time user experience score and real-time key performance indicators for the target service reported by the UE.

[0200] Step c18: The base station uses the received resource scheduling model to perform prediction processing on the real-time user experience score, real-time key performance indicators, and the corresponding real-time network renewal information of the base station, matches the corresponding grid, and outputs the optimal performance key indicators.

[0201] Step c19: The base station performs network scheduling based on the optimal performance key indicators.

[0202] Among them, the network scheduling process mainly includes:

[0203] 1. Scheduling resource allocation: Set and allocate resources according to the optimal performance key indicators output by the model, such as the optimal user transmission rate, the lowest bit error rate, and the lowest service response delay.

[0204] 2. Scheduling user priority adjustment: Add a weight factor to the existing scheduling algorithm to dynamically adjust the influence degree of different factors on the scheduling priority.

[0205] For example, the calculation method of the scheduling priority can be defined as:

[0206] P = α1*f(A) + α2*f(B) + α3*f(C) + α4*f(D)

[0207] Where A, B, C, and D respectively represent different factors considered in the scheduling priority calculation formula. For example, they can be channel condition, channel quality, service bearing level requirement, scheduler waiting delay of the terminal to be scheduled, etc. in sequence; α1, α2, α3, and α4 are respectively the weight factors corresponding to each factor, and their values are all less than 1. The influence degree of different factors on the priority can be flexibly adjusted according to the currently real-time feedback situation.

[0208] It should be further noted that when the real-time user service experience score value is lower than the average score of this type of service, the weight factor of the service priority factor in the scheduling priority algorithm can be increased, so as to improve the scheduling priority of this service in the next scheduling cycle and ensure the user's service experience.

[0209] In this way, through the artificial intelligence model, a flexible example adjustment of the network scheduling strategy can be realized according to the current user experience evaluation situation, and flexible selection of multiple types of guarantee factors can be combined with the current overall situation of the cell to ensure the consistent service experience of various services in multiple scenarios, reduce the proportion of users with poor quality, improve the network performance, and thus ensure the user's usage experience effect.

[0210] It should be noted that the descriptions of the same steps and the same content in this embodiment and other embodiments can refer to the descriptions in other embodiments, and will not be repeated here.

[0211] The resource scheduling method provided by the embodiment of the present application, after obtaining the key network performance indicators corresponding to the target service currently used by the terminal device, determines the first terminal application data based on the key network performance indicators, and sends the first terminal application data to the resource scheduling node. After receiving the first terminal application data based on the target service sent by the terminal device, the resource scheduling node obtains the current first network environment data, and determines the trained resource scheduling model. The identification information of the target service, the first terminal application data, and the first network environment data are input into the resource scheduling model for prediction to obtain the target resource scheduling parameters. Finally, based on the target resource scheduling parameters, a network scheduling operation for the terminal device is executed. In this way, the resource scheduling node calls the resource scheduling model to perform model prediction on the real-time feedback data of the terminal device, that is, the first terminal application data about the target service and the data of the first network environment where the resource scheduling node is located, and obtains the prediction result, so as to adjust the network scheduling operation for the terminal device according to the prediction result, solves the problem that the user experience is not considered in the current resource scheduling process, and proposes a method for dynamically scheduling network resources according to the user's real-time feedback and user experience. While fully considering the user's usage experience, it also ensures the service efficiency of the network system.

[0212] Based on the foregoing embodiments, an embodiment of the present application provides a first resource scheduling device, which is applied to a resource scheduling node and can be applied to Figure 1 , Figures 4 - 8 in the resource scheduling method provided in the corresponding embodiment. As shown in Figure 13 , the first resource scheduling device 5 at least includes: a receiving unit 51, a first obtaining unit 52, a first determining unit 53, a prediction unit 54, and an execution unit 55; where:

[0213] The receiving unit 51 is configured to receive first terminal application data based on a target service sent by a terminal device; where the first terminal application data at least includes a usage experience evaluation value of the target service;

[0214] The first obtaining unit 52 is configured to obtain current first network environment data;

[0215] The first determining unit 53 is configured to determine a trained resource scheduling model; where the resource scheduling model is obtained by training a model at a management node for a target cell where the terminal device is located;

[0216] The prediction unit 54 is configured to input identification information of the target service, the first terminal application data, and the first network environment data into the first resource scheduling model for prediction to obtain target resource scheduling parameters; where the target resource scheduling parameters correspond to the target resources of the target service of the terminal device;

[0217] The execution unit 55 is configured to perform a network scheduling operation for the terminal device based on the target resource scheduling parameters.

[0218] In other embodiments of the present application, the execution unit includes: an allocation module, a first determination module, and a second determination module; where:

[0219] The allocation module is configured to allocate resources for the target cell according to the target resource scheduling parameters;

[0220] The first determination module is configured to determine a first weight coefficient of a priority determination parameter for the resource scheduling priority; where the priority determination parameter includes an evaluation parameter corresponding to the usage experience evaluation value;

[0221] The second determination module is configured to determine the scheduling priority of the target service based on the priority determination parameter and the first weight coefficient.

[0222] In other embodiments of the present application, the first determination module is specifically configured to implement the following steps:

[0223] If the usage experience evaluation value is less than or equal to a preset threshold, increase the current first weight coefficient of the corresponding evaluation parameter by a preset adjustment step.

[0224] In other embodiments of the present application, the first resource determination device further includes: a sending unit and a storage unit; where:

[0225] The sending unit is configured to send a downlink control message to at least one access device within the target cell that is currently accessed; where the at least one access device includes a terminal device, and the downlink control message is used to instruct each access device to report second terminal application data;

[0226] The receiving unit is further configured to receive the second terminal application data reported by each access device;

[0227] The first determination unit is further configured to determine second network environment data when each second terminal application data is received;

[0228] The storage unit is configured to store each second terminal application data and the corresponding second network environment data.

[0229] In other embodiments of the present application, the sending unit is further configured to perform the following steps:

[0230] Send the stored set of terminal application data and the set of network environment data corresponding to the target cell to the management node according to a preset period.

[0231] In other embodiments of the present application, the first determination unit includes: a receiving module and an updating module; where:

[0232] The receiving module is configured to receive a reference scheduling model sent by the management node;

[0233] The updating module is configured to update the resource scheduling model to the reference scheduling model.

[0234] It should be noted that for the specific implementation process of the interaction between the units and modules in the embodiments of the present application, reference may be made to Figure 1 、 Figures 4 - 8 the implementation process in the resource scheduling method provided in the corresponding embodiments, which will not be elaborated here.

[0235] The first resource scheduling device provided by the embodiment of the present application, after receiving the first terminal application data based on the target service sent by the terminal device through the resource scheduling node, obtains the current first network environment data, determines the trained resource scheduling model, inputs the identification information of the target service, the first terminal application data, and the first network environment data into the resource scheduling model for prediction to obtain the target resource scheduling parameters, and finally performs a network scheduling operation for the terminal device based on the target resource scheduling parameters. In this way, the resource scheduling node calls the resource scheduling model to perform model prediction on the real-time feedback data of the terminal device, that is, the first terminal application data regarding the target service and the first network environment and data where the resource scheduling node is located, and obtains the prediction result, so as to adjust the network scheduling operation for the terminal device according to the prediction result, solves the problems such as not considering the user experience in the current resource scheduling process, and proposes a method for dynamically scheduling network resources according to the user's real-time feedback and user experience. While fully considering the user experience, it also ensures the service efficiency of the network system.

[0236] Based on the foregoing embodiments, the embodiment of the present application provides a second resource scheduling device, which is applied to the management node, and this device can be applied to Figure 2 、 Figures 4 - 8 the resource scheduling method provided in the corresponding embodiment, as shown in Figure 14 The second resource scheduling device 6 at least includes: an update unit 61 and a first sending unit 62; where:

[0237] The update unit 61 is used to, if an update instruction for updating the model is detected, use the set of terminal application data and network environment parameters of the target cell sent by the resource scheduling node to perform model training and update on the corresponding resource scheduling model to obtain a reference scheduling model; where the set of terminal application data at least includes the values of evaluation parameters of at least one provided network service;

[0238] The first sending unit 62 is used to send the reference scheduling model to the resource scheduling node; where the reference scheduling model is used to enable the resource scheduling node to predict the target resource scheduling parameters for network scheduling of the target cell.

[0239] In other embodiments of the present application, the second resource scheduling device further includes: a third acquisition unit and a model training unit; where:

[0240] The third acquisition unit is used to acquire the sample data to be trained; where the sample data to be trained includes historical terminal application data and historical network environment data in units of cells, and the historical terminal application data at least includes historical usage experience evaluation values for network services;

[0241] A model training unit is configured to perform model training on a model to be trained using sample data to be trained, so as to obtain a resource scheduling model. The model to be trained is used to determine, after classifying data according to features, the optimal network environment parameters corresponding to the highest usage experience evaluation value in each category.

[0242] It should be noted that for the specific implementation process of the interaction between units and modules in the embodiments of the present application, reference may be made to Figure 2 , Figures 4 - 8 the implementation process in the resource scheduling method provided in the corresponding embodiment, which will not be elaborated here.

[0243] For the second resource scheduling device provided in the embodiments of the present application, if an update instruction for updating the model is detected, the management node uses the terminal application data set and network environment parameter set of the target cell sent by the resource scheduling node to perform model training and update on the corresponding resource scheduling model, obtains a reference scheduling model, and sends the reference scheduling model to the resource scheduling node, so that the resource scheduling node calls the reference scheduling model to predict the target scheduling parameters for network scheduling of the target cell, and performs corresponding network scheduling operations. In this way, the resource scheduling node calls the resource scheduling model to perform model prediction on the real-time feedback data of the terminal device, that is, the first terminal application data regarding the target service and the first network environment and data where the resource scheduling node is located, obtains a prediction result, and adjusts the network scheduling operation for the terminal device according to the prediction result, solving the problem that the user experience is not considered in the current resource scheduling process, and proposing a method for dynamically scheduling network resources according to the user's real-time feedback and user experience. While fully considering the user's usage experience, it also ensures the service efficiency of the network system.

[0244] Based on the foregoing embodiments, the embodiments of the present application provide a third resource scheduling device, which is applied to a terminal device. This device can be applied to Figures 3 - 8 the resource scheduling method provided in the corresponding embodiment. As shown in Figure 15 , this third resource scheduling device 7 at least includes: a second acquisition unit 71, a second determination unit 72, and a second sending unit 73. Among them:

[0245] The second acquisition unit 71 is configured to acquire the key network performance indicators corresponding to the target service currently in use.

[0246] The second determination unit 72 is configured to determine the first terminal application data based on the key network performance indicators.

[0247] The second sending unit 73 is configured to send the first terminal application data to the resource scheduling node, so that the resource scheduling node predicts the network resources of the target cell currently accessed based on the first terminal application data.

[0248] In other embodiments of the present application, the second determination unit includes: a third determination module and a calculation module; where:

[0249] The third determination module is configured to determine the score corresponding to each index in the key network performance indicators.

[0250] The third determination module is further configured to determine the second weight coefficient corresponding to each index in the key network performance indicators.

[0251] The calculation module is configured to perform weighted calculation on the score corresponding to each index and the corresponding second weight coefficient to obtain the usage experience evaluation value of the target service.

[0252] The third determination module is further configured to determine that the first terminal application data includes the usage experience evaluation value and the key network performance indicators.

[0253] It should be noted that for the specific implementation process of the interaction between units and modules in the embodiments of the present application, reference may be made to Figures 3 - 8 the implementation process in the resource scheduling method provided in the corresponding embodiment, which will not be elaborated here.

[0254] After the third resource scheduling device provided in the embodiments of the present application obtains the key network performance indicators corresponding to the target service currently used through the terminal device, based on the key network performance indicators, it determines the first terminal application data and sends the first terminal application data to the resource scheduling node. In this way, the resource scheduling node calls the resource scheduling model to perform model prediction on the real-time feedback data of the terminal device, that is, the first terminal application data regarding the target service and the first network environment and data where the resource scheduling node is located, to obtain a prediction result, so as to adjust the network scheduling operation for the terminal device according to the prediction result, solving the problems such as not considering the user experience in the current resource scheduling process, and proposing a method for dynamically scheduling network resources according to the user's real-time feedback and user experience. While fully considering the user's usage experience, it also ensures the service efficiency of the network system.

[0255] Based on the foregoing embodiments, an embodiment of the present application provides a resource scheduling system, which can be applied to Figures 1 - 8 the resource scheduling method provided in the corresponding embodiment, with reference to Figure 16 as shown, the system 8 may at least include: a management node 81, at least one resource scheduling node 82, and at least one access device 83 including terminal devices; where:

[0256] The management node 81 is configured to implement the implementation process in the resource scheduling method provided in the corresponding embodiment such as Figure 1 , Figures 4 - 7 the corresponding embodiment, which will not be elaborated here;

[0257] The resource scheduling node 82 is configured to implement asFigure 2 , Figures 4 - 7 The implementation process in the resource scheduling method provided by the corresponding embodiment will not be elaborated here;

[0258] The terminal device 83 is used to implement the implementation process in the resource scheduling method provided by the corresponding embodiment, which will not be elaborated here. Figures 3 - 7 The implementation process in the resource scheduling method provided by the corresponding embodiment will not be elaborated here.

[0259] Based on the foregoing embodiments, an embodiment of the present application provides a computer-readable storage medium, simply referred to as a storage medium. The computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the implementation process in the resource scheduling method provided by the corresponding embodiment, which will not be elaborated here. Figures 1 - 8 The implementation process in the resource scheduling method provided by the corresponding embodiment will not be elaborated here.

[0260] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) containing computer-usable program code.

[0261] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of the processes and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the specified functions in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0262] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured product including an instruction device, and the instruction device implements the specified functions in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0263] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, so that a series of operation steps are executed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions for implementing the functions specified in one process or a plurality of processes and / or blocks Figure 1 one process or a plurality of processes and / or blocks Figure 1 steps for implementing the functions specified in one block or a plurality of blocks.

[0264] As described above, the above are only preferred embodiments of the present application, and are not intended to limit the protection scope of the present application.

Claims

1. A resource scheduling method, characterized in that, The method is applied to a resource scheduling node, and the method includes: Receiving first terminal application data of a target service sent by a terminal device; wherein, the first terminal application data at least includes a usage experience evaluation value of the target service; Obtaining current first network environment data; Determining a trained resource scheduling model; wherein, the resource scheduling model is obtained by model training at a management node for a target cell where the terminal device is located; Inputting identification information of the target service, the first terminal application data, and the first network environment data into the resource scheduling model for prediction to obtain target resource scheduling parameters; wherein, the target resource scheduling parameters correspond to target resources of the target service of the terminal device; Performing a network scheduling operation for the terminal device based on the target resource scheduling parameters.

2. The method according to claim 1, characterized in that, The performing a network scheduling operation for the terminal device based on the target resource scheduling parameters includes: Allocating resources for the target cell according to the target resource scheduling parameters; Determining a first weight coefficient of a priority determination parameter for determining a resource scheduling priority; wherein, the priority determination parameter includes an evaluation parameter corresponding to the usage experience evaluation value; Determining the scheduling priority of the target service based on the priority determination parameter and the first weight coefficient.

3. The method according to claim 2, wherein The determining a first weight coefficient of a priority determination parameter for determining a resource scheduling priority includes: If the usage experience evaluation value is less than or equal to a preset threshold, increasing the current first weight coefficient of the corresponding evaluation parameter by a preset adjustment step.

4. The method according to claim 1, characterized in that, The method further includes: Sending a downlink control message to at least one access device currently accessing in the target cell; wherein, at least one of the access devices includes the terminal device, and the downlink control message is used to instruct each access device to report second terminal application data; Receiving the second terminal application data reported by each access device; Determining second network environment data when each second terminal application data is received; Storing each second terminal application data and the corresponding second network environment data.

5. The method according to any one of claims 1 to 4, characterized in that, The method further includes: Sending a set of terminal application data and a set of network environment data corresponding to the target cell stored according to a preset period to the management node.

6. The method according to claim 1, wherein The determining a trained resource scheduling model includes: Receiving a reference scheduling model sent by the management node; Updating the resource scheduling model to the reference scheduling model.

7. A resource scheduling method, characterized in that, The method is applied to a management node, and the method includes: If an update instruction for updating the model is detected, using a set of terminal application data and a set of network environment parameters of a target cell sent by a resource scheduling node to perform model training and updating on a corresponding resource scheduling model to obtain a reference scheduling model; wherein, the set of terminal application data at least includes values of evaluation parameters of at least one network service provided; Send the reference scheduling model to the resource scheduling node; wherein, the reference scheduling model is used to enable the resource scheduling node to predict the target resource scheduling parameters for network scheduling of the target cell.

8. The method according to claim 7, wherein The method further includes: Obtain the sample data to be trained; wherein, the sample data to be trained includes historical terminal application data and historical network environment data in units of cells, and the historical terminal application data at least includes historical usage experience evaluation values for network services. Use the sample data to be trained to train the model to be trained, and obtain the resource scheduling model; wherein, the model to be trained is used to determine the optimal network environment parameters corresponding to the highest usage experience evaluation value in each category after classifying the data according to features.

9. A resource scheduling method, characterized in that, The method is applied to a terminal device, and the method includes: Obtain the key network performance indicators corresponding to the target service currently in use. Based on the key network performance indicators, determine the first terminal application data. Send the first terminal application data to the resource scheduling node, so that the resource scheduling node predicts the network resources of the target cell currently accessed based on the first terminal application data.

10. The method according to claim 9, wherein The determining the first terminal application data based on the key network performance indicators includes: Determine the score corresponding to each indicator in the key network performance indicators. Determine the second weight coefficient corresponding to each indicator in the key network performance indicators. Perform weighted calculation on the score corresponding to each indicator and the corresponding second weight coefficient to obtain the usage experience evaluation value of the target service. Determine that the first terminal application data includes the usage experience evaluation value and the key network performance indicators.

11. A first resource scheduling device, characterized in that, The device is applied to a resource scheduling node, and the device includes: a receiving unit, a first obtaining unit, a first determining unit, a predicting unit, and an executing unit; wherein: The receiving unit is configured to receive the first terminal application data based on the target service sent by the terminal device; wherein, the first terminal application data at least includes the usage experience evaluation value of the target service. The first obtaining unit is configured to obtain the current first network environment data. The first determining unit is configured to determine the trained resource scheduling model; wherein, the resource scheduling model is obtained by model training at the management node for the target cell where the terminal device is located. The predicting unit is configured to input the identification information of the target service, the first terminal application data, and the first network environment data into the first resource scheduling model for prediction, and obtain the target resource scheduling parameters; wherein, the target resource scheduling parameters correspond to the target resources of the target service of the terminal device. The executing unit is configured to perform a network scheduling operation for the terminal device based on the target resource scheduling parameters.

12. A second resource scheduling device, characterized in that, The device is applied to a management node, and the device includes: an updating unit and a first sending unit; wherein: The updating unit is configured to, if an update instruction for updating the model is detected, perform model training and updating on the corresponding resource scheduling model by using the terminal application data set and the network environment parameter set of the target cell sent by the resource scheduling node, so as to obtain a reference scheduling model; wherein, the terminal application data set includes at least the values of the evaluation parameters of at least one provided network service. The first sending unit is configured to send the reference scheduling model to the resource scheduling node; wherein, the reference scheduling model is used to enable the resource scheduling node to predict the target resource scheduling parameters for network scheduling of the target cell.

13. A third resource scheduling device, characterized in that, The apparatus is applied to a terminal device, and the apparatus includes: a second obtaining unit, a second determining unit, and a second sending unit; wherein: The second obtaining unit is configured to obtain the key network performance indicators corresponding to the target service currently used. The second determining unit is configured to determine the first terminal application data based on the key network performance indicators. The second sending unit is configured to send the first terminal application data to the resource scheduling node, so that the resource scheduling node predicts the network resources of the target cell currently accessed based on the first terminal application data.

14. A resource scheduling system, characterized in that, The system includes a management node, at least one resource scheduling node, and at least one access device including a terminal device; wherein: The management node is configured to implement the steps of the resource scheduling method according to any one of claims 9 to 10. The resource scheduling node is configured to implement the steps of the resource scheduling method according to any one of claims 1 to 6. The terminal device is configured to implement the steps of the resource scheduling method according to any one of claims 7 to 8.

15. A storage medium, characterized in that, A resource scheduling program is stored on the storage medium, and when the resource scheduling program is executed by a processor, the steps of the resource scheduling method according to any one of claims 1 to 6, or claims 7 to 8, or claims 9 to 10 are implemented.

Images