CN116684903A

CN116684903A - Cell parameter processing method, device, equipment and storage medium

Info

Publication number: CN116684903A
Application number: CN202310816014.XA
Authority: CN
Inventors: 于洋; 祝海亮; 周奕昕; 赵煜; 盛莉莉; 黎越
Original assignee: China United Network Communications Group Co Ltd
Current assignee: China United Network Communications Group Co Ltd
Priority date: 2023-07-04
Filing date: 2023-07-04
Publication date: 2023-09-01

Abstract

The application provides a cell parameter processing method, device, equipment and storage medium, and belongs to the technical field of communication. The method comprises the following steps: receiving an optimization instruction sent by a terminal device, and converting the optimization instruction into an instruction characteristic by adopting a natural language understanding model; determining a target optimization strategy according to the instruction characteristics; acquiring data of each cell, wherein the cell data comprises cell type, cell parameters and operation data; determining policy details corresponding to each cell according to the target optimization policy and the cell type of each cell; determining at least one cell to be optimized according to the policy details and the operation data of each cell; and adjusting the cell parameters of each cell to be optimized according to the policy details of the cell to be optimized. The method solves the problem of high time cost of manually optimizing the cell parameters of the base station.

Description

Cell parameter processing method, device, equipment and storage medium

Technical Field

The present application relates to the field of communications technologies, and in particular, to a method, an apparatus, a device, and a storage medium for processing cell parameters.

Background

With the continuous development of information technology, a large number of services can be performed on a mobile network, and the smoothness of the mobile network becomes an important factor affecting the network service effect.

Currently, in order to increase the speed and stability of a mobile network in the prior art, it is generally necessary to manually analyze parameters of a base station cell, so as to optimize the base station cell.

However, the inventors found that at least the following technical problems exist in the prior art: the time cost for manually optimizing the cell parameters of the base station is high.

Disclosure of Invention

The application provides a cell parameter processing method, device, equipment and storage medium, which are used for solving the problem of high time cost of manually optimizing cell parameters of a base station.

In a first aspect, the present application provides a cell parameter processing method, including:

and receiving an optimization instruction sent by the terminal equipment, and converting the optimization instruction into instruction characteristics by adopting a natural language understanding model. And determining a target optimization strategy according to the instruction characteristics. And acquiring data of each cell, wherein the cell data comprises cell type, cell parameters and operation data. And determining policy details corresponding to each cell according to the target optimization policy and the cell type of each cell. And determining at least one cell to be optimized according to the policy details and the operation data of each cell. And adjusting the cell parameters of each cell to be optimized according to the policy details of the cell to be optimized.

In one possible implementation, the policy details include numerical thresholds and weights corresponding to various types of operational data. Correspondingly, determining at least one cell to be optimized according to the policy details and the operation data of each cell, including: and multiplying various operation data of the target cell by the corresponding weight to obtain various weighted values, wherein the target cell is any cell. And adding various weighted values of the target cell to obtain a numerical sum. And if the numerical sum is larger than the numerical threshold, determining the target cell as the cell to be optimized.

In one possible implementation, the policy details include optimization actions and decision criteria. Correspondingly, optimizing the cell parameters of each cell to be optimized according to the policy details and the cell parameters of the cell to be optimized, including: and adjusting the cell parameters of each cell to be optimized by applying the optimizing action to obtain the parameters to be determined of each cell to be optimized. And controlling each cell to be optimized to operate with the parameters to be determined. And acquiring new operation data of each cell to be optimized. And if the new operation data of the target cell to be adjusted does not meet the preset operation requirement, controlling the target cell to be adjusted to operate with the corresponding cell parameter, wherein the target cell to be adjusted is any cell to be adjusted. And if the new operation data of the target cell to be adjusted meets the preset operation requirement and meets the judgment standard, controlling the target cell to be adjusted to operate with the parameters to be determined. If the new operation data of the target cell to be adjusted meets the preset operation requirement and does not meet the judgment standard, the step of adjusting the cell parameters of the cell to be optimized to judge whether the new operation data meets the preset operation requirement is re-executed until the new operation data meets the preset operation requirement.

In one possible implementation manner, after the step of re-executing the step of adjusting the cell parameters of the cell to be optimized to determine whether the new operation data meets the preset operation requirement, the method further includes: if the step of adjusting the cell parameters of the cell to be optimized is re-executed to judge whether the new operation data accords with the preset operation requirement for m times, the new operation data does not accord with the preset operation requirement, and the target cell to be adjusted is controlled to operate with the corresponding cell parameters, wherein m is a positive integer.

In one possible implementation, determining a target optimization strategy based on instruction characteristics includes: inputting the instruction features into a feature conversion model obtained through pre-training to obtain a target optimization strategy.

In one possible implementation manner, determining policy details corresponding to each cell according to the target optimization policy and a cell type of each cell includes: and inputting a target optimization strategy and the cell type of the target cell into a preset strategy detail generation model to obtain the strategy details of the target cell, wherein the target cell is any cell.

In one possible implementation, before acquiring each cell data, the method further includes: and acquiring the operation data of each cell. And normalizing the operation data of each cell to obtain the normalized operation data of each cell. And clustering the cells by adopting standardized operation data of the cells to obtain cell types corresponding to the cells.

In a second aspect, the present application provides a cell parameter processing apparatus, including: the instruction conversion module is used for receiving the optimization instruction sent by the terminal equipment and converting the optimization instruction into instruction characteristics by adopting a natural language understanding model. And the strategy determining module is used for determining a target optimization strategy according to the instruction characteristics. The data acquisition module is used for acquiring the data of each cell, wherein the cell data comprises cell type, cell parameters and operation data. The detail determining module is used for determining policy details corresponding to each cell according to the target optimization policy and the cell type of each cell. And the cell determining module is used for determining at least one cell to be optimized according to the policy details and the operation data of each cell. And the cell optimization module is used for adjusting the cell parameters of each cell to be optimized according to the policy details of the cell to be optimized.

In a third aspect, the present application provides an electronic device comprising: a processor, and a memory communicatively coupled to the processor. The memory stores computer-executable instructions. The processor executes computer-executable instructions stored in the memory to cause the processor to perform the cell parameter processing method as described in the first aspect.

In a fourth aspect, the present application provides a computer-readable storage medium having stored therein computer-executable instructions which, when executed by a processor, are adapted to carry out the cell parameter processing method as described in the first aspect.

The cell parameter processing method, the device, the equipment and the storage medium provided by the application convert the optimization instruction into the instruction characteristic by receiving the optimization instruction sent by the terminal equipment, find the corresponding target optimization strategy by adopting the instruction characteristic, obtain the strategy details corresponding to each cell by the target optimization strategy and the cell type of each cell, screen the cells to be optimized in all the cells by the strategy details and the operation data corresponding to each cell, and adjust the cell parameters of each cell to be optimized by the strategy details of the cells to be optimized, thereby realizing automatic adjustment of the cell parameters of the cells, adopting different adjustment modes aiming at different cells, and reducing the time cost of adjusting the cell parameters by staff.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the application and together with the description, serve to explain the principles of the application.

Fig. 1 is a schematic diagram of an application scenario of a cell parameter processing method according to an embodiment of the present application;

fig. 2 is a flow chart of a cell parameter processing method according to an embodiment of the present application;

FIG. 3 is a schematic diagram of an instruction feature obtained by converting an optimized instruction according to an embodiment of the present application;

fig. 4 is a schematic diagram of a process for adjusting cell parameters of a cell to be optimized according to an embodiment of the present application;

fig. 5 is a schematic structural diagram of a cell parameter processing device according to an embodiment of the present application;

fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present application.

Specific embodiments of the present application have been shown by way of the above drawings and will be described in more detail below. The drawings and the written description are not intended to limit the scope of the inventive concepts in any way, but rather to illustrate the inventive concepts to those skilled in the art by reference to the specific embodiments.

Detailed Description

Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary examples do not represent all implementations consistent with the application. Rather, they are merely examples of apparatus and methods consistent with aspects of the application as detailed in the accompanying claims.

With the continuous development of network services, a mobile network becomes an important way for providing services to users, and in order to ensure the stability of the mobile network, parameters of a base station cell need to be adjusted, so as to improve the coverage area and signal strength of the base station cell.

Currently, the adjustment of the cell parameters of the base station is usually performed manually, and the manual adjustment of the cell parameters of the base station requires workers to perform on site, so that the time cost is high.

In order to solve the technical problems, the inventor proposes the following technical ideas: receiving an instruction input by a worker in a terminal device, converting the instruction into an instruction characteristic, obtaining a corresponding optimization strategy according to the instruction characteristic, determining different strategy details by combining the optimization strategy and a cell type, finding a cell to be optimized in all cells by adopting the strategy details and the operation data of the cell, and adjusting the cell parameters of the cell to be optimized by adopting the strategy details.

Fig. 1 is a schematic application scenario diagram of a cell parameter processing method according to an embodiment of the present application. As in fig. 1, in this scenario, it includes: terminal equipment 101, server 102, data acquisition device 103.

The terminal device 101 may include a computer, a server, a tablet, a mobile phone, a palm top (Personal Digital Assistant, PDA), a notebook, etc., which can input, transmit, etc., data.

The server 102 may be implemented by a server or a cluster of multiple servers, and may be replaced by a computer with a relatively high computing power, a notebook computer, or the like.

The data acquisition device 103 may comprise a server and may also comprise a database.

In a specific implementation process, the terminal device 101 is configured to receive an optimization instruction input by a worker or a user, and send the added optimization instruction to the server 102.

A server 102, configured to read cell data corresponding to each cell from the data acquisition device 103. And the cell data and the optimizing instruction are used for adjusting the cell parameters of each cell.

It will be appreciated that the structure illustrated in the embodiments of the present application does not constitute a specific limitation on the cell parameter processing method. In other possible embodiments of the present application, the architecture may include more or less components than those illustrated, or some components may be combined, some components may be split, or different component arrangements may be specifically determined according to the actual application scenario, and the present application is not limited herein. The components shown in fig. 1 may be implemented in hardware, software, or a combination of software and hardware.

The following describes the technical scheme of the present application and how the technical scheme of the present application solves the above technical problems in detail with specific embodiments. The following embodiments may be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments. Embodiments of the present application will be described below with reference to the accompanying drawings.

Fig. 2 is a flowchart of a cell parameter processing method according to an embodiment of the present application. The execution subject of the embodiment of the present application may be the server 102 in fig. 1, or may be a computer and/or a mobile phone, which is not particularly limited in this embodiment. As shown in fig. 2, the method includes: step S201 to step S206.

S201: and receiving an optimization instruction sent by the terminal equipment, and converting the optimization instruction into instruction characteristics by adopting a natural language understanding model.

In this step, the optimization instruction may be sent in the format of a message or a character string. The optimization instruction can be obtained by a user inputting natural language through voice or characters or by dragging, clicking and selecting keywords on a humanized graphical interface.

Converting the optimization instructions into instruction features may include generalizing the implementation of the optimization instructions to perform some operation on some object to achieve or avoid some result. The wireless network optimization intent can thus be expressed as an abstract model in the form of [ results, operations, objects ]. The "result" tag refers to the service state that the user expects the network to reach; the operation label refers to a networking strategy formulated in a network in order to realize service requirements; the "object" tag refers to the physical device and related resources that need to be manipulated in order to execute the networking policy.

For example, the optimization instruction is: the wireless connection rate of the area A is excellent, an optimization instruction is input into a natural language understanding model, the obtained object is the area A, the wireless connection rate is the operation is the achievement, and the target is excellent.

Fig. 3 is a schematic instruction feature diagram obtained by converting an optimized instruction according to an embodiment of the present application. As shown in fig. 3, the converted instruction features include "service type", "performance index", "expected state", "space-time constraint", "topology", "access mode", "resource allocation", "network constraint", "physical node", "radio resource", "storage resource", "computing resource" in the "object" in the "result".

S202: and determining a target optimization strategy according to the instruction characteristics.

In this step, a corresponding relationship between a preset instruction feature and a target optimization strategy may be searched according to the instruction feature, so as to obtain the target optimization strategy. The correspondence between the instruction features and the target optimization strategy may be preset by a worker.

S203: and acquiring data of each cell, wherein the cell data comprises cell type, cell parameters and operation data.

In this step, each cell data is acquired, and the cell data of each cell may be read from the data acquisition device 103. The process of reading the data of each cell may be to use a pre-stored cell identifier to read the corresponding cell data.

The cell type may be obtained by pre-clustering, and the cell parameters may include an elevation angle of the cell, a position of the cell, a direction of the cell, and the like. The operational data may include setup success rate, drop rate, rebuild rate, etc.

S204: and determining policy details corresponding to each cell according to the target optimization policy and the cell type of each cell.

According to the target optimization strategy and the cell type, searching the corresponding relation between the preset optimization strategy, the cell type and the strategy details to obtain the strategy details corresponding to each cell.

The corresponding relation between the optimization strategy, the cell type and the strategy details comprises strategy details corresponding to each group of optimization strategies and cell types. Policy details may include policy identification, monitoring objects, policy enforcement points, policy trigger conditions, policy operations, etc. Policy identification is used to define and identify a policy rule. The monitoring object indicates that the policy monitoring object is a characteristic value capable of indicating the current operation state of the network. The policy enforcement point represents an accessible network element node, a base station cell, in the policy enforcement point network. The policy triggering condition represents a predefined network state of the triggering policy, for example, the performance index is smaller than a certain threshold, the packet loss rate is larger than a certain threshold, the cell unavailable time is larger than a certain time threshold, etc., and the policy triggering condition can be a policy event triggering mechanism, a policy event and a policy condition, or can be a complex and combined analysis method of the policy event and the condition. The policy operations represent adjustment actions to be performed according to policy conditions, such as switching parameter adjustment, power adjustment, antenna adjustment, and the like. The policy execution is to execute the action of a certain policy when the trigger condition of the policy is met, and the policy operation is based on the network state perceived in real time, so as to dynamically execute the configuration action.

S205: and determining at least one cell to be optimized according to the policy details and the operation data of each cell.

In this step, the cell meeting the policy triggering condition may be determined by using the policy triggering condition and the operation data in the policy details corresponding to each cell. And determining the cell meeting the strategy triggering condition as the cell to be optimized.

S206: and adjusting the cell parameters of each cell to be optimized according to the policy details of the cell to be optimized.

In this step, the cell parameters of each cell to be optimized may be adjusted by using the above policy operations, for example, handover parameter adjustment, power adjustment, antenna adjustment, etc.

As can be seen from the description of the above embodiments, in the embodiment of the present application, by receiving an optimization instruction sent by a terminal device, converting the optimization instruction into an instruction feature, finding a corresponding target optimization policy by using the instruction feature, obtaining policy details corresponding to each cell by using the target optimization policy and a cell type of each cell, screening out cells to be optimized in all the cells by using the policy details and operation data corresponding to each cell, and adjusting cell parameters of each cell to be optimized by using the policy details of the cell to be optimized, thereby implementing automatic adjustment of cell parameters of the cells, and reducing time cost for adjusting cell parameters by staff by adopting different adjustment modes for different cells.

In one possible implementation, in step S201, the natural language understanding model may use a named entity recognition (Named Entity Recognition, NER) algorithm or incorporate a BLSTM (Bidirectional Long Short-Term Memory, two-way long and short Term Memory neural network). The natural language understanding model includes three subtasks: domain identification, intent identification, and slot filling. The number of nodes that the BLSTM uses to memorize and store past states may be 64.

Domain recognition, intent recognition, is typically modeled as a sentence-based classification problem that determines to which domain the user input belongs. In different fields, the system may be designed for different kinds of intents and different tanks. After the domain information is identified, the intention input by the user is judged and classified, so that the calculated amount is reduced, the matching efficiency of the system is improved, and the machine is facilitated to make a next decision. Whereas slot filling is based on word or word classification, the purpose of semantic slot filling is to extract important information related to a task entered by a user. The slot name is usually a semantic tag oriented to a specific task and is important content that a machine wants to acquire in the interaction process.

For example, if the optimization instruction is "how much the voice call drop rate is in VoLTE in south tokyo today", the result after the natural language model processing should be: intention field: "performance intent", intent classification: "query index", semantic slot filling: "{ time: today's site: nanjing index: voLTE dropped call rate }). Wherein VoLTE is a Voice over Long-Term Evolution (Voice over Long-Term Evolution).

The essence of semantic slot filling can be seen as named entity recognition (Named Entity Recognition, NER) in the field of natural language processing, i.e. by recognizing entities in text that have a specific meaning and assigning each word a corresponding named entity tag. NER finds out the vocabulary consistent with or relevant to the key element expression in the user intention input as the intention key word through the natural language processing procedures such as part-of-speech analysis, language word segmentation, dictionary inquiry and the like. And labeling key element labels of the feature conversion model according to the category of the named entity to which the key words belong, and laying a foundation for final translation of the intention. And selecting a joint recognition algorithm of combining the bidirectional cyclic neural network with the conditional random field. The two-way dependency of the text context can be captured, the constraint condition of the sequence annotation context can be learned, the attention mechanism can be added for improving the performance, and the time slot gating mechanism can be adopted.

In one possible implementation, the policy details include numerical thresholds and weights corresponding to various types of operational data.

The numerical threshold and the weights corresponding to the various parameters may be parameters in the policy triggering condition. There are many types of operational data.

Accordingly, in the step S205, at least one cell to be optimized is determined according to the policy details and the operation data of each cell, including:

S2051: and multiplying the data values of various operation data of the target cell with the corresponding weights to obtain various weighted values, wherein the target cell is any cell.

In this step, the running data, such as packet loss rate and call drop rate, may be preset by the staff, or may be pre-generated by the program.

S2052: and adding various weighted values of the target cell to obtain a numerical sum.

In this step, for example, the respective weighted values of the target cell are 0.3, 0.5 and 0.7, respectively, and the sum of the values is 1.5.

Step S2051 and step S2052 can be expressed as the following formulas:

X＝k ₁ x ₁ +k ₂ x ₂ +k ₃ x ₃ +…+k _n x _n

wherein X represents the sum of values, k _n Weights, x, representing class n operational data _n A data value representing the n-th class of operational data. n represents the total number of kinds of operation data.

S2053: and if the numerical sum is larger than the numerical threshold, determining the target cell as the cell to be optimized.

In this step, the numerical sum may be preset by a worker, or may be obtained by program optimization.

As can be seen from the description of the foregoing embodiments, in the embodiments of the present application, the operation data of the target cell is multiplied by the corresponding weights to obtain weighted values, and the weighted values are added to obtain a numerical sum, and if the numerical sum is greater than the numerical threshold, the target cell is determined to be the cell to be optimized, so that the effect of screening the cell to be optimized from all the cells by using the operation data of the cell and reducing the subsequent calculation amount is achieved.

In a possible implementation manner, the policy details include numerical thresholds corresponding to various types of operation data, and accordingly, in step S205, at least one cell to be optimized is determined according to the policy details and the operation data of each cell, including: and if any type of operation data of the target cell exceeds the corresponding numerical threshold, determining the target cell as the cell to be optimized.

Wherein the numerical threshold may be greater than a certain value or less than a certain value, and exceeding the numerical threshold may be a condition that the numerical threshold is met.

In one possible implementation, the policy details include optimization actions and decision criteria.

The optimization action may be an adjustment action of a cell parameter, for example, decreasing the cell elevation angle by 5 °, increasing the horizontal orientation angle by 3 °, or the like.

Accordingly, in the step S206, optimizing the cell parameters of each cell to be optimized according to the policy details and the cell parameters of the cell to be optimized, including: steps S2061 to S2065.

S2061: and adjusting the cell parameters of each cell to be optimized by applying the optimizing action to obtain the parameters to be determined of each cell to be optimized.

In this step, the cell parameters of each cell to be optimized may be changed by using the optimization actions corresponding to each cell to be optimized, and the obtained new cell parameters are determined as the parameters to be determined.

For example, the original cell parameter of the target cell is 15 ° elevation angle, the optimization action is to reduce the elevation angle by 3 °, and the obtained parameter to be determined is 12 ° elevation angle. For another example, if the original cell parameter of the target cell is 100 ° horizontal direction angle and the optimization is to increase the horizontal direction angle by 5 °, the obtained parameter to be determined is 105 °.

S2062: and controlling each cell to be optimized to operate with the parameters to be determined.

In this step, the cell parameters of each cell to be optimized may be adjusted, so that the cell parameters of each cell to be optimized are the same as the corresponding parameters to be determined, and the cell parameters are operated.

S2063: and acquiring new operation data of each cell to be optimized.

In this step, the identifier of the cell to be optimized may be used to read the corresponding newly added operation data from the data acquisition device. Or the identification of the cell to be optimized can directly send the operation data acquisition request to the cell to be optimized and receive the operation data sent by the cell to be optimized.

The newly added operation data may be operation data generated after the cell to be optimized operates with the parameter to be determined.

S2064: and if the new operation data of the target cell to be adjusted does not meet the preset operation requirement, controlling the target cell to be adjusted to operate with the corresponding cell parameter, wherein the target cell to be adjusted is any cell to be adjusted.

In this step, the preset operation requirement may include basic performance indexes such as a set-up success rate, a dropped-line rate, a rebuilding rate, and the like. And controlling the target cell to be adjusted to return the cell parameters to be adopted to the original cell parameters by the corresponding cell parameters.

The preset operation requirements are for example, the disconnection rate is less than 5%, the reconstruction rate is greater than 90%, the connection rate is greater than 99.8%, the switching success rate is greater than 99%, etc.

In a possible implementation manner, the method further comprises the step of adding the identification of the target cell to be adjusted to the list to be optimized after the step, so that the manual optimization is more convenient to follow.

S2065: and if the new operation data of the target cell to be adjusted meets the preset operation requirement and meets the judgment standard, controlling the target cell to be adjusted to operate with the parameters to be determined.

In this step, the controlling the target cell to be adjusted to operate with the parameter to be determined may be sending a control instruction to the target cell to be adjusted, so that the target cell to be adjusted operates with the parameter to be determined in the control instruction.

S2066: if the new operation data of the target cell to be adjusted meets the preset operation requirement and does not meet the judgment standard, the step of adjusting the cell parameters of the cell to be optimized to judge whether the new operation data meets the preset operation requirement is re-executed until the new operation data meets the preset operation requirement.

In this step, operation requirements such as high packet loss on the air interface, high packet loss rate on the air interface, and non-low access cell are preset. The step of adjusting the cell parameters of the cell to be optimized to determine whether the new operation data meets the preset operation requirement may be the steps S2064 to S2065 described above.

The air interface identifier is a visual term, is a radio transmission specification between a base station and a mobile phone, and the packet loss rate is a data packet loss rate.

In a possible implementation manner, if the preset operation requirements are met after the steps S2064 to S2065 are re-performed for a preset number of times, the identification of the target cell to be adjusted is added to the list to be optimized, thereby facilitating the manual optimization in the following steps

As can be seen from the description of the foregoing embodiments, in the embodiments of the present application, the cell parameters of each cell to be optimized are adjusted by applying the optimization actions in the policy details, so as to obtain the parameters to be optimized of each cell to be optimized, the cell to be optimized is controlled to operate by the parameters to be optimized, new operation data is obtained, whether the obtained operation data of the target cell to be adjusted meets the preset operation requirement is judged, if not, the target cell to be adjusted is controlled to operate with the corresponding original cell parameters, if the obtained operation data meets the preset operation requirement, the target cell to be adjusted is controlled to operate with the parameters to be determined according to the judgment criterion in the policy details, and if the obtained operation data meets the preset operation requirement, the step of adjusting the cell parameters of the cell to be optimized to judge whether the new operation data meets the preset operation requirement is re-executed until the new operation data meets the preset operation requirement, the cell parameters of the iterative optimization cell are implemented, the stability and coverage rate of the mobile network are improved, and the time of the base station of the optimization cell is reduced.

Fig. 4 is a schematic diagram of a process for adjusting cell parameters of a cell to be optimized according to an embodiment of the present application. As shown in fig. 4, the process of adjusting the cell parameters of the cell to be optimized includes:

s401: parameter checking: cell data of each cell is obtained, and preset rules are adopted to normalize cell parameters.

The cell parameters may include handover threshold, timer setting, algorithm setting, power setting, etc.

S402: model setting: the method comprises the steps of classifying scenes of a whole network cell, setting personalized initial thresholds according to different scene strategies according to expert experience, and forming a strategy model of parameter adjustment by setting a parameter base line L, a single adjustment step length S and an adjustment offset f, wherein the steps are as follows:

setting the cell parameter value to be adjusted as Q, setting the parameter adjustment target value as D and setting the adjustment sequence as O. Wherein the initial value of the adjusting sequence is 1, 1 is overlapped once every cycle, and n times are overlapped at most according to the set value n, and the formula is as follows:

O＝O+1，1≤O≤O _n

if L-f is less than Q < L+f, setting the parameter adjustment target value as a parameter baseline value L, wherein the formula is as follows:

D＝L

if Q > L+f or Q < L-f, the parameter base line value L, the adjustment step S and the adjustment sequence O form a parameter adjustment target value D. Wherein the formula is as follows:

D＝L+S*O

The adjustment strategy is shown in table 1.

Table 1 adjustment strategy example table

Parameter name	Threshold value	Base line	Step size	Adjustment sequence	Target value
						A2	-115	-115	-2	1	-117
A4	-117	-120	-1	2	-119

S403: and (5) judging indexes. The index determination process is similar to the above steps S2052 to S2053, and will not be described here again.

S404: and (3) adjustment implementation: and adjusting the cell parameters by adopting a target optimization strategy and strategy details corresponding to the cell type.

S405: and (3) index observation: after the cell parameters are adjusted, judging whether the operation data of the cell meet the preset operation requirement or not in the time interval T1, if not, carrying out strategy rollback, operating with the original cell parameters, and writing the identification of the cell into a list to be manually interfered. If the preset operation requirement is met, step S406 is performed.

S406: and (3) secondary discrimination: and judging whether the judgment standard in the policy details is met, if so, controlling the target cell to be adjusted to run with the parameters to be determined, and if not, executing the step S407.

S407: and (3) cyclic adjustment: and re-executing the step of adjusting the cell parameters of the cell to be optimized to judge whether the new operation data meets the preset operation requirement or not until the new operation data meets the preset operation requirement or is still not met after repeated execution for n times, and adopting the original cell parameters for operation.

S408: and (3) data acquisition: after the step S406, if the judgment criteria in the policy details are met, the cell parameters are added to the training set.

S409: machine learning: and substituting different results of different strategies of the acquired certain scene into a machine learning model for iterative training to finally form a new full-network scene strategy, and applying the new full-network scene strategy to all cells of the scene to form a new parameter strategy.

In a possible implementation manner, after the step of re-executing the step of adjusting the cell parameters of the cell to be optimized to determine whether the new operation data meets the preset operation requirement in step S2066, the method further includes:

s2067: and if the step of adjusting the cell parameters of the cell to be optimized is re-executed to judge whether the new operation data accords with the preset operation requirement for m times, the new operation data does not accord with the preset operation requirement, and the target cell to be adjusted is controlled to operate with the corresponding cell parameters, wherein m is a positive integer.

In this step, the execution count may be incremented by 1 after each execution of steps S2064 to S2065 to obtain a new execution count, and when the execution count reaches m, if the new operation data still does not meet the preset operation requirement, the cell operation is controlled by using the cell parameters in step S203.

After this step, the identity of the target cell to be adjusted may also be added to the list to be optimized, so that the manual optimization is more convenient to follow.

As can be seen from the description of the above embodiment, in the embodiment of the present application, by adopting the original cell parameter to control the operation of the target cell when the cell parameter m is repeatedly adjusted and still does not meet the preset operation requirement, the cell parameter is restored under the condition that the cell optimization cannot meet the requirement for multiple times, and the effect of avoiding the decrease of the cell performance is achieved.

In a possible implementation manner, in step S202, determining, according to the instruction feature, a target optimization policy includes:

s2021: inputting the instruction features into a feature conversion model obtained through pre-training to obtain a target optimization strategy.

In the step, the feature conversion model can be obtained by training a long-term and short-term memory network, a feedforward neural network, a feedback neural network and the like. In the feature conversion model, the encoder and the decoder can both be composed of a layer of long-short-term memory neural network, and the number of hidden layers can be 32, or more or less hidden layers can be provided. During training, the effect of the model is assessed using bilingual assessment alternatives (Bilingual Evaluation Understudy, BLEU) and cross entropy loss functions. The feature conversion model can be trained by adopting the existing experimental data, wherein the experimental data comprises instruction features obtained through experiments and corresponding optimization strategies.

In a possible implementation manner, the step S202 may include searching, according to the instruction features, a corresponding relationship between preset instruction features and an optimization policy, to obtain a corresponding target optimization policy.

As can be seen from the description of the above embodiment, in the embodiment of the present application, the instruction feature is input into the feature conversion model to obtain the target optimization strategy, so that the optimization strategy obtained by the instruction feature is realized, and when the cell optimization is required, the input instruction obtains the corresponding optimization strategy, thereby reducing the time for manually searching the optimization strategy.

In a possible implementation manner, in the step S204, determining policy details corresponding to each cell according to the target optimization policy and the cell type of each cell includes:

s2041: and inputting a target optimization strategy and the cell type of the target cell into a preset strategy detail generation model to obtain the strategy details of the target cell, wherein the target cell is any cell.

In this step, the policy detail generating model may be a feedforward neural network or a feedback neural network, or may be obtained by training the corresponding relation of the optimization policy, the cell type and the policy detail in the experimental data.

From the description of the above embodiment, it can be known that, in the embodiment of the present application, the target optimization strategy and the cell type of the target cell are input into the preset strategy detail generation model, so as to obtain strategy details, and different strategy details are obtained for different cell types and optimization strategies, so that the adjustment method of the cell parameters is more targeted.

In a possible implementation manner, the step S204 determines policy details corresponding to each cell according to the target optimization policy and the cell type of each cell, including:

s2042: searching a preset corresponding relation among the optimization strategy, the cell type and the strategy details according to the target optimization strategy and the cell type of the target cell to obtain the strategy details corresponding to the target optimization strategy and the cell type of the target cell.

The optimization strategies, the cell types and the corresponding relations of the strategy details can be stored in a table and key value peer-to-peer format, and each group of optimization strategies and the cell types correspond to one strategy detail.

In a possible implementation manner, before the step S203 of acquiring each cell data, the method further includes:

s210: and acquiring the operation data of each cell.

In this step, the identification of the cell may be used to read the corresponding operation data from the data acquisition device or database.

S211: and normalizing the operation data of each cell to obtain the normalized operation data of each cell.

In this step, it may include: filling null values of various data in the running data by adopting an average value of corresponding classes, converting the character string type into a numerical value type, respectively carrying out high call drop, low access, uplink high packet loss, respectively representing downlink high packet loss by numbers 1,2,3 and 4, converting the data type of an index column into a floating point type, and removing abnormal values in the running data. And may also include removing less important types of operational data from the operational data.

The abnormal value removal can be performed by using a preset library or script. Removing the less important type of operational data may be obtaining the importance of the data feature using a LASSO (least absolute shrinkage and selection algorithm, least Absolute Shrinkage and Selection Operator) algorithm, and removing the less important feature using the obtained importance, where the less important feature may be a preset feature with the lowest importance, and may also be z% feature types with the lowest importance, where z is an integer. The feature types with low importance can be removed, and the importance of various features can be drawn into pictures, and the removed feature types can be selected manually.

S212: and clustering the cells by adopting standardized operation data of the cells to obtain cell types corresponding to the cells.

In this step, the corresponding cell type may be obtained by clustering each cell by adopting a KNN (K Nearest Neighbor ) algorithm, a K-means (K-means) algorithm, or the like. The KNN algorithm can adopt parameters obtained by experimental data training, and the training data can be obtained according to the following steps of 8:2 is divided into a training set and a testing set, and in the process of training to obtain parameters, the model can be optimized in a cross-validation and grid search mode, so that the parameters of the model are obtained.

As can be seen from the description of the foregoing embodiments, in the embodiments of the present application, operation data of each cell is obtained by obtaining the operation data of each cell, and standardized operation data of each cell is obtained, and the standardized operation data is used to cluster each cell, so as to obtain a cell type corresponding to each cell, so that the cell parameters can be adjusted by using the cell type of each cell subsequently.

In the process of training a clustering algorithm, a training set is input into a model for training, the score of the model is calculated by using a score function, the accuracy of the model is determined by adopting a test set under the condition that the score is higher than a preset value, and the model is stored after the accuracy exceeds an accuracy threshold.

Fig. 5 is a schematic structural diagram of a cell parameter processing device according to an embodiment of the present application. As shown in fig. 5, the cell parameter processing apparatus 500 includes: instruction conversion module 501, policy determination module 502, data acquisition module 503, detail determination module 504, cell determination module 505, and cell optimization module 506.

The instruction conversion module 501 is configured to receive an optimization instruction sent by a terminal device, and convert the optimization instruction into an instruction feature by adopting a natural language understanding model.

The policy determining module 502 is configured to determine a target optimization policy according to the instruction features.

A data obtaining module 503, configured to obtain each cell data, where the cell data includes a cell type, a cell parameter, and operation data.

The detail determining module 504 is configured to determine policy details corresponding to each cell according to the target optimization policy and a cell type of each cell.

The cell determining module 505 is configured to determine at least one cell to be optimized according to policy details and operation data of each cell.

The cell optimization module 506 is configured to adjust cell parameters of each cell to be optimized according to policy details of the cell to be optimized.

The device provided in this embodiment may be used to implement the technical solution of the foregoing method embodiment, and its implementation principle and technical effects are similar, and this embodiment will not be described herein again.

In one possible implementation, the policy details include numerical thresholds and weights corresponding to various types of operational data. The cell determining module 505 is specifically configured to multiply each type of operation data of the target cell with a corresponding weight to obtain each type of weighted value, where the target cell is any cell. And adding various weighted values of the target cell to obtain a numerical sum. And if the numerical sum is larger than the numerical threshold, determining the target cell as the cell to be optimized.

In one possible implementation, the policy details include optimization actions and decision criteria. The cell optimization module 506 is configured to apply an optimization action to adjust cell parameters of each cell to be optimized, and obtain parameters to be determined of each cell to be optimized. And controlling each cell to be optimized to operate with the parameters to be determined. And acquiring new operation data of each cell to be optimized. And if the new operation data of the target cell to be adjusted does not meet the preset operation requirement, controlling the target cell to be adjusted to operate with the corresponding cell parameter, wherein the target cell to be adjusted is any cell to be adjusted. And if the new operation data of the target cell to be adjusted meets the preset operation requirement and meets the judgment standard, controlling the target cell to be adjusted to operate with the parameters to be determined. If the new operation data of the target cell to be adjusted meets the preset operation requirement and does not meet the judgment standard, the step of adjusting the cell parameters of the cell to be optimized to judge whether the new operation data meets the preset operation requirement is re-executed until the new operation data meets the preset operation requirement.

In one possible implementation, the cell parameter processing apparatus 500 further includes: parameter rollback module 507.

And a parameter rollback module 507, configured to control the target cell to be adjusted to operate with the corresponding cell parameter if the new operation data does not meet the preset operation requirement after the step of adjusting the cell parameter of the cell to be optimized to determine whether the new operation data meets the preset operation requirement is performed for m times, where m is a positive integer.

In one possible implementation, the policy determining module 502 is configured to input the instruction feature into a feature transformation model that is trained in advance, to obtain the target optimization policy.

In a possible implementation manner, the detail determining module 504 is configured to input a target optimization policy and a cell type of a target cell into a preset policy detail generating model to obtain policy details of the target cell, where the target cell is any cell.

In one possible implementation, the cell parameter processing apparatus 500 further includes: and the cell clustering module 508 is used for acquiring the operation data of each cell. And normalizing the operation data of each cell to obtain the normalized operation data of each cell. And clustering the cells by adopting standardized operation data of the cells to obtain cell types corresponding to the cells.

In order to achieve the above embodiment, the embodiment of the present application further provides an electronic device.

Referring to fig. 6, there is shown a schematic structural diagram of an electronic device 600 suitable for implementing an embodiment of the present application, where the electronic device 600 may be a terminal device or a server. The terminal device may include, but is not limited to, a mobile terminal such as a mobile phone, a notebook computer, a digital broadcast receiver, a personal digital assistant (Personal Digital Assistant, PDA for short), a tablet (Portable Android Device, PAD for short), a portable multimedia player (Portable Media Player, PMP for short), an in-vehicle terminal (e.g., an in-vehicle navigation terminal), and the like, and a fixed terminal such as a digital TV, a desktop computer, and the like. The electronic device shown in fig. 6 is only an example and should not be construed as limiting the functionality and scope of use of the embodiments of the application.

As shown in fig. 6, the electronic device 600 may include a processor (e.g., a central processing unit, a graphics processor, etc.) 601, and a Memory 602 communicatively connected to the processor, which may perform various appropriate actions and processes according to a program stored in the Memory 602, a computer executing instructions, or a program loaded from a storage 608 into a random access Memory (Random Access Memory, abbreviated as RAM) 603, to implement the cell parameter processing method in any of the above embodiments, where the Memory may be a Read Only Memory (ROM). In the RAM603, various programs and data required for the operation of the electronic apparatus 600 are also stored. The processing device 601, the memory 602, and the RAM603 are connected to each other via a bus 604. An input/output (I/O) interface 605 is also connected to bus 604.

In general, the following devices may be connected to the I/O interface 605: input devices 606 including, for example, a touch screen, touchpad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, and the like; an output device 607 including, for example, a liquid crystal display (Liquid Crystal Display, LCD for short), a speaker, a vibrator, and the like; storage 608 including, for example, magnetic tape, hard disk, etc.; and a communication device 609. The communication means 609 may allow the electronic device 600 to communicate with other devices wirelessly or by wire to exchange data. While fig. 6 shows an electronic device 600 having various means, it is to be understood that not all of the illustrated means are required to be implemented or provided. More or fewer devices may be implemented or provided instead.

In particular, according to embodiments of the present application, the processes described above with reference to flowcharts may be implemented as computer software programs. For example, embodiments of the present application include a computer program product comprising a computer program embodied on a computer readable storage medium, the computer program comprising program code for performing the method shown in the flowcharts. In such embodiments, the computer program may be downloaded and installed from a network via communication device 609, or from storage device 608, or from memory 602. The above-described functions defined in the method of the embodiment of the present application are performed when the computer program is executed by the processing means 601.

The computer readable storage medium of the present application may be a computer readable signal medium or a computer storage medium, or any combination of the two. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In the present application, however, the computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, with the computer-readable program code embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable storage medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable storage medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, fiber optic cables, RF (radio frequency), and the like, or any suitable combination of the foregoing.

The computer-readable storage medium may be contained in the electronic device; or may exist alone without being incorporated into the electronic device.

The computer-readable storage medium carries one or more programs which, when executed by the electronic device, cause the electronic device to perform the methods shown in the above-described embodiments.

Computer program code for carrying out operations of the present application may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, smalltalk, C ++ and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a local area network (Local Area Network, LAN for short) or a wide area network (Wide Area Network, WAN for short), or it may be connected to an external computer (e.g., connected via the internet using an internet service provider).

The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The modules involved in the embodiments of the present application may be implemented in software or in hardware. Where the names of the units do not constitute a limitation of the module itself in some cases, for example, the policy determination module may also be described as a "target optimization policy determination module".

The functions described above herein may be performed, at least in part, by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: a Field Programmable Gate Array (FPGA), an Application Specific Integrated Circuit (ASIC), an Application Specific Standard Product (ASSP), a system on a chip (SOC), a Complex Programmable Logic Device (CPLD), and the like.

The present application also provides a computer readable storage medium, in which computer executable instructions are stored, when a processor executes the computer executable instructions, the technical scheme of the cell parameter processing method in any of the above embodiments is implemented, and the implementation principle and the beneficial effects are similar to those of the cell parameter processing method, and can be seen from the implementation principle and the beneficial effects of the cell parameter processing method, which are not described herein.

In the context of the present application, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

The application also provides a computer program product, which comprises a computer program, when the computer program is executed by a processor, the technical scheme of the cell parameter processing method in any of the above embodiments is realized, the realization principle and the beneficial effects are similar to those of the cell parameter processing method, and the realization principle and the beneficial effects of the cell parameter processing method can be seen, and the detailed description is omitted herein.

The above description is only illustrative of the preferred embodiments of the present application and of the principles of the technology employed. It will be appreciated by persons skilled in the art that the scope of the disclosure referred to in the present application is not limited to the specific combinations of technical features described above, but also covers other technical features formed by any combination of the technical features described above or their equivalents without departing from the spirit of the disclosure. Such as the above-mentioned features and the technical features disclosed in the present application (but not limited to) having similar functions are replaced with each other.

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. This application is intended to cover any variations, uses, or adaptations of the application following, in general, the principles of the application and including such departures from the present disclosure as come within known or customary practice within the art to which the application pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the application being indicated by the following claims.

It is to be understood that the application is not limited to the precise arrangements and instrumentalities shown in the drawings, which have been described above, and that various modifications and changes may be effected without departing from the scope thereof. The scope of the application is limited only by the appended claims.

Claims

1. A cell parameter processing method, comprising:

receiving an optimization instruction sent by a terminal device, and converting the optimization instruction into an instruction characteristic by adopting a natural language understanding model;

determining a target optimization strategy according to the instruction characteristics;

acquiring data of each cell, wherein the cell data comprises cell type, cell parameters and operation data;

determining policy details corresponding to each cell according to the target optimization policy and the cell type of each cell;

determining at least one cell to be optimized according to the policy details and the operation data of each cell;

and adjusting the cell parameters of each cell to be optimized according to the policy details of the cell to be optimized.

2. The method of claim 1, wherein the policy details include numerical thresholds and weights corresponding to each type of operation data;

correspondingly, the determining at least one cell to be optimized according to the policy details and the operation data of each cell comprises the following steps:

Multiplying various operation data of a target cell by corresponding weights to obtain various weighted values, wherein the target cell is any cell;

adding various weighted values of the target cell to obtain a numerical sum;

and if the numerical sum is larger than the numerical threshold, determining the target cell as a cell to be optimized.

3. The method of claim 1, wherein the policy details include optimization actions and decision criteria;

correspondingly, optimizing the cell parameters of each cell to be optimized according to the policy details and the cell parameters of the cell to be optimized, including:

the optimization action is applied to adjust the cell parameters of each cell to be optimized, and the parameters to be determined of each cell to be optimized are obtained;

controlling each cell to be optimized to run according to the parameters to be determined;

acquiring new operation data of each cell to be optimized;

if the new operation data of the target cell to be adjusted does not meet the preset operation requirement, controlling the target cell to be adjusted to operate with the corresponding cell parameters, wherein the target cell to be adjusted is any cell to be adjusted;

if the new operation data of the target cell to be adjusted meets the preset operation requirement and meets the judgment standard, controlling the target cell to be adjusted to operate with the parameters to be determined;

And if the new operation data of the target cell to be adjusted meets the preset operation requirement and does not meet the judgment standard, the step of adjusting the cell parameters of the cell to be optimized to judge whether the new operation data meets the preset operation requirement is re-executed until the new operation data meets the preset operation requirement.

4. A method according to claim 3, further comprising, after said step of re-executing the step of adjusting the cell parameters of the cell to be optimized to determine whether the new operation data meets the preset operation requirement:

and if the step of adjusting the cell parameters of the cell to be optimized is re-executed to judge whether the new operation data accords with the preset operation requirement for m times, and then the new operation data does not accord with the preset operation requirement, controlling the target cell to be adjusted to operate with the corresponding cell parameters, wherein m is a positive integer.

5. The method of any one of claims 1 to 4, wherein said determining a target optimization strategy based on said instruction characteristics comprises:

inputting the instruction features into a feature conversion model obtained through pre-training to obtain a target optimization strategy.

6. The method according to any one of claims 1 to 4, wherein determining policy details corresponding to each cell according to the target optimization policy and a cell type of each cell comprises:

And inputting the target optimization strategy and the cell type of the target cell into a preset strategy detail generation model to obtain the strategy details of the target cell, wherein the target cell is any cell.

7. The method according to any one of claims 1 to 4, further comprising, prior to said acquiring each cell data:

acquiring operation data of each cell;

normalizing the operation data of each cell to obtain normalized operation data of each cell;

and clustering the cells by adopting standardized operation data of the cells to obtain cell types corresponding to the cells.

8. A cell parameter processing apparatus, comprising:

the instruction conversion module is used for receiving an optimization instruction sent by the terminal equipment and converting the optimization instruction into an instruction characteristic by adopting a natural language understanding model;

the strategy determining module is used for determining a target optimization strategy according to the instruction characteristics;

the data acquisition module is used for acquiring the data of each cell, wherein the cell data comprises cell types, cell parameters and operation data;

the detail determining module is used for determining policy details corresponding to each cell according to the target optimization policy and the cell type of each cell;

The cell determining module is used for determining at least one cell to be optimized according to the policy details and the operation data of each cell;

and the cell optimization module is used for adjusting the cell parameters of each cell to be optimized according to the policy details of the cell to be optimized.

9. An electronic device, comprising: a processor, and a memory communicatively coupled to the processor;

the memory stores computer-executable instructions;

the processor executing computer-executable instructions stored in the memory, causing the processor to perform the cell parameter processing method according to any one of claims 1 to 7.

10. A computer readable storage medium, wherein computer executable instructions are stored in the computer readable storage medium, which when executed by a processor is configured to implement the cell parameter processing method according to any one of claims 1 to 7.