Resource scheduling method, device and electronic equipment

Abstract

The embodiment of the present invention discloses a resource scheduling method, device and electronic device, the method comprising: obtaining a service data stream generated by a target user device executing a target service, and identifying the user behavior corresponding to the service data stream based on a pre-trained service identification model to obtain the target user behavior corresponding to the service data stream; determining the quality detection result of the target service based on preset key quality indicators and the service data corresponding to the target user behavior in the service data stream; if the quality detection result does not meet the preset quality requirement conditions, determining a parameter adjustment strategy for the target service based on the quality detection result, and executing the parameter adjustment strategy to perform resource scheduling for the target service. Through this method, targeted network quality detection and parameter adjustment can be performed to improve the network optimization effect and user experience for the target service.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a resource scheduling method, device and electronic equipment. Background Art

[0002] As an emerging service, data services have penetrated into all areas of people's work and life and become an important service on mobile phones. The number of users of some specific services has shown a rapid growth, and network optimization issues for these specific services have become a focus of operators.

[0003] Currently, based on user complaints about poor network quality when using specific services, the network parameters and configuration parameters on the wireless side can be obtained from the Operation Maintenance Center (OMC) on the network management side. Based on the obtained data, poor quality cells can be identified and a network optimization plan can be formulated based on the obtained data to achieve network optimization for executing specific services in poor quality cells.

[0004] However, the network quality requirements may be different for different specific services. For example, for gaming services, the download rate requirement is not high, but the packet transmission delay requirement is high. For video services, the download rate requirement is high, and the packet transmission delay requirement is low. Moreover, the network quality requirements may also be different for different operations in the same type of service. Therefore, network optimization for executing specific services based only on network parameters and configuration parameters will result in poor optimization effect and poor user experience. Summary of the Invention

[0005] The purpose of the embodiments of the present invention is to provide a resource scheduling method, device and electronic device to solve the problems of poor optimization effect and poor user experience when performing network optimization for specific services in the prior art.

[0006] To solve the above technical problems, the embodiments of the present invention are implemented as follows:

[0007] In a first aspect, an embodiment of the present invention provides a resource scheduling method, the method comprising:

[0008] Obtaining a service data stream generated by a target user device executing a target service, and identifying user behavior corresponding to the service data stream based on a pre-trained service identification model to obtain a target user behavior corresponding to the service data stream, wherein the service identification model is obtained by training a model constructed using a preset machine learning algorithm based on historical service data streams of the target service;

[0009] Determining a quality detection result of the target service based on preset key quality indicators and service data corresponding to the target user behavior in the service data stream;

[0010] In a case where the quality detection result does not meet the preset quality requirement condition, a parameter adjustment strategy for the target service is determined based on the quality detection result, and the parameter adjustment strategy is executed to perform resource scheduling for the target service.

[0011] Optionally, determining the quality detection result of the target service based on preset key quality indicators and service data corresponding to the target user behavior in the service data stream includes:

[0012] Determining a first sub-quality detection result of the target service based on preset key quality indicators and service data corresponding to the target user behavior in the service data stream;

[0013] Obtaining network parameter information for executing the target service, and determining the data transmission status during the execution of the target service based on a pre-trained data transmission status prediction model and the network parameter information, wherein the data transmission status prediction model is obtained by training a model constructed using a preset machine learning algorithm based on historical network parameter information of the target service;

[0014] The quality detection result is determined based on the first sub-quality detection result and the data transmission status.

[0015] Optionally, before identifying the user behavior corresponding to the service data flow based on the pre-trained service identification model to obtain the target user behavior corresponding to the service data flow, the method further includes:

[0016] Acquire historical business data streams for executing the target business;

[0017] Based on the protocol type corresponding to the historical service data contained in the historical service data stream, the historical service data stream is divided into one or more data classes, each of the data classes contains one or more historical service data, and different data classes correspond to different protocol types;

[0018] Obtaining a sub-service identification model corresponding to the data class, and training the sub-service identification model based on historical service data contained in the data class to obtain the trained sub-service identification model corresponding to the data class;

[0019] The pre-trained service identification model is used to identify the user behavior corresponding to the service data flow to obtain the target user behavior corresponding to the service data flow, including:

[0020] Determining the data class to which the service data belongs based on a protocol type corresponding to the service data contained in the service data stream;

[0021] Inputting the business data into the sub-business identification model corresponding to the data class to which the business data belongs for identification, thereby obtaining target sub-user behavior;

[0022] Based on the target sub-user behavior, a target user behavior corresponding to the service data flow is determined.

[0023] Optionally, determining the first sub-quality detection result of the target service based on a preset key quality indicator and service data corresponding to the target user behavior in the service data stream includes:

[0024] Obtaining a quality threshold and a preset weight corresponding to each of the key quality indicators;

[0025] A first sub-quality detection result of the target service is determined based on the service data corresponding to the target user behavior, the quality threshold corresponding to the key quality indicator, and a preset weight.

[0026] Optionally, determining the first sub-quality detection result of the target service based on the service data corresponding to the target user behavior, the quality threshold corresponding to the key quality indicator, and a preset weight includes:

[0027] Determining a quality detection score corresponding to each key quality indicator based on the business data corresponding to the target user behavior, a quality threshold corresponding to each key quality indicator, and a preset weight;

[0028] Substitute the quality inspection score corresponding to each key quality indicator into the formula

[0029]

[0030] Get the target quality score corresponding to each key quality indicator, where X is the target quality score corresponding to the key quality indicator, X test is the quality detection score corresponding to the key quality indicator, I is the first preset score, Q is the second preset score, and T is the third preset score;

[0031] Based on the target quality score corresponding to each key quality indicator, a first sub-quality detection result of the target service is determined.

[0032] Optionally, before determining the data transmission status during the execution of the target service based on the pre-trained data transmission status prediction model and the network parameter information, the method further includes:

[0033] Acquire historical network parameter information during the execution of the target service;

[0034] The preset recurrent neural network model is trained based on the historical network parameter information to obtain the data transmission situation prediction model.

[0035] Optionally, determining a parameter adjustment strategy for the target service based on the quality detection result includes:

[0036] Determining a first quality detection result of the target service based on the service data flow and preset key performance indicators;

[0037] Based on the first quality detection result and the quality detection result, a parameter adjustment strategy for the target service is determined, where the parameter adjustment strategy includes at least one or more strategies of a cell failure optimization strategy, a service balancing adjustment strategy, a carrier scheduling strategy, and a planning adjustment strategy.

[0038] In a second aspect, an embodiment of the present invention provides a resource scheduling device, the device comprising:

[0039] A first acquisition module is configured to acquire a service data stream generated by a target user device executing a target service, and identify user behavior corresponding to the service data stream based on a pre-trained service identification model to obtain a target user behavior corresponding to the service data stream, wherein the service identification model is obtained by training a model constructed using a preset machine learning algorithm based on historical service data streams of the target service;

[0040] A first detection module is configured to determine a quality detection result of the target service based on preset key quality indicators and service data corresponding to the target user behavior in the service data stream;

[0041] The parameter adjustment module is used to determine a parameter adjustment strategy for the target business based on the quality detection result when the quality detection result does not meet the preset quality requirement condition, and execute the parameter adjustment strategy to schedule resources for the target business.

[0042] Optionally, the first detection module is configured to:

[0043] Determining a first sub-quality detection result of the target service based on preset key quality indicators and service data corresponding to the target user behavior in the service data stream;

[0044] Obtaining network parameter information for executing the target service, and determining the data transmission status during the execution of the target service based on a pre-trained data transmission status prediction model and the network parameter information, wherein the data transmission status prediction model is obtained by training a model constructed using a preset machine learning algorithm based on historical network parameter information of the target service;

[0045] The quality detection result is determined based on the first sub-quality detection result and the data transmission status.

[0046] Optionally, the device further includes:

[0047] A second acquisition module is used to acquire a historical business data stream for executing the target business;

[0048] a classification module, configured to divide the historical service data stream into one or more data classes based on a protocol type corresponding to the historical service data contained in the historical service data stream, each data class containing one or more historical service data, and different data classes corresponding to different protocol types;

[0049] a first training module, configured to obtain a sub-service identification model corresponding to the data class, and train the sub-service identification model based on historical service data contained in the data class to obtain the trained sub-service identification model corresponding to the data class;

[0050] The first acquisition module is configured to:

[0051] Determining the data class to which the service data belongs based on a protocol type corresponding to the service data contained in the service data stream;

[0052] Inputting the business data into the sub-business identification model corresponding to the data class to which the business data belongs for identification, thereby obtaining target sub-user behavior;

[0053] Based on the target sub-user behavior, a target user behavior corresponding to the service data flow is determined.

[0054] Optionally, the first detection module is configured to:

[0055] Obtaining a quality threshold and a preset weight corresponding to each of the key quality indicators;

[0056] A first sub-quality detection result of the target service is determined based on the service data corresponding to the target user behavior, the quality threshold corresponding to the key quality indicator, and a preset weight.

[0057] Optionally, the first detection module is configured to:

[0058] Determining a quality detection score corresponding to each key quality indicator based on the business data corresponding to the target user behavior, a quality threshold corresponding to each key quality indicator, and a preset weight;

[0059] Substitute the quality inspection score corresponding to each key quality indicator into the formula

[0060]

[0061] Get the target quality score corresponding to each key quality indicator, where X is the target quality score corresponding to the key quality indicator, X test is the quality detection score corresponding to the key quality indicator, I is the first preset score, Q is the second preset score, and T is the third preset score;

[0062] Based on the target quality score corresponding to each key quality indicator, a first sub-quality detection result of the target service is determined.

[0063] Optionally, the device further includes:

[0064] A third acquisition module is used to obtain historical network parameter information during the execution of the target service;

[0065] The second training module is used to train the preset recurrent neural network model based on the historical network parameter information to obtain the data transmission situation prediction model.

[0066] Optionally, the parameter adjustment module is used to:

[0067] Determining a first quality detection result of the target service based on the service data flow and preset key performance indicators;

[0068] Based on the first quality detection result and the quality detection result, a parameter adjustment strategy for the target service is determined, wherein the parameter adjustment strategy includes at least one or more strategies selected from a cell failure optimization strategy, a service balancing adjustment strategy, a carrier scheduling strategy, and a planning adjustment strategy. In a third aspect, an embodiment of the present invention provides an electronic device comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein when the computer program is executed by the processor, the steps of the resource scheduling method provided in the above embodiment are implemented.

[0069] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the resource scheduling method provided in the above embodiment are implemented.

[0070] It can be seen from the technical solution provided by the above embodiments of the present invention that the embodiments of the present invention obtain the service data stream generated by the target user device executing the target service, and identify the user behavior corresponding to the service data stream based on the pre-trained service identification model to obtain the target user behavior corresponding to the service data stream. The service identification model is obtained by training a model constructed by a preset machine learning algorithm based on the historical service data stream of the target service. The quality detection result of the target service is determined based on the preset key quality indicators and the service data corresponding to the target user behavior in the service data stream. If the quality detection result does not meet the preset quality requirement conditions, the parameter adjustment strategy for the target service is determined based on the quality detection result, and the parameter adjustment strategy is executed to perform resource scheduling for the target service. In this way, the quality detection result for the target service can be determined based on the identified target user behavior of the target service and the corresponding service data. If the quality result does not meet the preset quality requirement conditions, the parameter adjustment strategy is determined based on the quality detection result, that is, the target user behavior of the target service can be targeted for parameter adjustment to achieve good network optimization effect and improve user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0072] Figure 1 A schematic diagram of a resource scheduling method according to the present invention;

[0073] Figure 2 A schematic diagram of another resource scheduling method of the present invention;

[0074] Figure 3 This is a structural diagram of a resource scheduling device according to the present invention;

[0075] Figure 4 The figure is a schematic structural diagram of an electronic device of the present invention. DETAILED DESCRIPTION

[0076] Embodiments of the present invention provide a resource scheduling method, device, and electronic device.

[0077] In order to enable those skilled in the art to better understand the technical solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0078] Example 1

[0079] like Figure 1 As shown, an embodiment of the present invention provides a resource scheduling method for implementing resource scheduling between cells corresponding to different base stations. The execution subject of the method can be a server, which can be an independent server or a server cluster composed of multiple servers. The method can specifically include the following steps:

[0080] In S102, a service data flow generated by a target user device executing a target service is obtained, and user behavior corresponding to the service data flow is identified based on a pre-trained service identification model to obtain a target user behavior corresponding to the service data flow.

[0081] Among them, the business identification model can be obtained by training a model constructed by a preset machine learning algorithm based on the historical business data flow of the target business. The target device can be any device that can execute the target business. The target business can be any business that the target user can use. For example, the target user device can be a mobile terminal device such as a mobile phone, tablet computer, or a terminal device such as a personal computer. The target business can be the business corresponding to the application installed in the target user device or the executable business in the application. For example, the target business can be the business corresponding to the game application installed in the target user device. The business data flow generated by executing the target business can be composed of a series of business data (such as traffic data, etc.) generated by executing the target business. The user behavior can be the user operation received by the target user device during the execution of the target business.

[0082] In practice, data services, as an emerging service, have penetrated into all areas of people's work and life, becoming a key feature of mobile phones. The number of users of certain specific services has seen explosive growth, making network optimization for these services a key concern for operators. Currently, based on user complaints about poor network quality when using specific services, operators can obtain wireless network parameters and configuration parameters from the Operation Maintenance Center (OMC) on the network management side. This data can then be used to identify cells with poor quality and develop network optimization plans based on this data to optimize the network for specific services in these cells.

[0083] However, for different specific services, the network quality requirements may be different. For example, for gaming services, the download rate requirement is not high, but the packet transmission delay requirement is high. For video services, the download rate requirement is high, but the packet transmission delay requirement is low. Moreover, for different operations in the same type of service, the network quality requirements may also be different. Therefore, if the network is optimized for executing specific services based only on network parameters and configuration parameters, there will be problems such as poor optimization effect and poor user experience. To this end, the embodiment of the present invention provides a technical solution that can solve the above problems, which may specifically include the following contents:

[0084] For example, if the target user device is a mobile phone, the target user can launch a game application installed on the phone and perform operations such as logging in and playing games within the game application. The server can receive the traffic data generated by the target user device performing these operations. The traffic data flow obtained by the server within a preset time is the service data flow generated by the target user device performing the target service.

[0085] After obtaining the business data flow generated by executing the target business, behavior recognition can be performed on the business data flow based on a pre-trained business recognition model to obtain the target user behavior corresponding to the business data flow. For example, the identified target user behavior may include the login operation, battle operation, etc. performed by the target user on the target user device.

[0086] Since not all operations during the user's use of the target service will directly affect the user's experience, for example, when using gaming services, it is the battle operations that affect the user's perception. The long delays in login and matching operations do not significantly affect the user's perception of the service. Therefore, based on DPI-based parsing and identification of the code stream, through artificial intelligence machine recognition technology, the business data stream generated during the execution of the target service can be accurately obtained to achieve refined identification of the target service. In addition, in the identification process, adding understanding and statistical word segmentation methods can make the identification more accurate and achieve in-depth identification of the target service, including the identification of various target user behaviors, such as login and battle operations in gaming services, login and video playback operations in video services, and other precise action recognition. This can effectively identify target user behaviors that directly affect user perception, provide basic data support for determining the quality inspection results of the target service, and be more targeted.

[0087] In S104 , a quality detection result of the target service is determined based on preset key quality indicators and service data corresponding to the target user behavior in the service data stream.

[0088] Among them, the preset key quality indicators (KQI) may include indicators such as the number of freezes and the duration of freezes.

[0089] In practice, key performance indicators (KPIs) can only reflect the performance of network equipment (i.e., the control plane). Often, excellent KPIs may result in poor service quality. Therefore, by focusing on KQIs for both the user and control planes, quality testing results can be determined to better reflect the user experience.

[0090] Different KQIs can be preset according to the target business. For example, for gaming businesses, the most direct KQI indicators that affect user perception of competitive games may include the number of freezes, freeze duration, and battle delay. For video businesses, the most direct KQI indicators that affect user perception of video services may include the initial video buffering time, the number of freezes, and freeze duration.

[0091] After identifying the target user behavior corresponding to the target business, the KQI corresponding to the target business can be obtained, and the business data corresponding to the KQI can be extracted from the business data corresponding to the target user behavior. The quality detection result of the target business can be determined based on the extracted business data.

[0092] For example, KQI includes the number of freezes, and the target user behavior is a login operation. The business data corresponding to the login operation can be obtained, and the number of freezes can be extracted from the business data corresponding to the login operation. Based on the relationship between the number of freezes and the preset freeze threshold, the quality detection result of the login operation is determined (if the number of freezes is not less than the preset freeze threshold, the quality detection result is poor; if the number of freezes is less than the preset freeze threshold, the quality detection result is excellent). For example, if the number of freezes in the login operation is 5 and the preset freeze threshold is 2, the quality detection result of the login operation is poor.

[0093] Furthermore, if there are multiple target user behaviors corresponding to a target service, the quality test results corresponding to each target user behavior can be obtained, and the quality test result of the target service can be determined based on the quality test results of each target user behavior. For example, if there are four target user behaviors corresponding to a target service, and the quality test results of three of them are excellent, and the quality test result of one target user behavior is poor, the quality test result of the target service can be determined to be excellent.

[0094] The above-mentioned method for determining the quality detection result of the target service is an optional and feasible determination method. In actual application scenarios, there may be a variety of different determination methods, which may vary according to different actual application scenarios. The embodiments of the present invention do not make specific limitations thereto.

[0095] In S106 , when the quality detection result does not meet the preset quality requirement condition, a parameter adjustment strategy for the target service is determined based on the quality detection result, and the parameter adjustment strategy is executed to perform resource scheduling for the target service.

[0096] During implementation, if the quality test result does not meet the preset quality requirement, a parameter adjustment policy corresponding to the quality test result of the target service can be determined based on a preset correspondence between the quality test result and the parameter adjustment policy, and the parameter adjustment policy can be executed. The preset correspondence between the quality test result and the parameter adjustment policy can be determined based on historical quality test results and historical parameter adjustment policies.

[0097] The above-mentioned method for determining the parameter adjustment strategy of the target business is an optional and feasible determination method. In actual application scenarios, there may be a variety of different determination methods, which may vary according to the actual application scenarios. The embodiments of the present invention do not make specific limitations accordingly.

[0098] An embodiment of the present invention provides a resource scheduling method, which obtains a service data stream generated by a target user device executing a target service, and identifies the user behavior corresponding to the service data stream based on a pre-trained service identification model to obtain the target user behavior corresponding to the service data stream. The service identification model is obtained by training a model constructed by a preset machine learning algorithm based on the historical service data stream of the target service. The quality detection result of the target service is determined based on preset key quality indicators and service data corresponding to the target user behavior in the service data stream. If the quality detection result does not meet the preset quality requirement conditions, a parameter adjustment strategy for the target service is determined based on the quality detection result, and the parameter adjustment strategy is executed to perform resource scheduling for the target service. In this way, the quality detection result for the target service can be determined based on the identified target user behavior of the target service and the corresponding service data. If the quality result does not meet the preset quality requirement conditions, a parameter adjustment strategy is determined based on the quality detection result, that is, targeted parameter adjustment can be performed for the target user behavior of the target service to achieve good network optimization effects and improve user experience.

[0099] Example 2

[0100] like Figure 2 As shown, an embodiment of the present invention provides a resource scheduling method for implementing resource scheduling between cells corresponding to different base stations. The execution subject of the method can be a server, which can be an independent server or a server cluster composed of multiple servers. The method can specifically include the following steps:

[0101] In S202, a historical business data flow of executing a target business is obtained.

[0102] To improve data processing efficiency during implementation, automated testing tools can be used to simulate the target business execution process and capture the corresponding historical business data stream. However, since the automated testing tools themselves are very limited in their functionality and scope of application, secondary development can be conducted based on the automated testing tools to capture the business data stream generated by the target business execution. For example, a list of all clickable controls in the target business execution process can be used to perform sequential click operations. Targeted secondary development can also be conducted based on the differences in the execution process of different target businesses, such as whether a welcome page needs to be skipped by sliding or clicking.

[0103] In addition, the historical service data stream may also be pre-stored historical data generated by the user equipment during the execution of the target service. The embodiment of the present invention does not specifically limit the method for obtaining the historical service data stream.

[0104] In S204 , the historical service data stream is divided into one or more data classes based on the protocol type corresponding to the historical service data contained in the historical service data stream.

[0105] Each data class may contain one or more historical business data, and different data classes correspond to different protocol types.

[0106] In practice, cluster analysis can be performed on historical service data streams to classify them into one or more data classes. For example, first, the service data in the historical service data streams can be divided into encrypted data and non-encrypted data by counting the bytes in the payload of the historical service data streams. For example, the payloads in the first n packets of a connection can be counted to obtain the probability distribution of the bytes, and the mean square error of the probability distribution can be calculated to determine whether the connection is encrypted.

[0107] Then, the unencrypted historical business data can be obtained and clustered based on the protocol type. For example, clustering can be performed based on tuple, payload, pattern, and timestamp respectively. The result of clustering is one or more data classes.

[0108] In addition, there may be some historical business data in the historical business data stream that has not been clustered into one or more of the above data classes. The data class to which the unclustered historical business data belongs can be determined based on the similarity between the historical business data and the historical business data contained in each data class.

[0109] In S206 , a sub-business identification model corresponding to the data class is obtained, and the sub-business identification model is trained based on the historical business data included in the data class to obtain a trained sub-business identification model corresponding to the data class.

[0110] In implementation, the sub-business identification model corresponding to each data class can be determined based on the protocol type corresponding to the data class. For example, the sub-business identification model corresponding to the data class corresponding to the TCP / UDP protocol can be a clustering model built based on sequence pattern mining; the sub-business identification model corresponding to the data class corresponding to the HTTP protocol can be a classification model built based on a decision tree.

[0111] For example, for a data class corresponding to the TCP / UDP protocol, we can obtain the historical business data contained in that data class and, based on a specified number of bytes before and after the payload in the obtained historical business data, perform clustering using a sequential pattern mining approach. This involves first finding all frequent item sets (the frequency of these item sets is at least the same as a predefined minimum support), and then generating strong association rules from the frequent item sets (these rules must meet a preset support and a preset confidence level). The determined strong association rules can then serve as the sub-business identification model corresponding to that data class.

[0112] For the data class corresponding to HTTP, the historical business data contained in the data class can be obtained, and then the header field of the HTTP data packet in the historical business data can be input into the pre-built decision tree model. Among them, the decision tree model can use the priority defined by the mobile feature library (for example, the priority of HOST is higher than that of User-Agent, and the priority of User-Agent is higher than that of URL, etc.) to split the tree nodes. In this way, the root node to the leaf node of the decision tree constitutes a complete HTTP rule.

[0113] In S208 , a service data flow generated by the target user equipment executing the target service is obtained.

[0114] In S210 , the data class to which the service data belongs is determined based on the protocol type corresponding to the service data contained in the service data flow.

[0115] In implementation, the service data stream may include multiple service data corresponding to different protocol types, and the data class to which each service data in the service data stream belongs can be obtained separately.

[0116] In S212 , the business data is input into a sub-business identification model corresponding to the data class to which the business data belongs for identification, thereby obtaining target sub-user behavior.

[0117] In S214 , based on the target sub-user behavior, the target user behavior corresponding to the service data flow is determined.

[0118] In implementation, the target user behavior may include one or more target sub-user behaviors.

[0119] In S216 , a first sub-quality detection result of the target service is determined based on the preset key quality indicators and the service data corresponding to the target user behavior in the service data stream.

[0120] In practical applications, the processing method of the above S216 can be various. An optional implementation method is provided below. For details, please refer to the following steps 1 and 5.

[0121] Step 1: Obtain the quality threshold and preset weight corresponding to each key quality indicator.

[0122] Step 2: Determine a first sub-quality detection result of the target service based on service data corresponding to the target user behavior, quality thresholds corresponding to key quality indicators, and preset weights.

[0123] Step three: Determine the quality detection score corresponding to each key quality indicator based on the business data corresponding to the target user behavior, the quality threshold corresponding to each key quality indicator, and the preset weight.

[0124] Step 4: Substitute the quality inspection score corresponding to each key quality indicator into the formula

[0125]

[0126] Get the target quality score corresponding to each key quality indicator, where X is the target quality score corresponding to the key quality indicator, X test is the quality detection score corresponding to the key quality indicator, I is the first preset score, Q is the second preset score, and T is the third preset score.

[0127] Among them, I, Q and T can be used to represent the satisfaction of executing the target business, among which I represents a higher satisfaction than Q, and Q represents a higher satisfaction than T. The specific scores of I, Q and T can vary depending on the target business or target user behavior, and the embodiment of the present invention does not make specific limitations on this.

[0128] Step 5: Based on the target quality score corresponding to each key quality indicator, determine the first sub-quality detection result of the target business.

[0129] In practice, by determining the first sub-quality detection result of the target service according to the above method, it is possible to achieve differentiated and accurate quality assessment for the target service, improve the accuracy of quality detection and the accuracy of network optimization. sex.

[0130] In S218, historical network parameter information during the execution of the target service is obtained.

[0131] During implementation, the packet capture data (pcap) generated by executing the target service can be collected through a preset interface (such as the S1-U interface), and the historical network parameter information during the execution of the target service can be determined based on the packet capture data, such as cell capacity utilization, perUE cache status, RB occupancy, etc.

[0132] In S220, a preset recurrent neural network model is trained based on historical network parameter information to obtain a data transmission situation prediction model.

[0133] During implementation, the data transmission situation of the target business is highly correlated with the base station scheduling. Factors such as network congestion and poor channel conditions will affect the data transmission situation of the target business. These influencing factors are not instantaneous, but rather a gradual process. During this process, the data transmission situation will change continuously and regularly.

[0134] Therefore, these changes can be learned through machine learning to predict the data transmission required to ensure that the target user equipment performs the target service.

[0135] Before model training, historical network parameter information and statistical results can be associated through timestamps, and the label of each frame of historical network parameter information can be divided into stuck or non-stuck. The frames between the start and end positions of the stuck time when executing the target business are recorded as stuck frames with a label of 1, and the frames in the non-stuck time are recorded as smooth frames with a label of 0.

[0136] The labeled historical network parameter information is then fed into a pre-set recurrent neural network (LSTM) model for training. This allows the prediction of the lag state (i.e., data transmission status) at time t+1 using the historical network parameter information from the previous t time periods. Furthermore, a fully connected network can be added later to fine-tune the LSTM output and compensate for errors.

[0137] In S222, network parameter information for executing the target service is obtained, and data transmission conditions during the execution of the target service are determined based on a pre-trained network parameter information prediction model and the network parameter information.

[0138] In S224, a quality detection result is determined based on the first sub-quality detection result and the data transmission status.

[0139] In implementation, the first sub-quality detection result can be reflected by a quality score, and the data transmission situation can be reflected by a jamming situation. Therefore, when the first sub-quality detection result is less than a preset quality score threshold and the data transmission situation is jamming, it can be determined that the quality detection result is poor user perception and poor data transmission.

[0140] The above-mentioned method for determining the quality inspection results is an optional and feasible determination method. In actual application scenarios, there may be a variety of different determination methods, which may vary according to different actual application scenarios. The embodiment of the present invention does not make any specific limitations on this.

[0141] In S226 , when the quality detection result does not meet the preset quality requirement condition, a first quality detection result of the target service is determined based on the service data flow and the preset key performance indicators.

[0142] In implementation, the preset key performance indicators may include TCP connection success rate, perUE cache status, RB occupancy rate, etc. When the quality detection result does not meet the preset quality requirement conditions, the first quality detection result can be determined based on the preset key performance indicators and business data flow. For example, the first quality detection result can be determined by determining the corresponding performance indicator score based on the indicator threshold and the corresponding indicator weight.

[0143] In S228 , a parameter adjustment strategy for the target service is determined based on the first quality detection result and the quality detection result.

[0144] The parameter adjustment strategy includes at least one or more strategies among a cell failure optimization strategy, a service balancing adjustment strategy, a carrier scheduling strategy and a planning adjustment strategy.

[0145] During implementation, wireless statistical indicators such as cell-level interference statistics, capacity statistics, and traffic statistics can be obtained from the wireless network management side. Based on the first quality detection result and the quality detection result, the corresponding network parameters that need to be adjusted are determined, and based on the determined network parameters, a parameter adjustment strategy for the target service is constructed.

[0146] In addition, network optimization can be performed for target services through cell fault optimization strategies, service balancing adjustment strategies, carrier scheduling strategies, and planning adjustment strategies.

[0147] Among them, the cell fault optimization strategy can be to check whether there is any business that affects the execution of the target business at the current site, and then check whether there is any business that affects the execution of the target business in the neighboring area. According to the verification results, it is analyzed to determine whether there is a fault at the site. If there is a fault, a maintenance task work order is generated to the maintenance group to open up the interface between the maintenance group and the branch and comprehensive maintenance agency, so as to realize closed-loop processing of the fault.

[0148] The service balancing adjustment strategy can be to check whether the services of the same remote radio unit (RRU) in the cell and different frequencies are balanced, whether the services of different RRUs in the same direction are balanced, etc., and take corresponding service load balancing adjustments based on the verification results. Among them, if the services of adjacent cells are unbalanced, the switching threshold is used to control the service flow.

[0149] The carrier scheduling strategy can be to extract the load index of the cell to verify whether the cell is under high load. For example, it can check whether the hardware is faulty. If the hardware is not faulty, the data is re-planned (such as re-planning capacity expansion, adjusting switching parameters, coverage parameters, etc.) to reduce the cell load.

[0150] The planning adjustment strategy may be to adjust hardware replacement, add new sites, etc. when it is detected that the resource configuration of the cell and the neighboring cells is fully allocated.

[0151] In S230 , the parameter adjustment strategy is executed to perform resource scheduling for the target service.

[0152] In practice, since network resources are fixedly allocated to users and will not be released based on whether users use the target service, this will result in resource waste. For example, shopping malls have high resource requirements during the day and very low resource requirements at night. How to dynamically adjust network resources to meet the usage needs of the target service is very critical.

[0153] Based on data transmission conditions and statistical indicators from the wireless network, a base station scheduling algorithm can be configured to determine a parameter adjustment strategy for the target service. This parameter adjustment strategy can be implemented to intelligently adjust network parameters, achieving intelligent network optimization. Wireless statistical indicators such as cell-level interference, capacity, and traffic statistics can be obtained from the wireless network management system. Various network parameters can be extracted and compared with the statistical indicators and parameters. The relationship between network quality and traffic volume corresponding to different parameter settings can be analyzed to determine the parameter adjustment strategy.

[0154] After receiving the parameter adjustment policy for the target business, the intelligent dispatching center can automatically issue relevant processes according to the parameter adjustment policy and automatically perform parameter adjustment operations such as parameter tuning and dynamic scheduling of wireless resources.

[0155] First, it analyzes whether there are cell faults, then analyzes whether there are imbalances, and determines whether carrier scheduling conditions are met. If so, resource scheduling is performed. If not, optimization work orders are automatically generated based on different scenarios for optimization and adjustment. This improves resource utilization, specifically enhances user network experience, and helps retain existing customers and attract new ones.

[0156] An embodiment of the present invention provides a resource scheduling method, which obtains a service data stream generated by a target user device executing a target service, and identifies the user behavior corresponding to the service data stream based on a pre-trained service identification model to obtain the target user behavior corresponding to the service data stream. The service identification model is obtained by training a model constructed by a preset machine learning algorithm based on the historical service data stream of the target service. The quality detection result of the target service is determined based on preset key quality indicators and service data corresponding to the target user behavior in the service data stream. If the quality detection result does not meet the preset quality requirement conditions, a parameter adjustment strategy for the target service is determined based on the quality detection result, and the parameter adjustment strategy is executed to perform resource scheduling for the target service. In this way, the quality detection result for the target service can be determined based on the identified target user behavior of the target service and the corresponding service data. If the quality result does not meet the preset quality requirement conditions, a parameter adjustment strategy is determined based on the quality detection result, that is, targeted parameter adjustment can be performed for the target user behavior of the target service to achieve good network optimization effects and improve user experience.

[0157] Example 3

[0158] The above is a resource scheduling method provided by an embodiment of the present invention. Based on the same idea, an embodiment of the present invention also provides a resource scheduling device, such as Figure 3 shown.

[0159] The resource scheduling device includes: a first acquisition module 301, a first detection module 302 and a parameter adjustment module 303, wherein:

[0160] A first acquisition module 301 is configured to acquire a service data stream generated by a target user device executing a target service, and identify user behavior corresponding to the service data stream based on a pre-trained service identification model to obtain a target user behavior corresponding to the service data stream. The service identification model is obtained by training a model constructed using a preset machine learning algorithm based on historical service data streams of the target service.

[0161] A first detection module 302 is configured to determine a quality detection result of the target service based on preset key quality indicators and service data corresponding to the target user behavior in the service data stream;

[0162] The parameter adjustment module 303 is used to determine a parameter adjustment strategy for the target service based on the quality detection result when the quality detection result does not meet the preset quality requirement condition, and execute the parameter adjustment strategy to perform resource scheduling for the target service.

[0163] In this embodiment of the present invention, the first detection module 302 is configured to:

[0164] Determining a first sub-quality detection result of the target service based on preset key quality indicators and service data corresponding to the target user behavior in the service data stream;

[0165] Obtaining network parameter information for executing the target service, and determining the data transmission status during the execution of the target service based on a pre-trained data transmission status prediction model and the network parameter information, wherein the data transmission status prediction model is obtained by training a model constructed using a preset machine learning algorithm based on historical network parameter information of the target service;

[0166] The quality detection result is determined based on the first sub-quality detection result and the data transmission status.

[0167] In an embodiment of the present invention, the apparatus further includes:

[0168] A second acquisition module is used to acquire a historical business data stream for executing the target business;

[0169] a classification module, configured to divide the historical service data stream into one or more data classes based on a protocol type corresponding to the historical service data contained in the historical service data stream, each data class containing one or more historical service data, and different data classes corresponding to different protocol types;

[0170] a first training module, configured to obtain a sub-service identification model corresponding to the data class, and train the sub-service identification model based on historical service data contained in the data class to obtain the trained sub-service identification model corresponding to the data class;

[0171] The first acquisition module 301 is configured to:

[0172] Determining the data class to which the service data belongs based on a protocol type corresponding to the service data contained in the service data stream;

[0173] Inputting the business data into the sub-business identification model corresponding to the data class to which the business data belongs for identification, thereby obtaining target sub-user behavior;

[0174] Based on the target sub-user behavior, a target user behavior corresponding to the service data flow is determined.

[0175] In this embodiment of the present invention, the first detection module 302 is configured to:

[0176] Obtaining a quality threshold and a preset weight corresponding to each of the key quality indicators;

[0177] A first sub-quality detection result of the target service is determined based on the service data corresponding to the target user behavior, the quality threshold corresponding to the key quality indicator, and a preset weight.

[0178] In this embodiment of the present invention, the first detection module 302 is configured to:

[0179] Determining a quality detection score corresponding to each key quality indicator based on the business data corresponding to the target user behavior, a quality threshold corresponding to each key quality indicator, and a preset weight;

[0180] Substitute the quality inspection score corresponding to each key quality indicator into the formula

[0181]

[0182] Get the target quality score corresponding to each key quality indicator, where X is the target quality score corresponding to the key quality indicator, X test is the quality detection score corresponding to the key quality indicator, I is the first preset score, Q is the second preset score, and T is the third preset score;

[0183] Based on the target quality score corresponding to each key quality indicator, a first sub-quality detection result of the target service is determined.

[0184] In an embodiment of the present invention, the apparatus further includes:

[0185] A third acquisition module is used to obtain historical network parameter information during the execution of the target service;

[0186] The second training module is used to train the preset recurrent neural network model based on the historical network parameter information to obtain the data transmission situation prediction model.

[0187] In this embodiment of the present invention, the parameter adjustment module 303 is configured to:

[0188] Determining a first quality detection result of the target service based on the service data flow and preset key performance indicators;

[0189] Based on the first quality detection result and the quality detection result, a parameter adjustment strategy for the target service is determined, where the parameter adjustment strategy includes at least one or more strategies of a cell failure optimization strategy, a service balancing adjustment strategy, a carrier scheduling strategy, and a planning adjustment strategy.

[0190] An embodiment of the present invention provides a resource scheduling device that obtains a service data stream generated by a target user device executing a target service and identifies the user behavior corresponding to the service data stream based on a pre-trained service identification model to obtain a target user behavior corresponding to the service data stream. The service identification model is obtained by training a model constructed by a preset machine learning algorithm based on the historical service data stream of the target service. The device determines the quality detection result of the target service based on preset key quality indicators and service data corresponding to the target user behavior in the service data stream. If the quality detection result does not meet the preset quality requirement conditions, a parameter adjustment strategy for the target service is determined based on the quality detection result, and the parameter adjustment strategy is executed to schedule resources for the target service. In this way, the quality detection result for the target service can be determined based on the identified target user behavior of the target service and the corresponding service data. If the quality result does not meet the preset quality requirement conditions, a parameter adjustment strategy is determined based on the quality detection result, that is, targeted parameter adjustment can be performed for the target user behavior of the target service to achieve good network optimization effects and improve user experience.

[0191] Example 4

[0192] Figure 4 A schematic diagram of the hardware structure of an electronic device for implementing various embodiments of the present invention is provided below.

[0193] The electronic device 400 includes but is not limited to: a radio frequency unit 401, a network module 402, an audio output unit 403, an input unit 404, a sensor 405, a display unit 406, a user input unit 407, an interface unit 408, a memory 409, a processor 410, and a power supply 411. It will be understood by those skilled in the art that Figure 4 The electronic device structure shown in the figure does not constitute a limitation on the electronic device. The electronic device may include more or fewer components than shown, or may combine certain components or arrange the components differently. In the embodiments of the present invention, the electronic device includes but is not limited to a mobile phone, a tablet computer, a laptop computer, a PDA, an in-vehicle terminal, a wearable device, and a pedometer.

[0194] Among them, the processor 410 is used to: obtain the business data stream generated by the target user device executing the target business, and identify the user behavior corresponding to the business data stream based on a pre-trained business identification model to obtain the target user behavior corresponding to the business data stream, wherein the business identification model is obtained by training a model constructed by a preset machine learning algorithm based on the historical business data stream of the target business; determine the quality inspection result of the target business based on preset key quality indicators and the business data corresponding to the target user behavior in the business data stream; if the quality inspection result does not meet the preset quality requirement conditions, determine the parameter adjustment strategy for the target business based on the quality inspection result, and execute the parameter adjustment strategy to perform resource scheduling for the target business.

[0195] Processor 410 is further configured to: determine a first sub-quality detection result of the target service based on preset key quality indicators and service data corresponding to the target user behavior in the service data stream; obtain network parameter information for executing the target service, and determine the data transmission status during the execution of the target service based on a pre-trained data transmission status prediction model and the network parameter information, wherein the data transmission status prediction model is obtained by training a model constructed using a preset machine learning algorithm based on historical network parameter information of the target service; and determine the quality detection result based on the first sub-quality detection result and the data transmission status.

[0196] Processor 410 is further configured to: obtain a historical business data stream for executing the target business; divide the historical business data stream into one or more data classes based on a protocol type corresponding to the historical business data contained in the historical business data stream, each data class containing one or more historical business data, and different data classes corresponding to different protocol types; obtain a sub-business identification model corresponding to the data class, and train the sub-business identification model based on the historical business data contained in the data class to obtain the trained sub-business identification model corresponding to the data class; determine the data class to which the business data belongs based on the protocol type corresponding to the business data contained in the business data stream; input the business data into the sub-business identification model corresponding to the data class to which the business data belongs for identification to obtain a target sub-user behavior; and determine the target user behavior corresponding to the business data stream based on the target sub-user behavior.

[0197] The processor 410 is further configured to obtain a quality threshold and a preset weight corresponding to each of the key quality indicators; and determine a first sub-quality detection result of the target service based on the service data corresponding to the target user behavior, the quality threshold and the preset weight corresponding to the key quality indicator.

[0198] In addition, the processor 410 is further configured to: determine a quality detection score corresponding to each key quality indicator based on the business data corresponding to the target user behavior, the quality threshold corresponding to each key quality indicator, and a preset weight; substitute the quality detection score corresponding to each key quality indicator into the formula

[0199]

[0200] Get the target quality score corresponding to each key quality indicator, where X is the target quality score corresponding to the key quality indicator, X test is the quality detection score corresponding to the key quality indicator, I is the first preset score, Q is the second preset score, and T is the third preset score; based on the target quality score corresponding to each key quality indicator, determine the first sub-quality detection result of the target business.

[0201] In addition, the processor 410 is further used to: obtain historical network parameter information during the execution of the target business; and train a preset recurrent neural network model based on the historical network parameter information to obtain the data transmission situation prediction model.

[0202] In addition, the processor 410 is also used to: determine the first quality detection result of the target service based on the service data flow and preset key performance indicators; determine the parameter adjustment strategy for the target service based on the first quality detection result and the quality detection result, and the parameter adjustment strategy includes at least one or more strategies of cell fault optimization strategy, service balancing adjustment strategy, carrier scheduling strategy and planning adjustment strategy.

[0203] An embodiment of the present invention provides an electronic device that obtains a service data stream generated by a target user device executing a target service, and identifies the user behavior corresponding to the service data stream based on a pre-trained service identification model to obtain a target user behavior corresponding to the service data stream. The service identification model is obtained by training a model constructed by a preset machine learning algorithm based on the historical service data stream of the target service. The quality detection result of the target service is determined based on preset key quality indicators and service data corresponding to the target user behavior in the service data stream. If the quality detection result does not meet the preset quality requirement conditions, a parameter adjustment strategy for the target service is determined based on the quality detection result, and the parameter adjustment strategy is executed to schedule resources for the target service. In this way, the quality detection result for the target service can be determined based on the identified target user behavior of the target service and the corresponding service data. If the quality result does not meet the preset quality requirement conditions, a parameter adjustment strategy is determined based on the quality detection result, that is, targeted parameter adjustment can be performed for the target user behavior of the target service to achieve good network optimization effects and improve user experience.

[0204] It should be understood that in this embodiment of the present invention, the RF unit 401 can be used to receive and transmit signals during information transmission or calls. Specifically, it receives downlink data from the base station and transmits it to the processor 410 for processing; in addition, it transmits uplink data to the base station. Typically, the RF unit 401 includes, but is not limited to, an antenna, at least one amplifier, a transceiver, a coupler, a low-noise amplifier, a duplexer, and the like. Furthermore, the RF unit 401 can communicate with the network and other electronic devices via a wireless communication system.

[0205] The electronic device provides users with wireless broadband Internet access through the network module 402, such as helping users to send and receive emails, browse web pages, and access streaming media.

[0206] The audio output unit 403 can convert audio data received by the RF unit 401 or the network module 402 or stored in the memory 409 into an audio signal and output it as sound. In addition, the audio output unit 403 can also provide audio output related to a specific function performed by the electronic device 400 (for example, a call signal reception sound, a message reception sound, etc.). The audio output unit 403 includes a speaker, a buzzer, a receiver, etc.

[0207] The input unit 404 is used to receive audio or video signals. The input unit 404 may include a graphics processing unit (GPU) 4041 and a microphone 4042. The graphics processor 4041 processes image data of a still picture or video obtained by an image capture device (such as a camera) in a video capture mode or an image capture mode. The processed image frames can be displayed on the display unit 406. The image frames processed by the graphics processor 4041 can be stored in the memory 409 (or other storage medium) or transmitted via the radio frequency unit 401 or the network module 402. The microphone 4042 can receive sound and process such sound into audio data. The processed audio data can be converted into a format that can be sent to a mobile communication base station via the radio frequency unit 401 in the case of a telephone call mode.

[0208] The electronic device 400 also includes at least one sensor 405, such as a light sensor, a motion sensor, and other sensors. Specifically, the light sensor includes an ambient light sensor and a proximity sensor, wherein the ambient light sensor can adjust the brightness of the display panel 4061 according to the brightness of the ambient light, and the proximity sensor can turn off the display panel 4061 and / or the backlight when the electronic device 400 is moved to the ear. As a type of motion sensor, the accelerometer sensor can detect the magnitude of acceleration in all directions (generally three axes), and can detect the magnitude and direction of gravity when stationary. It can be used to identify the posture of the electronic device (such as horizontal and vertical screen switching, related games, magnetometer posture calibration), vibration recognition related functions (such as pedometer, tapping), etc.; the sensor 405 can also include a fingerprint sensor, a pressure sensor, an iris sensor, a molecular sensor, a gyroscope, a barometer, a hygrometer, a thermometer, an infrared sensor, etc., which will not be repeated here.

[0209] The display unit 406 is used to display information input by the user or information provided to the user. The display unit 406 may include a display panel 4061, which may be configured in the form of a liquid crystal display (LCD), an organic light-emitting diode (OLED), or the like.

[0210] The user input unit 407 can be used to receive input digital or character information, and to generate key signal input related to the user settings and function control of the electronic device. Specifically, the user input unit 407 includes a touch panel 4071 and other input devices 4072. The touch panel 4071, also known as a touch screen, can collect user touch operations on or near it (such as operations performed by the user using any suitable object or accessory such as a finger, stylus, etc. on or near the touch panel 4071). The touch panel 4071 may include two parts: a touch detection device and a touch controller. Among them, the touch detection device detects the user's touch direction and detects the signal caused by the touch operation, and transmits the signal to the touch controller; the touch controller receives the touch information from the touch detection device and converts it into touch point coordinates, which are then sent to the processor 410, which receives and executes the command sent by the processor 410. In addition, the touch panel 4071 can be implemented using various types such as resistive, capacitive, infrared, and surface acoustic wave. In addition to the touch panel 4071, the user input unit 407 may also include other input devices 4072. Specifically, other input devices 4072 may include but are not limited to a physical keyboard, function keys (such as volume control keys, switch keys, etc.), a trackball, a mouse, and a joystick, which are not described in detail here.

[0211] Furthermore, the touch panel 4071 may be overlaid on the display panel 4061. When the touch panel 4071 detects a touch operation on or near it, it transmits the information to the processor 410 to determine the type of touch event. Subsequently, the processor 410 provides corresponding visual output on the display panel 4061 according to the type of touch event. Figure 4 In the figure, the touch panel 4071 and the display panel 4061 are two independent components to realize the input and output functions of the electronic device. However, in some embodiments, the touch panel 4071 and the display panel 4061 can be integrated to realize the input and output functions of the electronic device, which is not limited here.

[0212] The interface unit 408 is an interface for connecting external devices to the electronic device 400. For example, the external devices may include a wired or wireless headset port, an external power supply (or battery charger) port, a wired or wireless data port, a memory card port, a port for connecting a device with an identification module, an audio input / output (I / O) port, a video I / O port, a headphone port, etc. The interface unit 408 may be used to receive input (e.g., data information, power, etc.) from the external device and transmit the received input to one or more elements within the electronic device 400, or may be used to transmit data between the electronic device 400 and the external device.

[0213] Memory 409 can be used to store software programs and various data. Memory 409 may primarily include a program storage area and a data storage area. The program storage area may store an operating system and at least one application required for a function (such as a sound playback function or an image playback function); the data storage area may store data generated based on the use of the mobile phone (such as audio data, a phone book, etc.). Furthermore, memory 409 may include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0214] Processor 410 is the control center of the electronic device. It connects all parts of the electronic device using various interfaces and circuits. By running or executing software programs and / or modules stored in memory 409 and accessing data stored in memory 409, it performs various functions of the electronic device and processes data, thereby monitoring the electronic device as a whole. Processor 410 may include one or more processing units; preferably, processor 410 may integrate an application processor and a modem processor, wherein the application processor primarily processes the operating system, user interface, and application programs, while the modem processor primarily handles wireless communications. It is understood that the modem processor may not be integrated into processor 410.

[0215] The electronic device 400 may also include a power supply 411 (such as a battery) to supply power to each component. Preferably, the power supply 411 may be logically connected to the processor 410 through a power management system, thereby enabling the power management system to manage charging, discharging, and power consumption.

[0216] Preferably, an embodiment of the present invention also provides an electronic device, including a processor 410, a memory 409, and a computer program stored in the memory 409 and executable on the processor 410. When the computer program is executed by the processor 410, each process of the above-mentioned resource scheduling method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be described here.

[0217] Example 5

[0218] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon. When executed by a processor, the computer program implements the various processes of the resource scheduling method embodiment described above and achieves the same technical effects. To avoid repetition, the details are not described here. The computer-readable storage medium may be, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0219] An embodiment of the present invention provides a computer-readable storage medium, which obtains a target user behavior corresponding to the service data stream by acquiring a service data stream generated by a target user device executing a target service, and identifies the user behavior corresponding to the service data stream based on a pre-trained service identification model, thereby obtaining the target user behavior corresponding to the service data stream. The service identification model is obtained by training a model constructed by a preset machine learning algorithm based on the historical service data stream of the target service. The quality detection result of the target service is determined based on preset key quality indicators and service data corresponding to the target user behavior in the service data stream. If the quality detection result does not meet the preset quality requirement conditions, a parameter adjustment strategy for the target service is determined based on the quality detection result, and the parameter adjustment strategy is executed to schedule resources for the target service. In this way, the quality detection result for the target service can be determined based on the identified target user behavior of the target service and the corresponding service data. If the quality result does not meet the preset quality requirement conditions, a parameter adjustment strategy is determined based on the quality detection result, that is, targeted parameter adjustment can be performed for the target user behavior of the target service to achieve good network optimization effects and improve user experience.

[0220] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0221] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as combinations of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0222] These computer program instructions may 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 produce an article of manufacture including an instruction device that implements the process. Figure 1a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0223] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0224] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0225] Memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

[0226] Computer-readable media include permanent and non-permanent, removable and non-removable media that can be used to store information by any method or technology. Information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include temporary computer-readable media (transitory media), such as modulated data signals and carrier waves.

[0227] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not preclude the presence of other identical elements in the process, method, commodity, or apparatus comprising the element.

[0228] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0229] The foregoing is merely an embodiment of the present invention and is not intended to limit the present invention. It will be apparent to those skilled in the art that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention are intended to be included within the scope of the claims of the present invention.

Claims

1. A resource scheduling method, characterized in that: The method comprises: Obtaining a service data stream generated by a target user device executing a target service, and identifying user behavior corresponding to the service data stream based on a pre-trained service identification model to obtain a target user behavior corresponding to the service data stream, wherein the service identification model is obtained by training a model constructed using a preset machine learning algorithm based on historical service data streams of the target service; Determining a quality detection result of the target service based on preset key quality indicators and service data corresponding to the target user behavior in the service data stream; If the quality detection result does not meet the preset quality requirement condition, determining a parameter adjustment strategy for the target service based on the quality detection result, and executing the parameter adjustment strategy to perform resource scheduling for the target service; The pre-trained service identification model is used to identify the user behavior corresponding to the service data flow to obtain the target user behavior corresponding to the service data flow, including: Obtaining the service data flow generated by the target user equipment executing the target service; Determine the data class to which the business data belongs based on the protocol type corresponding to the business data contained in the business data stream; The business data is input into the sub-business identification model corresponding to the data class to which the business data belongs for identification to obtain the target sub-user behavior; Based on the target sub-user behavior, determine the target user behavior corresponding to the business data flow; The determining of the quality detection result of the target service based on the preset key quality indicator and the service data corresponding to the target user behavior in the service data stream includes: Determining a first sub-quality detection result of the target service based on preset key quality indicators and service data corresponding to the target user behavior in the service data stream; Obtaining network parameter information for executing the target service, and determining the data transmission status during the execution of the target service based on a pre-trained data transmission status prediction model and the network parameter information; The quality detection result is determined based on the first sub-quality detection result and the data transmission status.

2. The method according to claim 1, characterized in that The data transmission situation prediction model is obtained by training a model constructed by a preset machine learning algorithm based on historical network parameter information of the target service.

3. The method according to claim 2, characterized in that Before identifying the user behavior corresponding to the service data flow based on the pre-trained service identification model to obtain the target user behavior corresponding to the service data flow, the method further includes: Acquire historical business data streams for executing the target business; Based on the protocol type corresponding to the historical service data contained in the historical service data stream, the historical service data stream is divided into one or more data classes, each of the data classes contains one or more historical service data, and different data classes correspond to different protocol types; A sub-business identification model corresponding to the data class is obtained, and the sub-business identification model is trained based on the historical business data contained in the data class to obtain the trained sub-business identification model corresponding to the data class.

4. The method according to claim 1, wherein The determining, based on a preset key quality indicator and the service data corresponding to the target user behavior in the service data stream, a first sub-quality detection result of the target service includes: Obtaining a quality threshold and a preset weight corresponding to each of the key quality indicators; A first sub-quality detection result of the target service is determined based on the service data corresponding to the target user behavior, the quality threshold corresponding to the key quality indicator, and a preset weight.

5. The method according to claim 4, characterized in that The determining, based on the service data corresponding to the target user behavior, the quality threshold corresponding to the key quality indicator, and the preset weight, of the first sub-quality detection result of the target service includes: Determining a quality detection score corresponding to each key quality indicator based on the business data corresponding to the target user behavior, a quality threshold corresponding to each key quality indicator, and a preset weight; Substitute the quality inspection score corresponding to each key quality indicator into the formula ， Obtain the target quality score corresponding to each key quality indicator, where X is the target quality score corresponding to the key quality indicator, is the quality detection score corresponding to the key quality indicator, I is the first preset score, Q is the second preset score, and T is the third preset score; Based on the target quality score corresponding to each key quality indicator, a first sub-quality detection result of the target service is determined.

6. The method according to claim 5, characterized in that Before determining the data transmission status during the execution of the target service based on the pre-trained data transmission status prediction model and the network parameter information, the method further includes: Acquire historical network parameter information during the execution of the target service; The preset recurrent neural network model is trained based on the historical network parameter information to obtain the data transmission situation prediction model.

7. The method according to claim 6, characterized in that The determining, based on the quality detection result, a parameter adjustment strategy for the target service includes: Determining a first quality detection result of the target service based on the service data flow and preset key performance indicators; Based on the first quality detection result and the quality detection result, a parameter adjustment strategy for the target service is determined, where the parameter adjustment strategy includes at least one or more strategies of a cell failure optimization strategy, a service balancing adjustment strategy, a carrier scheduling strategy, and a planning adjustment strategy.

8. A resource scheduling device, characterized in that: The device comprises: A first acquisition module is configured to acquire a service data stream generated by a target user device executing a target service, and identify user behavior corresponding to the service data stream based on a pre-trained service identification model to obtain a target user behavior corresponding to the service data stream, wherein the service identification model is obtained by training a model constructed using a preset machine learning algorithm based on historical service data streams of the target service; A first detection module is configured to determine a quality detection result of the target service based on preset key quality indicators and service data corresponding to the target user behavior in the service data stream; a parameter adjustment module, configured to determine, based on the quality detection result, a parameter adjustment strategy for the target service and execute the parameter adjustment strategy to perform resource scheduling for the target service if the quality detection result does not meet the preset quality requirement condition; The first acquisition module is further configured to acquire a service data flow generated by a target user equipment executing a target service; Determine the data class to which the business data belongs based on the protocol type corresponding to the business data contained in the business data stream; The business data is input into the sub-business identification model corresponding to the data class to which the business data belongs for identification to obtain the target sub-user behavior; Based on the target sub-user behavior, determine the target user behavior corresponding to the business data flow; The first detection module is further configured to determine a first sub-quality detection result of the target service based on preset key quality indicators and service data corresponding to the target user behavior in the service data stream; Obtaining network parameter information for executing the target service, and determining the data transmission status during the execution of the target service based on a pre-trained data transmission status prediction model and the network parameter information; The quality detection result is determined based on the first sub-quality detection result and the data transmission status.

9. An electronic device, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program implements the steps of the resource scheduling method according to any one of claims 1 to 7 when executed by the processor.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the resource scheduling method according to any one of claims 1 to 7 are implemented.

Images