Method and apparatus for determining configuration parameters of a serving cell, electronic device, and medium

Abstract

The application provides a method and device for determining configuration parameters of a serving cell, electronic equipment and a medium. The method comprises: obtaining a plurality of characteristic variable values of at least one parameter group of the serving cell to be configured, each parameter group corresponding to a trained neural network; inputting the plurality of characteristic variable values of at least one parameter group of the serving cell to be configured into the corresponding trained neural network to determine a target neuron node mapped by the serving cell to be configured; the target neuron node is a neuron node mapped by a normally operating serving cell most similar to the plurality of characteristic variable values of at least one parameter group of the serving cell to be configured; and storing parameter configuration values of at least one normally operating serving cell stored in the target neuron node as parameter configuration values corresponding to at least one parameter group of the serving cell to be configured. The method of the application can provide better communication services for users.

Description

TECHNICAL FIELD

[0001] The present application relates to base station technology, and particularly relates to a method and device for determining configuration parameters of a serving cell, electronic equipment and medium. BACKGROUND

[0002] A serving cell refers to an area covered by a base station or a part of a base station in cellular mobile communication, and a mobile station in the area can reliably communicate with the base station through a wireless channel. When a new base station appears or the performance of the base station is improved, the parameters of the base station need to be configured or reconfigured, so that each serving cell can operate according to the configured parameters to provide better communication services for users.

[0003] At present, the parameters of the serving cell are mainly configured according to default values or templates, and the default values and templates are determined according to experience.

[0004] However, the parameters configured according to experience without considering the actual situation of the serving cell cannot provide better communication services for users according to the characteristics of the serving cell. SUMMARY

[0005] The present application provides a method and device for determining configuration parameters of a serving cell, electronic equipment and medium, to solve the technical problem that the serving cell cannot provide better communication services for users according to its own characteristics in the prior art.

[0006] In a first aspect, the present application provides a method for determining configuration parameters of a serving cell, comprising:

[0007] obtaining a plurality of characteristic variable values of at least one parameter group of a serving cell to be configured, each parameter group corresponding to a trained neural network;

[0008] inputting the plurality of characteristic variable values of the at least one parameter group of the serving cell to be configured into the corresponding trained neural network, to determine a target neuron node mapped by the serving cell to be configured; the target neuron node is a neuron node mapped by a normally operating serving cell most similar to the plurality of characteristic variable values of the at least one parameter group of the serving cell to be configured; and a plurality of neuron nodes of an output layer of the trained neural network store parameter configuration values of at least one normally operating serving cell having a mapping relationship with the plurality of neuron nodes;

[0009] determining the parameter configuration values of the at least one normally operating serving cell stored in the target neuron node as the parameter configuration values corresponding to the at least one parameter group of the serving cell to be configured.

[0010] In a second aspect, the present application provides a configuration parameter determination apparatus for a serving cell, comprising:

[0011] a variable value obtaining module, configured to obtain a plurality of characteristic variable values of at least one parameter group of the serving cell to be configured, each parameter group corresponding to a trained neural network;

[0012] a neuron determination module, configured to input the plurality of characteristic variable values of at least one parameter group of the serving cell to be configured into the corresponding trained neural network, to determine a target neuron node mapped by the serving cell to be configured; the target neuron node is a neuron node mapped by a normally operating serving cell most similar to the plurality of characteristic variable values of at least one parameter group of the serving cell to be configured; and a plurality of neuron nodes of an output layer of the trained neural network store parameter configuration values of at least one normally operating serving cell having a mapping relationship with the plurality of neuron nodes;

[0013] a parameter configuration determination module, configured to determine the parameter configuration values of at least one normally operating serving cell stored by the target neuron node as parameter configuration values corresponding to at least one parameter group of the serving cell to be configured.

[0014] In a third aspect, the present application provides an electronic device, comprising: a processor, and a memory connected with the processor in communication;

[0015] the memory stores computer execution instructions;

[0016] the processor executes the computer execution instructions stored by the memory, to implement the method according to the first aspect.

[0017] In a fourth aspect, the present application provides a computer readable storage medium, wherein the computer readable storage medium stores computer execution instructions, and the computer execution instructions are executed by a processor to implement the method according to the first aspect.

[0018] The application provides a service cell configuration parameter determination method, device, electronic equipment and medium. The method comprises the following steps: obtaining a plurality of characteristic variable values of at least one parameter group of a service cell to be configured, each parameter group corresponding to a trained neural network; inputting the plurality of characteristic variable values of at least one parameter group of the service cell to be configured into the corresponding trained neural network to determine a target neuron node mapped by the service cell to be configured; the target neuron node is a neuron node mapped by a service cell in normal operation which is most similar to the plurality of characteristic variable values of at least one parameter group of the service cell to be configured; and a plurality of neuron nodes of an output layer of the trained neural network store parameter configuration values of at least one service cell in normal operation which have a mapping relationship with the plurality of neuron nodes; and determining the parameter configuration values of at least one service cell in normal operation stored by the target neuron node as the parameter configuration values corresponding to at least one parameter group of the service cell to be configured. Since the target neuron node mapped by the service cell to be configured is a neuron node mapped by a service cell in normal operation which is most similar to the plurality of characteristic variable values of at least one parameter group of the service cell to be configured, and the target neuron node stores the parameter configuration values of the most similar service cell in normal operation, the parameter configuration values of at least one service cell in normal operation stored by the target neuron node are determined as the parameter configuration values corresponding to at least one parameter group of the service cell to be configured, which is equivalent to configuring parameters according to the characteristics of the service cell to be configured, and combining the characteristics, thereby providing better communication services for users. BRIEF DESCRIPTION OF DRAWINGS

[0019] The accompanying drawings, which are incorporated herein and form part of the specification, illustrate embodiments consistent with the present application and, together with the description, further serve to explain the principles of the application.

[0020] Figure 1 An application scenario diagram of the service cell configuration parameter determination method according to an embodiment of the present application;

[0021] Figure 2 A flowchart of the service cell configuration parameter determination method according to an embodiment of the present application;

[0022] Figure 3 A flowchart of the service cell configuration parameter determination method according to another embodiment of the present application;

[0023] Figure 4 A structural diagram of the service cell configuration parameter determination method according to the present application;

[0024] Figure 5 A structural diagram of the electronic equipment used to implement the service cell configuration parameter determination method.

[0025] The specific embodiments of the application have been shown by way of example in the above figures, and will be described in greater detail below. These figures and this written description are not intended to limit the scope of the inventive concept in any way, but rather to illustrate the inventive concept by reference to specific embodiments. DETAILED DESCRIPTION

[0026] The exemplary embodiments will be described in detail herein with reference to the attached drawings. The same reference numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments are not meant to represent all embodiments consistent with the application. Rather, they are merely examples of apparatus and methods consistent with some aspects of the application as detailed in the appended claims.

[0027] In order to clearly understand the technical solutions of the present application, the prior art solutions are first described in detail.

[0028] In the conventional manner, the parameters of the serving cell are configured mainly according to the default values or by template, and the default values and the template are formulated according to experience. However, the parameters fixed according to experience without considering the actual situation of the serving cell will make the serving cell unable to provide better communication services for users in combination with its own characteristics when the serving cell is configured.

[0029] Therefore, in the face of the technical problems of the prior art, the inventors found, through creative research, that in order to provide better communication services for users in combination with the characteristics of the serving cell, the mapping relationship between the serving cell in normal operation and the neuron node is established in advance, that is, the plurality of neuron nodes of the output layer of the trained neural network store the parameter configuration values of at least one serving cell in normal operation which has a mapping relationship with the plurality of neuron nodes. When a serving cell needs to be configured, as the serving cell to be configured, the plurality of characteristic variable values of at least one parameter group are input into the neural network model corresponding to the at least one parameter group to determine the target neuron node mapped with the serving cell to be configured, so that the parameter configuration values of the serving cell in normal operation stored in the target neuron node are used for parameter configuration of the serving cell to be configured. Since the target neuron node mapped with the serving cell to be configured is the neuron node mapped with the most similar serving cell in normal operation to the plurality of characteristic variable values of at least one parameter group of the serving cell to be configured, and the target neuron node stores the parameter configuration values of the most similar serving cell in normal operation. Therefore, the parameter configuration values of at least one serving cell in normal operation stored in the target neuron node are determined as the parameter configuration values corresponding to at least one parameter group of the serving cell to be configured, which is equivalent to configuring parameters according to the characteristics of the serving cell to be configured, and combining the characteristics, so as to provide better communication services for users.

[0030] As shown in Figure 1 The application embodiment provides an application scenario of the configuration parameter determination method of the serving cell, and the application scenario includes an electronic device 10 and a data providing platform 20 in a corresponding network architecture. The electronic device 10 can be an electronic device of a base station, and the data providing platform 20 can have a plurality of data providing platforms 20. Each data providing platform 20 can provide a plurality of characteristic variables of different parameter groups, and each data providing platform 20 is in communication connection with the electronic device 10. When a new base station is built or the base station needs to be specially improved, the data providing platform 20 transmits cell data of at least one parameter group of the serving cell to be configured to the electronic device 10, the electronic device 10 acquires a plurality of characteristic variables of at least one parameter group of the serving cell to be configured, processes the plurality of characteristic variables of the at least one parameter group, and obtains a plurality of characteristic variable values of the at least one parameter group of the serving cell to be configured. The electronic device 10 inputs the plurality of characteristic variable values of the at least one parameter group of the serving cell to be configured into the corresponding trained neural network, determines the mapped target neuron node, and determines the parameter configuration values of at least one serving cell in normal operation stored in the target neuron node as the parameter configuration values corresponding to the at least one parameter group of the serving cell to be configured.

[0031] The technical solutions of the present application and how the technical solutions of the present application solve the above technical problems will be described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes can not be described again in some embodiments. The embodiments of the present application will be described below with reference to the drawings.

[0032] Figure 2 The method for determining configuration parameters of a serving cell provided by an embodiment of the present application, as shown in the method for determining configuration parameters of a serving cell provided by an embodiment of the present application, the execution subject of the method is an electronic device. The method for determining configuration parameters of a serving cell provided by the embodiment includes the following steps: Figure 2

[0033] Step 101, obtaining multiple characteristic variable values of at least one parameter group of a serving cell to be configured.

[0034] The serving cell refers to a cell provided by a base station to provide mobile communication services. The serving cell to be configured refers to a serving cell that needs to be configured. The serving cell to be configured can be a serving cell of a newly built base station, or a serving cell of an existing base station that needs to be reconfigured when the existing base station is upgraded in terms of performance.

[0035] The serving cell includes multiple parameters, and the multiple parameters are grouped to obtain multiple parameter groups, such as at least a coverage-related parameter group, a handover-related parameter group, a reselection group parameter group, a quality group parameter group, an interoperation group parameter group, and a capacity-related parameter group. Taking the coverage-related parameter group as an example, the coverage-related parameter group refers to parameters that can affect the coverage of the serving cell and have a low adjustment frequency. Optionally, the coverage-related parameter group includes station height, antenna tilt angle, station spacing, antenna gain, cell frequency band, coverage scenario, and population density. The station height refers to the difference between the horizontal distance of the cell and the coverage area. The antenna tilt angle refers to the angle between the antenna and the vertical direction. The station spacing refers to the average distance between the base station and the nearest N macro stations, and N is a natural positive integer. The cell frequency band refers to the fact that different frequency bands have large differences in propagation loss, and different frequency bands are used in different scenarios. The antenna gain refers to the ratio of the power density of the actual antenna to the ideal radiation unit at the same point in space under the condition that the input power is equal. The coverage scenario refers to the fact that the cell coverage scenario is rural or urban. The population density refers to the service load of the cell coverage area.

[0036] The characteristic variable value is obtained by standardizing the original value of the characteristic variable of the parameter group. Taking the coverage-related parameter group as an example, the characteristic variables include station height, antenna tilt angle, station spacing, antenna gain, and cell frequency band, and the characteristic variable values correspond to the values of station height, antenna tilt angle, station spacing, antenna gain, and cell frequency band. ​

[0037] The trained neural network can be based on a self-organizing mapping neural network, which belongs to an unsupervised learning method. Each parameter group of the to-be-configured serving cell corresponds to a trained neural network. For example, the coverage-related parameter group, the handover-related parameter group, the reselection group parameter group, the quality group parameter group, the interoperation group parameter group, and the capacity-related parameter group correspond to a trained neural network, respectively.

[0038] In step 102, the multiple feature variable values of at least one parameter group of the to-be-configured serving cell are input into the corresponding trained neural network to determine the target neuron node to which the to-be-configured serving cell is mapped.

[0039] The target neuron node refers to the neuron node in the trained neural network to which the parameter group of the to-be-configured serving cell is mapped. If the parameter group of the to-be-configured serving cell has multiple groups, the multiple groups are respectively input into the corresponding trained neural network to determine multiple target neuron nodes to which the to-be-configured serving cell is mapped.

[0040] The cell in normal operation refers to a serving cell that has been configured with parameter values and is in a normal operation state in the process of being put into use.

[0041] The target neuron node is the neuron node to which the serving cell in normal operation that is most similar to the multiple feature variable values of at least one parameter group of the to-be-configured serving cell is mapped. That is, for each parameter group of the to-be-configured serving cell, the target neuron node corresponding to each parameter group is the neuron node to which the serving cell in normal operation that is most similar to the multiple feature variable values of the parameter group of the to-be-configured serving cell is mapped. The target neuron node and the neuron node to which the serving cell in normal operation is mapped are neurons of the output layer of the trained neural network.

[0042] a neuron node mapped by the at least one parameter group of the to-be-configured serving cell, i.e., a target neuron node mapped by the at least one parameter group of the to-be-configured serving cell. For example, if there is only one parameter group, and it is a coverage-related parameter group, then the neuron node mapped by the coverage-related parameter group of the to-be-configured serving cell, i.e., the target neuron node mapped by the coverage-related parameter group of the to-be-configured serving cell, is the neuron node mapped by the coverage-related parameter group of the in-service serving cell that is most similar to the plurality of feature variable values of the coverage-related parameter group of the to-be-configured serving cell. If there are two parameter groups, a coverage-related parameter group and a handover-related parameter group, then the neuron node mapped by the coverage-related parameter group of the to-be-configured serving cell, i.e., the target neuron node mapped by the coverage-related parameter group of the to-be-configured serving cell, is the neuron node mapped by the in-service serving cell that is most similar to the plurality of feature variable values of the coverage-related parameter group of the to-be-configured serving cell; the neuron node mapped by the handover-related parameter group of the to-be-configured serving cell, i.e., the target neuron node mapped by the handover-related parameter group of the to-be-configured serving cell, is the neuron node mapped by the in-service serving cell that is most similar to the plurality of feature variable values of the handover-related parameter group of the to-be-configured serving cell. If there are more than two parameter groups, the process is similar to the case of two parameter groups, and will not be described in detail.

[0043] In addition, the plurality of neuron nodes of the output layer of the trained neural network store the parameter configuration values of at least one in-service serving cell that have a mapping relationship with the plurality of neuron nodes. Here, it can be understood that in the trained neural network, there is a neuron node that does not store the parameter configuration value of any one in-service serving cell that has a mapping relationship with the neuron node. In addition, any neuron node stores the parameter configuration value of at least one in-service serving cell that has a mapping relationship with the neuron node, which can be the parameter configuration value of one in-service serving cell or the parameter configuration values of multiple in-service serving cells.

[0044] In step 103, the parameter configuration value of at least one in-service serving cell stored in the target neuron node is determined as the parameter configuration value corresponding to the at least one parameter group of the to-be-configured serving cell.

[0045] In addition, the plurality of neuron nodes of the output layer of the trained neural network store the parameter configuration values of at least one in-service serving cell that have a mapping relationship with the plurality of neuron nodes. Here, it can be understood that in the trained neural network, there is a neuron node that does not store the parameter configuration value of any one in-service serving cell that has a mapping relationship with the neuron node. In addition, any neuron node stores the parameter configuration value of at least one in-service serving cell that has a mapping relationship with the neuron node, which can be the parameter configuration value of one in-service serving cell or the parameter configuration values of multiple in-service serving cells.

[0046] Exemplarily, a plurality of characteristic variable values of a coverage-related parameter group of the to-be-configured serving cell are acquired, and the plurality of characteristic variable values of the coverage-related parameter group of the to-be-configured serving cell are input into the corresponding trained neural network to determine a target neuron node to which the to-be-configured serving cell is mapped. The target neuron node is a neuron node to which a normally-operating serving cell most similar to the plurality of characteristic variable values of the coverage-related parameter group of the to-be-configured serving cell is mapped. At least one parameter configuration value of the normally-operating serving cell stored in the target neuron node is determined as a corresponding parameter configuration value of the coverage-related parameter group of the to-be-configured serving cell.

[0047] In the present application, a plurality of characteristic variable values of at least one parameter group of a to-be-configured serving cell are acquired, each parameter group corresponding to a trained neural network; the plurality of characteristic variable values of the at least one parameter group of the to-be-configured serving cell are input into the corresponding trained neural network to determine a target neuron node to which the to-be-configured serving cell is mapped; the target neuron node is a neuron node to which a normally-operating serving cell most similar to the plurality of characteristic variable values of the at least one parameter group of the to-be-configured serving cell is mapped; and a plurality of neuron nodes of an output layer of the trained neural network store parameter configuration values of at least one normally-operating serving cell having a mapping relationship with the plurality of neuron nodes; at least one parameter configuration value of the normally-operating serving cell stored in the target neuron node is determined as a corresponding parameter configuration value of the at least one parameter group of the to-be-configured serving cell. Since the target neuron node to which the to-be-configured serving cell is mapped is a neuron node to which a normally-operating serving cell most similar to the plurality of characteristic variable values of the at least one parameter group of the to-be-configured serving cell is mapped. The target neuron node stores the parameter configuration value of the most similar normally-operating serving cell. Therefore, determining the at least one parameter configuration value of the normally-operating serving cell stored in the target neuron node as the corresponding parameter configuration value of the at least one parameter group of the to-be-configured serving cell is equivalent to configuring parameters according to the characteristics of the to-be-configured serving cell itself, which combines the characteristics of the to-be-configured serving cell itself, thereby providing better communication services for users.

[0048] As an optional implementation manner, as shown in Figure 2 Before step 101, the present embodiment further includes the following steps:

[0049] Step 201: acquiring a plurality of cell data of the to-be-configured serving cell.

[0050] The multiple cell data of the to-be-configured serving cell at least include configuration data corresponding to the newly-built base station and survey data of the newly-built base station. The configuration data of the newly-built base station can be understood as planning data of the newly-built base station. The cell data of the to-be-configured serving cell can further include planning data of the newly-built base station, design data of the newly-built base station and audit data of the newly-built base station. The configuration data of the newly-built base station can be obtained from a planning platform, the design data of the newly-built base station can be obtained from a design platform, the survey data of the newly-built base station can be obtained from a survey platform, and the audit data of the newly-built base station can be obtained from an audit platform. The planning platform, the design platform, the survey platform and the audit platform are the multiple data providing platforms mentioned above.

[0051] The cell data are original numerical values of characteristic variables of parameter groups, and specifically include original numerical values of characteristic variables of a coverage-related parameter group, a handover-related parameter group, a reselection group parameter group, a quality group parameter group, an interoperation group parameter group and a capacity-related parameter group. Taking the coverage-related parameter group included in the cell data as an example, the cell data include specific original numerical values of station height, antenna tilt angle, inter-station distance, cell frequency band and the like.

[0052] In step 202, according to the multiple cell data of the to-be-configured serving cell and a pre-stored parameter table, a parameter group corresponding to each cell data in the to-be-configured serving cell is determined.

[0053] The parameter table stores a mapping relationship between cell data and parameter groups. According to the pre-stored parameter table, it can be determined which parameter groups the multiple cell data of the to-be-configured serving cell belong to, i.e., the parameter group corresponding to each cell data of the to-be-configured serving cell is determined, so that the to-be-configured serving cell is grouped by parameter groups, and at least one parameter group of the to-be-configured serving cell can be obtained.

[0054] In this embodiment, the multiple cell data of the to-be-configured serving cell are obtained, and the multiple cell data of the to-be-configured serving cell at least include configuration data corresponding to the newly-built base station and survey data of the newly-built base station. According to the multiple cell data of the to-be-configured serving cell and a pre-stored parameter table, a parameter group corresponding to each cell data in the to-be-configured serving cell is determined, and the parameter table stores a mapping relationship between cell data and parameter groups. After obtaining the multiple cell data of the to-be-configured serving cell, the multiple cell data of the to-be-configured serving cell are grouped by parameter groups to determine the parameter group corresponding to each cell data of the to-be-configured serving cell, so that the multiple characteristic variable values of each parameter group of the to-be-configured serving cell can be determined and input into the corresponding trained neural network in the subsequent process, and accurate parameter configuration values can be obtained.

[0055] As an optional implementation, in step 101, the following steps are included:

[0056] Step 301, obtaining a characteristic variable of at least one parameter group of the to-be-configured serving cell.

[0057] Wherein, the characteristic variable is a variable related to the parameter group, and for example of the related parameter group, the characteristic variable includes station height, antenna tilt angle, station spacing, antenna gain, and cell frequency band, etc.

[0058] Step 302, performing standardization processing on the characteristic variable of at least one parameter group of the to-be-configured serving cell, to obtain a plurality of characteristic variable values of at least one parameter group of the to-be-configured serving cell.

[0059] Wherein, the characteristic variable of at least one parameter group of the to-be-configured serving cell is a numerical characteristic variable. The characteristic variable value is a value obtained by standardizing the value of the numerical characteristic variable.

[0060] Specifically, the standardization processing is to map the value of the numerical characteristic variable to a standard normal distribution with a mean of 0 and a standard deviation of 1, and the purpose is to eliminate the characteristic variable with large value and large variance.

[0061] Standardization process: obtaining the mean and standard deviation of the plurality of characteristic variables of at least one parameter group of the to-be-configured serving cell; subtracting the mean from the plurality of characteristic variables of at least one parameter group of the to-be-configured serving cell respectively and dividing by the standard deviation, to obtain the plurality of characteristic variable values of at least one parameter group of the to-be-configured serving cell.

[0062] For example:

[0063] ①Calculate the mean μ of the numerical characteristic variable;

[0064] ②Calculate the standard deviation δ of the numerical characteristic variable;

[0065] ③Calculate the value X' of the standardization processing, X'=(X-μ) / δ, wherein X is the characteristic variable, and X' is the characteristic variable value.

[0066] Wherein, for the characteristic variable with large value and large variance, it is determined as an outlier sample and is removed, and the following formula can be used to calculate to determine whether the characteristic variable has large value or large variance:

[0067] Y i =1 / n+(X i -μ) / ∑(X-μ) 2

[0068] Y i is a leverage statistic, n is the total number of samples, x i is the value of the characteristic variable of the i-th sample, μ is the mean of the characteristic variable, and ∑(X-μ)2 The sum of squares of the difference between the characteristic variable and the mean value is calculated, and if the sample Y i If the sample Y

[0069] In this embodiment, the characteristic variable of at least one parameter group of the to-be-configured serving cell is obtained; and the characteristic variable of at least one parameter group of the to-be-configured serving cell is normalized to obtain a plurality of characteristic variable values of at least one parameter group of the to-be-configured serving cell. Since the plurality of characteristic variable values of at least one parameter group of the to-be-configured serving cell are obtained through normalization, it can be ensured that there is no obviously outlying sample between the plurality of characteristic variable values of each parameter group, which is beneficial to improve the accuracy of subsequent mapping of target neuron nodes.

[0070] As an optional implementation, before step 101, the following step is further included in this embodiment:

[0071] In step 401, the initial neural network corresponding to each parameter group is initialized, and the corresponding initial learning rate is established.

[0072] As described above, each parameter group corresponds to a trained neural network, and therefore before step 101, when training the neural network, a plurality of initial neural networks can be trained to obtain a plurality of trained neural networks. Before training, the neural network corresponding to each parameter group is initialized, and the initialization includes initializing the weight between each neuron, establishing the initial winning neighborhood, and establishing the initial learning rate.

[0073] For understanding of the self-organizing learning process of the self-organizing mapping neural network: in the training phase, a certain number of training set samples are randomly input to the model input layer, and for these specific samples, a certain neuron "wins" to produce the maximum response in the output layer. When the input sample changes, the winning neuron in the two-dimensional plane also changes, and the neurons around the winning neuron also produce a larger response due to the lateral mutual excitation effect, that is, the weight vectors connected to the winning neuron and all neurons in the winning neighborhood are adjusted in the direction of the input vector to different degrees, and the adjustment degree gradually decays depending on the distance of each neuron in the neighborhood from the winning neuron. The neural network adjusts the weight of the output layer neuron based on a large number of training samples through self-organizing, and finally makes the output layer neurons become input-sensitive nerve cells of specific patterns, so as to form an ordered feature map in the output layer that can reflect the sample pattern class distribution. In this embodiment, the ordered feature map is the mapping relationship between the neuron nodes of the output layer of the trained neural network and the serving cell in normal operation.

[0074] In step 402, the training sample set corresponding to each initial neural network is obtained.

[0075] wherein the sample in the training sample set is a plurality of feature variable values of the corresponding parameter group of the service cell in normal operation.

[0076] Step 403, iteratively learning each initial neural network by the sample of each training sample set respectively, and updating the corresponding learning rate.

[0077] wherein iteratively learning each initial neural network by the sample of each training sample set respectively, it is understood that the initial neural network corresponding to each parameter group, with the plurality of feature variable values of the parameter group as the training sample, iteratively learning, in the process of iteratively learning, the learning rate and the winning neighborhood are updated each time. The winning neighborhood is a neighborhood composed of a plurality of adjacent neuron nodes centered on the winning neuron node.

[0078] Specifically, the iteratively learning process:

[0079] 1) First, calculate the discriminant function value of the feature variable value of each input layer neuron node, and determine the specific neuron node with the minimum discriminant function value as the winner. The discriminant function value calculation method of the neuron node is:

[0080]

[0081] wherein w i，j is the connection weight between the input layer neuron node i and the output layer neuron node j, x i is the feature variable value of the input layer neuron node i, d j (x) is the discriminant function value of the neuron node j.

[0082] 2) Determine the weight adjustment neighborhood at time t (which can be understood as determining the weight adjustment neighborhood at the tth iteration with the winning neuron node as the center). Generally, the initial neighborhood N is large, and N gradually shrinks with the training time (iteration times) during the training process. The spatial position of the winning neighborhood:

[0083]

[0084] wherein t represents the training time, i.e. the iteration times, I(x) represents the winning neuron node, s represents the distance between the winning neuron node and the surrounding neuron nodes, σ0, τ σ are constants. T j,1 is the radius, with the winning neuron node at this time as the center, T j,1 is the radius, i.e. the weight adjustment neighborhood at the tth iteration is determined.

[0085] 3) adjusting the connection weights of the neuron nodes in the winning neighborhood spatial range, so that the winning neuron node and the neuron nodes in its neighborhood range are more likely to win again for the input of the similar pattern.

[0086] Wherein, when adjusting the connection weights, the distance between each neuron node and the winning neuron node can be referred to, and the weights between the neuron nodes with smaller distance to the winning neuron node and the winning neuron node are adjusted to be larger. The input response enhancement can be understood as that, for the input of the similar pattern, the winning neuron node and the neuron nodes in its neighborhood range are more likely to win again.

[0087] The specific adjustment value is:

[0088] Δw i，j = θ (t) * T j，I(x) (t) * (x i -w i，j )

[0089] Wherein, the learning rate dependent on the training times is defined as

[0090] Step 404, if the learning rate of each initial neural network is less than the corresponding target learning rate, it is determined that the plurality of trained neural networks are obtained.

[0091] Wherein, the foregoing iterative competitive learning process is repeated until the learning rate of each initial neural network is less than the corresponding target learning rate, and it is determined that the trained neural network corresponding to each parameter group is obtained, that is, the plurality of trained neural networks are obtained.

[0092] The result of the training is to map the feature variable value of the new input serving cell to the neuron node closest to the feature variable value in the specific application.

[0093] In the embodiment, the initial neural network corresponding to each parameter group is initialized, and the corresponding initial learning rate is established; the training sample set corresponding to each initial neural network is obtained, and the samples in the training sample set are a plurality of feature variable values of the corresponding parameter group of the serving cell in normal operation; each initial neural network is subjected to iterative competitive learning by the samples of each training sample set, and the corresponding learning rate is updated; if the learning rate of each initial neural network is less than the corresponding target learning rate, it is determined that the plurality of trained neural networks are obtained. Since the plurality of trained neural networks are obtained by continuously iterative competitive learning, the accuracy of the plurality of trained neural networks can be ensured.

[0094] As an optional implementation, in step 402, the embodiment includes the following steps:

[0095] Step 501, according to the KPI running index, from the multiple running service cells, multiple normal running service cells are screened.

[0096] Among them, the running service cell refers to the service cell that has been put into operation. The normal running service cell is a service cell with relatively good running condition among the running service cells. The KPI running index is a related index representing the running condition of the service cell. The KPI running index includes the accessibility index, the maintainability index and the integrity index. When the three indexes meet the requirements, the running service cell is determined as the normal running service cell.

[0097] The accessibility index includes but is not limited to: radio access rate, RAB congestion rate, paging congestion rate, RRC connection success rate, RRC connection reestablishment rate, RRC connection reestablishment failure rate and other accessibility index parameters. The maintainability index includes but is not limited to: uplink interference busy average, downlink sensing rate, uplink sensing rate, MR coverage rate and other maintainability index parameters. The integrity index includes but is not limited to: drop rate, drop rate, handover success rate, same frequency handover success rate, different frequency handover success rate and other integrity index parameters. When the running service cell meeting the requirements of the accessibility index, the maintainability index and the integrity index is screened, at least 4 index parameters of each type of index such as the accessibility index, the maintainability index and the integrity index are selected, for example, at least 4 index parameters are selected from the radio access rate, the RAB congestion rate, the paging congestion rate, the RRC connection success rate, the RRC connection reestablishment rate and the RRC connection reestablishment failure rate to participate in the screening.

[0098] Step 502, a plurality of parameter groups are obtained.

[0099] Among them, the plurality of parameter groups are obtained according to the functional role of the cell data of each normal running service cell. For example, the cell data related to the coverage capability of the service cell is divided into a coverage related parameter group, and the cell data related to the switching stability / safety of the service cell is divided into a switching related parameter group.

[0100] Step 503, according to each parameter group, the training sample set corresponding to each initial neural network is obtained.

[0101] That is, the training sample set of each initial neural network includes multiple characteristic variable values of the corresponding parameter group.

[0102] In the embodiment, according to the KPI operation index, a plurality of normally operating service cells are screened from a plurality of operating service cells, the KPI operation index is a related index representing the operation condition of the service cell; a plurality of parameter groups are obtained, each parameter group is obtained according to the function of the cell data of each normally operating service cell; and a training sample set corresponding to each initial neural network is obtained according to each parameter group. Since the training sample set for determining each initial neural network is determined based on the normally operating service cells screened according to the KPI operation index, the accuracy of each parameter group can be ensured, and the accuracy of the subsequently used parameter values of the trained neural network is improved.

[0103] As an optional implementation, in the embodiment, the method for determining the configuration parameters of the service cell further includes: if the target neuron node does not store the parameter configuration values of any normally operating service cell, at least one parameter configuration value of a normally operating service cell stored in a neuron node adjacent to the target neuron node is determined as the parameter configuration value corresponding to at least one parameter group of the service cell to be configured.

[0104] If the target neuron node of the service cell to be configured does not store the parameter configuration values of any normally operating service cell, at least one parameter configuration value of a normally operating service cell stored in a neuron node adjacent to the target neuron node and closest to the target neuron node is determined as the parameter configuration value corresponding to at least one parameter group of the service cell to be configured. For example, if the target neuron node of the coverage-related parameter group of the service cell to be configured does not store the parameter configuration values of any normally operating service cell, at least one parameter configuration value of a normally operating service cell stored in a neuron node adjacent to the target neuron node and closest to the target neuron node is determined as the parameter configuration value corresponding to the coverage-related parameter group of the service cell to be configured.

[0105] If the neuron node adjacent to the target neuron node and closest to the target neuron node does not store the parameter configuration values of any normally operating service cell, the next neuron node to the target neuron node is taken until a neuron node storing the parameter configuration values of any normally operating service cell is found.

[0106] In the embodiment, if the target neuron node does not store the parameter configuration value of at least one normally operating serving cell, the parameter configuration value of at least one normally operating serving cell stored in the neuron node adjacent to the target neuron node is determined as the parameter configuration value corresponding to the at least one parameter group of the serving cell to be configured. In the case that the target neuron node does not store the parameter configuration value, the parameter configuration value stored in the adjacent neuron node is obtained, so that the at least one parameter group of the serving cell to be configured can obtain the parameter configuration value.

[0107] As an optional implementation, in the embodiment, the method for determining the configuration parameter of the serving cell further includes: if any neuron node stores a plurality of parameter configuration values of normally operating serving cells, the mean or mode of the plurality of parameter configuration values of normally operating serving cells is determined as the plurality of parameter configuration values of normally operating serving cells stored in the any neuron node.

[0108] That is, if a neuron node stores a plurality of parameter configuration values of normally operating serving cells, the parameter configuration value stored in the neuron node is the mode or mean of the plurality of parameter configuration values of normally operating serving cells.

[0109] In the embodiment, if any neuron node stores a plurality of parameter configuration values of normally operating serving cells, the mean or mode of the plurality of parameter configuration values of normally operating serving cells is determined as the plurality of parameter configuration values of normally operating serving cells stored in the any neuron node. In the case that the neuron node has a plurality of parameter configuration values, the mean or mode of the plurality of parameter configuration values is stored, so that the parameter configuration is more standard and uniform when the serving cell to be configured is configured.

[0110] As an optional implementation, in the embodiment, after step 103, the following steps are further included:

[0111] The parameter configuration value corresponding to the at least one parameter group of the serving cell to be configured is input into the configuration system of the base station, so that the serving cell to be configured operates according to the parameter configuration value corresponding to the at least one parameter group of the serving cell to be configured.

[0112] The parameter configuration value corresponding to the at least one parameter group of the serving cell to be configured is input into the configuration system of the base station, so that the serving cell to be configured operates according to the parameter configuration value corresponding to the at least one parameter group of the serving cell to be configured.

[0113] When a plurality of parameter groups of the serving cell to be configured need to be set, the parameter configuration value corresponding to the plurality of parameter groups of the serving cell to be configured is input into the configuration system of the base station.

[0114] In the embodiment, the parameter configuration value corresponding to the at least one parameter group of the to-be-configured serving cell is input into the configuration system of the base station, so that the to-be-configured serving cell operates according to the parameter configuration value corresponding to the at least one parameter group of the to-be-configured serving cell. By inputting the parameter configuration value corresponding to the at least one parameter group of the to-be-configured serving cell into the configuration system of the base station, so that the to-be-configured serving cell operates according to the parameter configuration value, the subsequent normal operation of the to-be-configured serving cell can be ensured on the basis of ensuring the accuracy of the parameter configuration value, and the characteristics of the to-be-configured serving cell are combined, which is beneficial to providing better communication services for users.

[0115] Figure 4 is a structural schematic diagram of a configuration parameter determination apparatus of a serving cell provided in an embodiment of the present application, as shown in Figure 4 The configuration parameter determination apparatus 40 of the serving cell provided in the embodiment is located in an electronic device, and the configuration parameter determination apparatus 40 of the serving cell provided in the embodiment includes a variable value acquisition module 41, a neuron determination module 42, and a parameter configuration determination module 43. Among them:

[0116] The variable value acquisition module 41 is configured to acquire a plurality of characteristic variable values of at least one parameter group of a to-be-configured serving cell, and each parameter group corresponds to a trained neural network;

[0117] The neuron determination module 42 is configured to input the plurality of characteristic variable values of the at least one parameter group of the to-be-configured serving cell into the corresponding trained neural network, so as to determine a target neuron node mapped by the to-be-configured serving cell; the target neuron node is a neuron node mapped by a serving cell in normal operation which is most similar to the plurality of characteristic variable values of the at least one parameter group of the to-be-configured serving cell; and a plurality of neuron nodes of an output layer of the trained neural network store parameter configuration values of at least one serving cell in normal operation which have a mapping relationship with the plurality of neuron nodes;

[0118] The parameter configuration determination module 43 is configured to determine the parameter configuration values of the at least one serving cell in normal operation stored by the target neuron node as the parameter configuration values corresponding to the at least one parameter group of the to-be-configured serving cell.

[0119] Optionally, the configuration parameter determination apparatus of the serving cell further comprises a parameter group determination module configured to: obtain a plurality of cell data of the serving cell to be configured before obtaining a plurality of characteristic variable values of at least one parameter group of the serving cell to be configured, wherein the plurality of cell data of the serving cell to be configured at least comprises configuration data of a newly-built base station and survey data of the newly-built base station; and determine a parameter group corresponding to each cell data of the serving cell to be configured according to the plurality of cell data of the serving cell to be configured and a pre-stored parameter table, wherein the parameter table stores a mapping relationship between the cell data and the parameter group.

[0120] Optionally, when the parameter group determination module obtains the plurality of characteristic variable values of at least one parameter group of the serving cell to be configured, the parameter group determination module is specifically configured to: obtain characteristic variables of at least one parameter group of the serving cell to be configured; and perform standardization processing on the characteristic variables of at least one parameter group of the serving cell to be configured to obtain the plurality of characteristic variable values of at least one parameter group of the serving cell to be configured.

[0121] Optionally, the characteristic variables of at least one parameter group of the serving cell to be configured are numerical characteristic variables; and when the parameter group determination module performs standardization processing on the characteristic variables of at least one parameter group of the serving cell to be configured to obtain the plurality of characteristic variable values of at least one parameter group of the serving cell to be configured, the parameter group determination module is specifically configured to: obtain a mean value and a standard deviation of the plurality of characteristic variables of at least one parameter group of the serving cell to be configured; and subtract the plurality of characteristic variables of at least one parameter group of the serving cell to be configured from the mean value and then divide by the standard deviation to obtain the plurality of characteristic variable values of at least one parameter group of the serving cell to be configured.

[0122] Optionally, the configuration parameter determination apparatus of the serving cell further comprises a model training module configured to: initialize an initial neural network corresponding to each parameter group and establish a corresponding initial learning rate before obtaining the plurality of characteristic variable values of at least one parameter group of the serving cell to be configured; obtain a training sample set corresponding to each initial neural network, wherein a sample in the training sample set is a plurality of characteristic variable values of a corresponding parameter group of a serving cell in normal operation; perform iterative competitive learning on each initial neural network through the samples of each training sample set, and update a corresponding learning rate; and if the learning rate of each initial neural network is less than a corresponding target learning rate, determine that a plurality of trained neural networks are obtained.

[0123] Optionally, when the model training module obtains the training sample set corresponding to each initial neural network, it is specifically used to: select multiple normally operating serving cells from multiple operating serving cells according to KPI operating indicators, wherein the KPI operating indicators are relevant indicators characterizing the operating status of the serving cells; obtain multiple parameter groups, wherein each parameter group is divided according to the functional role of the cell data of each normally operating serving cell; and obtain the training sample set corresponding to each initial neural network according to each parameter group.

[0124] Optionally, the parameter configuration determination module 43 is further configured to: if the target neuron node does not store the parameter configuration value of any normally operating serving cell, determine the parameter configuration value of at least one normally operating serving cell stored in the neuron node adjacent to the target neuron node as the parameter configuration value corresponding to at least one parameter group of the serving cell to be configured.

[0125] Optionally, the parameter configuration determination module 43 is further configured to: if any neuron node stores parameter configuration values ​​of multiple normally operating serving cells, determine the mean or mode of the parameter configuration values ​​of the multiple normally operating serving cells as the parameter configuration values ​​of the multiple normally operating serving cells stored in the any neuron node.

[0126] Optionally, the serving cell configuration parameter determination device further includes a cell operation module, configured to: after determining the parameter configuration values ​​of at least one normally operating serving cell stored in the target neuron node as the parameter configuration values ​​corresponding to at least one parameter group of the serving cell to be configured, input the parameter configuration values ​​corresponding to at least one parameter group of the serving cell to be configured into the configuration system of the base station, so that the serving cell to be configured operates according to the parameter configuration values ​​corresponding to at least one parameter group of the serving cell to be configured.

[0127] Figure 5 This is a block diagram illustrating an electronic device according to an exemplary embodiment, the device being as follows: Figure 5 As shown, the electronic device includes: a memory 51 and a processor 52; the memory 51 is a memory for storing processor-executable instructions; the processor 52 is used to run computer programs or instructions to implement the method for determining the configuration parameters of the serving cell provided in any of the above embodiments.

[0128] The memory 51 is used to store programs. Specifically, the program may include program code, which includes computer operation instructions. The memory 51 may include high-speed RAM, and may also include non-volatile memory, such as at least one disk storage device.

[0129] The processor 52 can be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to perform the operations of the embodiments of the present disclosure.

[0130] Optionally, in a specific implementation, if the memory 51 and the processor 52 are implemented independently, the memory 51 and the processor 52 can be connected to each other through a bus 53 and complete communication between each other. The bus 53 can be an industry standard architecture (ISA) bus 53, a peripheral component (PCI) bus 53, or an extended industry standard architecture (EISA) bus 53, etc. The bus 53 can be divided into an address bus 53, a data bus 53, a control bus 53, etc. For ease of representation, Figure 5 In the figure, only one thick line is used to represent the bus 53, but it does not mean that there is only one bus 53 or only one type of bus 53.

[0131] Optionally, in a specific implementation, if the memory 51 and the processor 52 are integrated on a chip, the memory 51 and the processor 52 can complete communication between each other through an internal interface.

[0132] A non-transitory computer readable storage medium, when the instructions in the storage medium are executed by a processor of an electronic device, enable the electronic device to perform the method for determining a configuration parameter of a serving cell of the electronic device.

[0133] Other embodiments of the application will be apparent to those skilled in the art from consideration of the specification and practice of the application disclosed herein. It is intended that the specification and examples be considered as exemplary only, with the true scope and spirit of the application being indicated by the following claims. Other embodiments of the application will be apparent to those skilled in the art from consideration of the specification and practice of the application disclosed herein. It is intended that the specification and examples be considered as exemplary only, with the true scope and spirit of the application being indicated by the following claims.

[0134] It should be understood that the application is not limited to the precise construction that has been described above and illustrated in the accompanying drawings, and that various modifications and changes can be made by those skilled in the art without departing from the scope of the application. The scope of the application should only be limited by the appended claims.

Claims

1. A method for determining configuration parameters of a serving cell, characterized in that, The method includes: Obtain multiple feature variable values ​​of at least one parameter group of the serving cell to be configured, each parameter group corresponding to a trained neural network, wherein the trained neural network is a self-organizing map neural network. Multiple feature variable values ​​of at least one parameter group of the serving cell to be configured are input into the corresponding pre-trained neural network to determine the target neuron node mapped by the serving cell to be configured; the target neuron node is the neuron node mapped by the normally operating serving cell that is most similar to the multiple feature variable values ​​of at least one parameter group of the serving cell to be configured; and the multiple neuron nodes of the output layer of the pre-trained neural network store the parameter configuration values ​​of at least one normally operating serving cell that have a mapping relationship with the multiple neuron nodes. The parameter configuration values ​​of at least one normally operating serving cell stored in the target neuron node are determined as the parameter configuration values ​​corresponding to at least one parameter group of the serving cell to be configured. Before obtaining the values ​​of multiple feature variables of at least one parameter group of the serving cell to be configured, the method further includes: Initialize the neural network corresponding to each parameter group and establish the corresponding initial learning rate; Obtain the training sample set corresponding to each initial neural network, wherein the samples in the training sample set are multiple feature variable values ​​of the parameter group corresponding to the serving cell in normal operation; Each initial neural network is iteratively and competitively learned using samples from each of the training sample sets, and the corresponding learning rate is updated accordingly. If the learning rate of each initial neural network is less than the corresponding target learning rate, then multiple trained neural networks are determined to be obtained. The neural network, through self-organization, adjusts the weights of the output layer neurons based on the training samples, making each neuron in the output layer a nerve cell sensitive to specific pattern inputs. This results in the formation of an ordered feature map in the output layer that reflects the distribution of sample pattern classes. The ordered feature map is a mapping relationship between the neuron nodes of the output layer of the trained neural network and the serving cells in normal operation.

2. The method according to claim 1, characterized in that, Before obtaining the values ​​of multiple feature variables for at least one parameter group of the serving cell to be configured, the method further includes: Obtain multiple cell data of the serving cell to be configured, wherein the multiple cell data of the serving cell to be configured includes at least the configuration data of the corresponding newly built base station and the survey data of the newly built base station; Based on the multiple cell data of the serving cell to be configured and the pre-stored parameter table, the parameter group corresponding to each cell data in the serving cell to be configured is determined. The parameter table stores the mapping relationship between cell data and parameter groups.

3. The method according to claim 2, characterized in that, The step of obtaining multiple feature variable values ​​of at least one parameter group of the serving cell to be configured includes: Obtain the feature variables of at least one parameter group of the serving cell to be configured; The feature variables of at least one parameter group of the serving cell to be configured are standardized to obtain multiple feature variable values ​​of at least one parameter group of the serving cell to be configured.

4. The method according to claim 3, characterized in that, At least one parameter group of the serving cell to be configured has numerical feature variables. The standardization process for the feature variables of at least one parameter group of the serving cell to be configured, to obtain multiple feature variable values ​​for at least one parameter group of the serving cell to be configured, includes: Obtain the mean and standard deviation of multiple feature variables of at least one parameter group of the serving cell to be configured; The values ​​of multiple feature variables of at least one parameter group of the serving cell to be configured are obtained by subtracting the mean from the mean and then dividing by the standard deviation.

5. The method according to claim 1, characterized in that, The step of obtaining the training sample set corresponding to each initial neural network includes: Based on the KPI performance indicators, multiple serving cells that are operating normally are selected from multiple operating serving cells. The KPI performance indicators are relevant indicators that characterize the operating status of the serving cells. Multiple parameter groups are obtained, each parameter group being divided according to the functional role of the cell data of each normally operating serving cell; Based on each set of parameters, the training sample set corresponding to each initial neural network is obtained.

6. The method according to claim 5, characterized in that, The method further includes: If the target neuron node does not store the parameter configuration value of any normally operating serving cell, then the parameter configuration value of at least one normally operating serving cell stored in the neuron node adjacent to the target neuron node is determined as the parameter configuration value corresponding to at least one parameter group of the serving cell to be configured.

7. The method according to any one of claims 1-6, characterized in that, If any neuron node stores parameter configuration values ​​of multiple normally operating serving cells, then the mean or mode of the parameter configuration values ​​of the multiple normally operating serving cells is determined as the parameter configuration values ​​of the multiple normally operating serving cells stored in the neuron node.

8. The method according to claim 7, characterized in that, After determining the parameter configuration values ​​of at least one normally operating serving cell stored in the target neuron node as the parameter configuration values ​​corresponding to at least one parameter group of the serving cell to be configured, the method further includes: The parameter configuration values ​​corresponding to at least one parameter group of the serving cell to be configured are input into the configuration system of the base station so that the serving cell to be configured operates according to the parameter configuration values ​​corresponding to at least one parameter group of the serving cell to be configured.

9. A device for determining configuration parameters of a serving cell, characterized in that, The device includes: The variable value acquisition module is used to acquire multiple feature variable values ​​of at least one parameter group of the serving cell to be configured, each parameter group corresponds to a trained neural network, wherein the trained neural network is a self-organizing map neural network. A neuron determination module is used to input multiple feature variable values ​​of at least one parameter group of the serving cell to be configured into a corresponding pre-trained neural network to determine the target neuron node mapped by the serving cell to be configured; the target neuron node is the neuron node mapped by the normally operating serving cell that is most similar to the multiple feature variable values ​​of at least one parameter group of the serving cell to be configured; and the multiple neuron nodes of the output layer of the pre-trained neural network store the parameter configuration values ​​of at least one normally operating serving cell that has a mapping relationship with the neuron node; The parameter configuration determination module is used to determine the parameter configuration values ​​of at least one normally operating serving cell stored in the target neuron node as the parameter configuration values ​​corresponding to at least one parameter group of the serving cell to be configured. The model training module is used to initialize the initial neural network corresponding to each parameter group and establish the corresponding initial learning rate before obtaining the multiple feature variable values ​​of at least one parameter group of the serving cell to be configured; obtain the training sample set corresponding to each initial neural network, wherein the samples in the training sample set are the multiple feature variable values ​​of the corresponding parameter group of the serving cell in normal operation; perform iterative competitive learning on each initial neural network using the samples of each training sample set, and update the corresponding learning rate; if the learning rate of each initial neural network is less than the corresponding target learning rate, then multiple pre-trained neural networks are determined to be obtained. The neural network, through self-organization, adjusts the weights of the output layer neurons based on the training samples, making each neuron in the output layer a nerve cell sensitive to specific pattern inputs. This results in the formation of an ordered feature map in the output layer that reflects the distribution of sample pattern classes. The ordered feature map is a mapping relationship between the neuron nodes of the output layer of the trained neural network and the serving cells in normal operation.

10. An electronic device, comprising: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1-8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-8.

Images