CN113286315B - Load balance judging method, device, equipment and storage medium - Google Patents

Abstract

The application provides a load balance judging method, a device, equipment and a storage medium, wherein the method obtains current network perception data of all users in a cell to be judged, wherein the current network perception data comprises current voice service data and current data service data; inputting current network perception data into a preset network perception discrimination model, and obtaining perception weights of all users in a preset network area according to the output of the preset network perception discrimination model; determining the number of the perception difference users in the cell to be judged according to a preset network perception threshold and perception weight; according to the number of the users with poor perception and the total number of the users of the cell to be judged, whether the cell to be judged needs to be subjected to load balancing or not is determined, the load condition of the cell can be judged and balanced more accurately, the condition of unbalanced load is effectively solved, network resource distribution is more reasonable, and network perception of users is improved.

Description

Load balancing judgment method, device, equipment and storage medium

technical field

The present invention relates to the field of communication technologies, and in particular, to a load balance judgment method, apparatus, device and storage medium.

Background technique

With the rapid development of wireless communication technology, the number of users of the 4th generation mobile communication technology (4G) and the 5th generation mobile communication technology (5G) is also increasing rapidly. It is also getting higher and higher, and load balancing is required to meet the network requirements of each user.

The existing load balancing judging method is to judge according to the network indicators on the base station side. If the total network indicators collected by the base station side reach the preset load balancing standard, load balancing operations are performed on users under the base station.

However, the existing load balancing method has a single method for judging whether the load is balanced, and cannot effectively solve the situation of unbalanced load, resulting in unreasonable allocation of network resources in the area and poor network perception of users.

SUMMARY OF THE INVENTION

The present application provides a load balancing judgment method, device, equipment and storage medium, so as to solve the problem that the existing load balancing method has a single method for judging whether the load is balanced, and cannot effectively solve the situation of unbalanced load, resulting in unreasonable distribution of network resources in the area. , The technical problem of poor user network perception.

In a first aspect, an embodiment of the present application provides a load balancing judgment method, including:

Acquiring current network perception data of all users in the cell to be determined, wherein the current network perception data includes current voice service data and current data service data;

Inputting the current network perception data into a preset network perception discrimination model, and obtaining the perception weights of all users in the preset network area according to the output of the preset network perception discrimination model;

Determine the number of users with poor perception in the to-be-determined cell according to the preset network perception threshold and the perception weight;

According to the number of users with poor perception and the total number of users of the cell to be determined, it is determined whether the cell to be determined needs to perform load balancing.

Here, before load balancing is performed in this embodiment of the present application, the current network perception data in the cell, including the current voice service data and the current data service data, is first obtained, and the perception weight is predicted for the users of the current cell according to a preset network perception discrimination model , and then determine the number of users with poor perception in the cell according to the preset network perception threshold and perception weight, and judge the load situation in the cell according to the number of users with poor perception and the total number of users in the cell. The embodiment of the present application combines voice services and data services. In two aspects, considering the real usage of users in the network interaction process, and combining the bearing capacity of voice services and the bearing capacity of data services to predict the load situation, the bearing situation of cell services can be effectively judged. At the same time, the embodiments of the present application Combined with the total number of users in the cell, the load situation can be determined according to the proportion, which can more accurately judge and balance the load situation of the cell, effectively solve the situation of unbalanced load, make the network resource allocation more reasonable, and improve the user's network perception.

Optionally, before the inputting the current network perception data into the preset network perception discrimination model, the method further includes:

obtaining historical network perception data of users in the cell to be judged and historical perception weights corresponding to the historical network perception data, wherein the historical network perception data includes historical voice service data and historical data service data;

Model training is performed according to the historical network perception data and the historical network perception weights corresponding to the historical network perception data to obtain a preset network perception discrimination model.

Here, in this embodiment of the present application, before performing the perception weight prediction of the current network perception data, a preset network perception discrimination model is first established to perform accurate and convenient perception weight prediction. The historical network perception weight corresponding to the data can be trained by wandering, and an accurate discriminant model can be obtained. In addition, the historical perception data here includes historical voice service data and historical data service data. Use requirements, determine the user's network perception in the existing network, avoid the existing network to judge based on the network indicators on the base station side, may perform load balancing operations when the user's network perception is good, and the user's network perception is not good. The phenomenon that the load balancing operation will not be performed at times will further improve the accuracy of the judgment of whether the load is balanced, make the network resource allocation more reasonable, and improve the user's network perception.

Optionally, performing model training according to the historical network perception data and historical perception weights corresponding to the historical network perception data, including:

performing data preprocessing on the historical network perception data and the perception weights corresponding to the historical network perception data to obtain the processed historical network perception data and the processed historical perception weights;

Taking the processed historical network perception data as an input, and using the processed historical perception weight as an output, model training is performed to obtain the preset network perception discrimination model.

Here, the embodiment of the present application also preprocesses the above data before performing model training according to the historical network perception data and the perception weights corresponding to the historical network perception data. This preprocessing may be data cleaning to obtain data in a uniform format. Improve the accuracy and convenience of feature training. This preprocessing can also be feature engineering training to obtain more dimensional feature data to improve the accuracy of feature training. It can also be other types of data preprocessing or different The combination of preprocessing methods and data preprocessing can improve the training speed of the model and the accuracy of model weights, and further improve the efficiency of load balancing judgment.

Optionally, the determining the number of users with poor perception in the to-be-determined cell according to a preset network perception threshold and the perception weight includes:

According to the preset network perception threshold and the perception weight, the current network perception interval of all users is determined, wherein the current network perception interval includes a perception difference interval;

It is determined that the number of users whose perception weight is within the range of the perception difference interval is the number of users with perception difference in the cell to be determined.

Here, when judging users with poor perception in the cell in this embodiment of the present application, different network perception intervals can be determined by preset network perception thresholds, where the network perception intervals include perception difference intervals, and the preset network perception threshold here can be It is determined according to the actual situation, and no specific restrictions are made here. If the user's perception weight is within the perception difference interval, then the user can be determined to be a perceptually poor user, and by judging according to the interval, the perceptually poor user can be determined effectively and accurately. The situation is adjusted to suit the perceived needs of different environments.

Optionally, determining whether the to-be-determined cell needs to perform load balancing according to the number of perceived poor users and the total number of users of the to-be-determined cell includes:

According to the number of users with poor perception and the total number of users in the cell to be determined, determine the proportion of users with poor perception in the cell to be determined;

determining the load state of the to-be-determined cell according to the percentage of users with poor perception;

According to the load status, it is determined whether the to-be-determined cell needs to perform load balancing.

Here, the embodiment of the present application determines whether the cell needs to perform load balancing according to the proportion of the number of users with poor perception. Considering the overall situation of the cell, the load balancing is more reasonable and effective.

Optionally, the determining the load state of the to-be-determined cell according to the percentage of users with poor perception includes:

If the proportion of users with poor perception is greater than the first preset proportion of users, it is determined that the load state of the to-be-determined cell is high load;

If the perceptually poor user ratio is less than or equal to a first preset user ratio and greater than or equal to a second preset user ratio, determining that the load status of the to-be-determined cell is normal load;

If the proportion of users with poor perception is less than the second preset proportion of users, it is determined that the load state of the to-be-determined cell is low load.

Here, the embodiment of the present application determines the load condition of the cell by combining different percentages of users with poor perception. Different load balancing methods can be performed on the cell according to different load conditions, and the load balancing method is more flexible, accurate and effective.

In a second aspect, an embodiment of the present application provides a load balancing judging device, including:

an acquisition module to acquire current network perception data of all users in the cell to be determined, wherein the current network perception data includes current voice service data and current data service data;

an input module, configured to input the current network perception data into a preset network perception discrimination model, and obtain the perception weights of all users in the preset network area according to the output of the preset network perception discrimination model;

a first determining module, configured to determine the number of users with poor perception in the to-be-determined cell according to a preset network perception threshold and the perception weight;

The second determining module is configured to determine whether the cell to be determined needs to perform load balancing according to the number of users with poor perception and the total number of users of the cell to be determined.

Optionally, before the input module inputs the current network perception data into a preset network perception discrimination model, the above-mentioned device further includes:

A training module, configured to acquire historical network perception data of users in the cell to be determined and historical perception weights corresponding to the historical network perception data, wherein the historical network perception data includes historical voice service data and historical data service data; Model training is performed according to the historical network perception data and the historical network perception weights corresponding to the historical network perception data to obtain a preset network perception discrimination model.

Optionally, the training module is specifically used for:

Perform data preprocessing on the historical network perception data and the perception weights corresponding to the historical network perception data to obtain processed historical network perception data and processed historical network perception weights; take the processed historical network perception data as Input, take the processed historical perception weight as output, perform model training, and obtain the preset network perception discrimination model.

Optionally, the first determining module is specifically configured to:

According to the preset network perception threshold and the perception weight, the current network perception interval of all users is determined, wherein the current network perception interval includes a perception difference interval;

It is determined that the number of users whose perception weight is within the range of the perception difference interval is the number of users with perception difference in the cell to be determined.

Optionally, the second determining module is specifically configured to:

According to the number of users with poor perception and the total number of users in the cell to be determined, determine the proportion of users with poor perception in the cell to be determined;

determining the load state of the to-be-determined cell according to the percentage of users with poor perception;

According to the load status, it is determined whether the to-be-determined cell needs to perform load balancing.

Optionally, the second determining module is specifically configured to:

If the proportion of users with poor perception is greater than the first preset proportion of users, it is determined that the load state of the to-be-determined cell is high load;

If the perceptually poor user ratio is less than or equal to a first preset user ratio and greater than or equal to a second preset user ratio, determining that the load status of the to-be-determined cell is normal load;

If the proportion of users with poor perception is less than the second preset proportion of users, it is determined that the load state of the to-be-determined cell is low load.

In a third aspect, an embodiment of the present application provides a load balancing judgment device, including: at least one processor and a memory;

the memory stores computer-executable instructions;

The at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor executes the load balancing determination method described in the first aspect and various possible designs of the first aspect.

In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, where computer-executable instructions are stored in the computer-readable storage medium, and when a processor executes the computer-executable instructions, the first aspect and the first The load balancing judgment method described in the aspect of various possible designs.

In a fifth aspect, an embodiment of the present invention provides a computer program product, including a computer program that, when executed by a processor, implements the load balancing determination method described in the first aspect and various possible designs of the first aspect .

The load balancing judgment method, device, device, and storage medium provided by the embodiments of the present application, wherein, before performing load balancing, the method first obtains current network perception data in a cell, including current voice service data and current data service data, according to the pre- Suppose the network perception discrimination model predicts the perception weight of users in the current cell, and then determines the number of users with poor perception in the cell according to the preset network perception threshold and perception weight, and determines the number of users with poor perception in the cell according to the number of users with poor perception and the total number of users in the cell. In the embodiment of the present application, the voice service and the data service are combined, and the real usage of the user in the network interaction process is considered, and the load condition is predicted based on the bearing capacity of the voice service and the bearing capacity of the data service. It can effectively judge the load situation of the cell service. At the same time, the embodiment of the present application can determine the load situation according to the proportion in combination with the total number of users in the cell, which can more accurately judge and balance the load situation of the cell, and effectively solve the problem of unbalanced load. This makes the network resource allocation more reasonable and improves the user's network awareness.

Description of drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only These are some embodiments of the present application, and for those of ordinary skill in the art, other drawings can also be obtained from these drawings without any creative effort.

FIG. 1 is a schematic diagram of the architecture of a load balancing system provided by an embodiment of the present application;

FIG. 2 is a schematic flowchart of a load balancing judgment method provided by an embodiment of the present application;

FIG. 3 is a schematic flowchart of another load balancing judgment method provided by an embodiment of the present application;

FIG. 4 is a schematic structural diagram of a load balancing judging device provided by an embodiment of the present application;

FIG. 5 is a schematic structural diagram of a load balancing judgment device provided by an embodiment of the present application.

The above-mentioned drawings have shown clear embodiments of the present disclosure, and will be described in more detail hereinafter. These drawings and written descriptions are not intended to limit the scope of the disclosed concepts in any way, but rather to illustrate the disclosed concepts to those skilled in the art by referring to specific embodiments.

Detailed ways

Exemplary embodiments will be described in detail herein, examples of which are illustrated in the accompanying drawings. Where the following description refers to the drawings, the same numerals in different drawings refer to the same or similar elements unless otherwise indicated. The implementations described in the illustrative examples below are not intended to represent all implementations consistent with this disclosure. Rather, they are merely examples of apparatus and methods consistent with some aspects of the present disclosure as recited in the appended claims.

The terms "first", "second", "third" and "fourth", etc. (if any) in the description and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It is to be understood that data so used may be interchanged under appropriate circumstances so that the embodiments of the application described herein can be practiced in sequences other than those illustrated or described herein. Furthermore, the terms "comprising" and "having" and any variations thereof, are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those expressly listed Rather, those steps or units may include other steps or units not expressly listed or inherent to these processes, methods, products or devices.

The complexity of the network structure and the diversification of services make the improvement of user perception more and more complicated. In order to meet the daily network requirements of users, improve the network perception of users, and provide users with a better user experience, the base station needs to perform network load balancing.

The existing load balancing judgment method is based on the network index on the base station side. If the index reaches the specified standard of load balancing, the corresponding load balancing operation will be performed even if the user's perception of the base station is normal. The technical method is single, does not take into account the real usage of the user, and will increase the complexity of the network. On the contrary, when users use terminals to connect to the network, they generally interact with data services and voice services. The difference is that the base station cells have different bearing capacities for data services and voice services. The same balancing standard for groups with different emphasis on services is not only The current situation of unbalanced load cannot be effectively solved, and it will also lead to unreasonable allocation of network resources in the area, which affects the network perception of users.

In order to solve the above problem, the embodiments of the present application provide a load balancing judgment method, device, equipment, and storage medium, wherein the method performs load balancing judgment and operation in the dimension of user perception, according to the user's different Internet services and voice services. According to the user's perception, it is judged whether the base station cell is in a load imbalance state, and the unbalanced base station cell can be balanced.

Optionally, FIG. 1 is a schematic diagram of an architecture of a load balancing system provided by an embodiment of the present application. In FIG. 1 , the above architecture includes a base station 101 , a cell 1010 , a cell 1011 , a cell 1012 , a user equipment 10100 , a user equipment 10101 , and a user equipment 10102 .

The above system architecture is only schematic, a base station includes 3 sectors, each sector includes cells, and the number of cells is determined according to the situation. Here, 3 cells are used for illustration. Both contain multiple user devices.

Specifically, in the implementation process, the load balancing for the base station 101 may be implemented through a server connected to the base station 101 .

It can be understood that the structures illustrated in the embodiments of the present application do not constitute a specific limitation on the architecture of the load balancing system. In some other feasible embodiments of the present application, the above architecture may include more or less components than shown in the figure, or combine some components, or separate some components, or arrange different components, depending on the actual application. The scene is determined, and there is no restriction here. The components shown in Figure 1 may be implemented in hardware, software, or a combination of software and hardware.

In addition, the network architecture and service scenarios described in the embodiments of the present application are for the purpose of illustrating the technical solutions of the embodiments of the present application more clearly, and do not constitute limitations on the technical solutions provided by the embodiments of the present application. With the evolution of the network architecture and the emergence of new service scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

The technical solutions of the present application are described in detail below in conjunction with specific embodiments:

Optionally, FIG. 2 is a schematic flowchart of a load balance judgment method provided by an embodiment of the present application. The execution subject of the embodiment of the present application may be a server, and the specific execution subject may be determined according to an actual application scenario. As shown in Figure 2, the method includes the following steps:

S201: Acquire current network perception data of all users in the cell to be determined.

The current network perception data includes current voice service data and current data service data.

Optionally, user information may be determined according to user terminals under the base station within a certain period, and the network perception data may also include basic network indicator data, including basic network indicators, data service indicators and voice services of each user terminal under the base station. index.

S202: Input the current network perception data into the preset network perception discrimination model, and obtain the perception weights of all users in the preset network area according to the output of the preset network perception discrimination model.

Optionally, the preset network-aware discriminant model can be established for a variety of classification algorithms, such as a decision tree (Decision Tree, DT) classification algorithm, a naive Bayes (Navie Bayes, NB) classification algorithm, and a support vector machine (Support Vector Machine). Machine, SVM) classification algorithm, neural network (Neural Network, NN) classification algorithm, XGBOOST classification algorithm, etc., which are not specifically limited in this embodiment of the present application.

S203: Determine the number of users with poor perception in the to-be-determined cell according to a preset network perception threshold and a perception weight.

Optionally, determining the number of users with poor perception in the cell to be determined according to the preset network perception threshold and the perception weight includes: determining the current network perception interval of all users according to the preset network perception threshold and perception weight, wherein the current network perception interval Including the perceptual difference interval; determining the number of users whose perception weight is within the range of the perceptual difference interval is the number of perceptually poor users in the cell to be judged.

The preset network perception threshold can be set according to requirements, and further, the user network perception threshold can be set according to the requirements of users in the cell coverage area for data services and voice services. Exemplarily, if the users in the area do not use the network frequently, for example, the area is a residential area, a park, etc., the user's network perception threshold can be set to 0.95; if the users in the area use the network more frequently, for example If the area is a dense area of calls or Internet access, the user network perception threshold can be lowered from 0.95 to 0.85. The larger the threshold, the higher the user's demand for the network, and the easier it is to have poor user network perception.

Specifically, according to the preset network perception threshold and the user network perception weight, the network perception interval corresponding to the network perception weight is determined; the network perception interval may include: poor perception, poor perception, average perception, better perception, excellent perception, and user network perception The interval is used to represent the mapping of the user network perception weight to the perception interval. Exemplarily, the user's network perception threshold can be mapped according to the network perception interval. For example, for the residential areas, parks and other areas where the network is not frequently used in the above example, the perception threshold can be [0, 0.2), [0.2, 0.4), [0.4, 0.6), [0.6, 0.8), and [0.8, 1) users are mapped to the [perceptually superior, better, moderate, poorer, and poorer] perception intervals, respectively.

Here, when judging users with poor perception in the cell in this embodiment of the present application, different network perception intervals can be determined by preset network perception thresholds, where the network perception intervals include perception difference intervals, and the preset network perception threshold here can be It is determined according to the actual situation, and no specific restrictions are made here. If the user's perception weight is within the perception difference interval, then the user can be determined to be a perceptually poor user, and by judging according to the interval, the perceptually poor user can be determined effectively and accurately. The situation is adjusted to suit the perceived needs of different environments.

S204: Determine whether the cell to be judged needs to perform load balancing according to the number of perceived poor users and the total number of users of the cell to be judged.

Optionally, determining whether the to-be-determined cell needs to perform load balancing according to the number of perceived poor users and the total number of users of the to-be-determined cell includes: determining the perception of the to-be-determined cell according to the number of perceived poor users and the total number of users of the to-be-determined cell. The ratio of poor users; according to the ratio of perceived poor users, the load status of the cell to be judged is determined; according to the load status, it is determined whether the cell to be judged needs to perform load balancing.

Specifically, based on the cell used by the user as a benchmark, the total number of users and the number of users with poor perception in the cell are aggregated, and the proportion of users with poor perception in the cell is calculated according to the following formula. Further, the corresponding cells under the same sector and under the same coverage of the base station are target cells that can be used for load balancing operations respectively. Calculated as follows:

Here, the embodiment of the present application determines whether the cell needs to perform load balancing according to the proportion of the number of users with poor perception. Considering the overall situation of the cell, the load balancing is more reasonable and effective.

Optionally, determining the load status of the cell to be judged according to the proportion of users with poor perception, including:

If the proportion of users with poor perception is greater than the first preset proportion of users, it is determined that the load status of the cell to be determined is high load;

If the ratio of perceived poor users is less than or equal to the first preset user ratio and greater than or equal to the second preset user ratio, it is determined that the load status of the cell to be judged is normal load;

If the ratio of perceived poor users is less than the second preset ratio of users, it is determined that the load status of the cell to be determined is low load.

In this step, the preset cell network sensing threshold may be set according to requirements, and further, the cell network sensing threshold may be set according to the scene attribute of the area where the cell is located. Exemplarily, if users in the area are not sensitive to network usage, the cell network perception threshold can be set to a larger value; if users in the area are more sensitive to network usage, the cell network perception threshold can be set to smaller value. The smaller the threshold, the more sensitive the users in the area are to network perception, and the greater the impact of the proportion of users with poor perception on the judgment of load balancing. Specifically, the cell load status is determined according to the preset cell network perception threshold and the ratio of users with poor cell perception; the cell load status includes: low load, normal load, and high load, and the cell load status is used to represent the ratio of users with poor cell perception to the load status mapping. Exemplarily, the proportion of users with poor cell perception can be mapped according to the cell load state. [0.05, 1) users are respectively mapped to the [low load, normal load, high load] load state interval; for the user network perception insensitive area in the above example, the maximum value of the low load interval and the normal load interval can be appropriately increased .

Specifically, according to the cells with unbalanced load, users with high weights are balanced to other cells with the same coverage under the same sector through the user network perception weight. Exemplarily, for cells with unbalanced loads, users with high network perception weights will be preferentially screened. The users have high network demands, and such users will be preferentially balanced to other cells with the same coverage under the same sector of the base station.

Here, the embodiment of the present application determines the load condition of the cell by combining different percentages of users with poor perception. Different load balancing methods can be performed on the cell according to different load conditions, and the load balancing method is more flexible, accurate and effective.

Before performing load balancing in this embodiment of the present application, first acquire current network perception data in the cell, including current voice service data and current data service data, and predict the perception weight of users in the current cell according to a preset network perception discrimination model, and then The number of users with poor perception in the cell is determined according to the preset network perception threshold and the perception weight, and the load situation in the cell is judged according to the number of users with poor perception and the total number of users in the cell. The embodiment of the present application combines two aspects of voice service and data service , taking into account the real usage of users in the network interaction process, and combining the bearing capacity of voice services and the bearing capacity of data services to predict the load situation, it is possible to effectively judge the bearing situation of cell services. The total number of users can be determined in proportion to the load situation, which can more accurately judge and balance the cell load situation, effectively solve the unbalanced load situation, make network resource allocation more reasonable, and improve user network perception.

Optionally, in the embodiment of the present application, a preset network perception discrimination model may also be established in advance to perform accurate and convenient perception weight prediction. Correspondingly, FIG. 3 is a flowchart of another load balancing determination method provided by the embodiment of the present application. The schematic diagram, as shown in Figure 3, the method includes:

S301: Acquire current network perception data of all users in the cell to be determined.

The current network perception data includes current voice service data and current data service data.

S302: Obtain historical network perception data of users in the cell to be determined and historical perception weights corresponding to the historical network perception data.

The historical network perception data includes historical voice service data and historical data service data.

S303: Perform model training according to the historical network perception data and the historical network perception weights corresponding to the historical network perception data to obtain a preset network perception discrimination model.

Optionally, model training is performed according to historical network perception data and historical network perception weights corresponding to the historical network perception data, including:

Perform data preprocessing on the historical network perception data and the perception weights corresponding to the historical network perception data to obtain the processed historical network perception data and the processed historical perception weights;

Taking the processed historical network perception data as input, and using the processed historical perception weight as output, model training is performed to obtain a preset network perception discrimination model.

Here, the embodiment of the present application also preprocesses the above data before performing model training according to the historical network perception data and the perception weights corresponding to the historical network perception data. This preprocessing may be data cleaning to obtain data in a uniform format. Improve the accuracy and convenience of feature training. This preprocessing can also be feature engineering training to obtain more dimensional feature data to improve the accuracy of feature training. It can also be other types of data preprocessing or different The combination of preprocessing methods and data preprocessing can improve the training speed of the model and the accuracy of model weights, and further improve the efficiency of load balancing judgment.

The method for establishing a classification model provided by this embodiment may include the following steps:

Obtain the historical network perception data of users in the area, and establish a user network interaction database; perform preprocessing according to basic network indicators, data service indicators and voice service indicator data to obtain data in a regular format; according to the data service characteristics and voice services in a regular format Perform feature engineering on features to obtain more dimensional feature data; use the processed derivative data as training sample features, use the existing network perception standards of data services and voice services as training sample labels, and build a user perception discrimination model based on a classification algorithm.

Optionally, the preprocessing of the basic network indicators, data service indicators and voice service indicator data includes: cleaning the data to obtain data in a uniform format.

Specifically, data cleaning can be achieved by field filtering, filtering duplicate values, filling null values, and detecting outliers.

Optionally, feature engineering may be performed according to data service features and voice service features in a regular format to obtain feature data of more dimensions.

In this step, the feature engineering of the data service features and the voice service features with regular formats includes: combining and deriving features on the data to obtain feature data with more dimensions.

Specifically, feature engineering can be implemented by combining features in pairs, binning features, discretizing features, and derivation of features in the time dimension.

Optionally, the preset network-aware discrimination model may be a classification model, and the model is obtained by collecting user samples in the network interaction database and training according to the user samples. The user samples include basic network indicators of a large number of users in historical base station cells, The data service uses data and the voice service uses data. Based on the obtained user samples as training sample characteristic data, the obtained perception discrimination model is established with the existing network perception standards of user data services and voice services as the training sample labels.

In a possible embodiment, XGBOOST is selected as the algorithm model for data training. During the training process, only one tree is trained at a time, that is, a weak classifier, and the final prediction result is the sum of all trees. And try to reduce the residual error in each round of training the weak classifier, so that the answer is closer to the real answer. In the process of reducing the residual error, the Taylor simplification of the objective function is used to derive the first-order and second-order derivatives, and the original traversal of the sample is transformed into the traversal of the leaf nodes to transform the objective function and solve it. In the process, a greedy strategy is used to build a tree from the root node layer by layer. When the amount of data is large, an approximation algorithm is used to select candidate split points for optimization.

According to the processed derivative data as the training sample features, and the existing network perception standards of data services and voice services as the training sample labels, multiple regression trees are established through the XGBOOST algorithm. It should be noted that the training sample label can be the user's perception state: good or bad perception, corresponding to the classification result 1 or 0 in the model. Furthermore, in the model training process, the sum of the prediction results of multiple regression trees indicates that the sample belongs to the probability value of the perceptually good standard, and in the model prediction process of the new sample, the user's perceptual weight in the current network is the probability value.

S304: Input the current network perception data into the preset network perception discrimination model, and obtain the perception weights of all users in the preset network area according to the output of the preset network perception discrimination model.

S305: Determine the number of users with poor perception in the to-be-determined cell according to a preset network perception threshold and a perception weight.

S306: Determine whether the cell to be judged needs to perform load balancing according to the number of perceived poor users and the total number of users of the cell to be judged.

Before performing the sensing weight prediction of the current network sensing data in this embodiment of the present application, a preset network sensing discriminant model is first established to perform accurate and convenient sensing weight prediction. In addition, the historical perception data here includes historical voice service data and historical data service data, which can be used according to the different needs of users in Internet data services and voice services. , to judge the user's network perception in the existing network, avoiding the existing network to judge based on the network indicators on the base station side, load balancing operations may be performed when the user's network perception is good, and reverse when the user's network perception is not good. The occurrence of no load balancing operation further improves the accuracy of judging whether the load is balanced, makes network resource allocation more reasonable, and improves user network perception.

FIG. 4 is a schematic structural diagram of a load balancing judging apparatus provided by an embodiment of the present application. As shown in FIG. 4 , the apparatus in this embodiment of the present application includes: an acquisition module 401 , an input module 402 , a first determination module 403 and a second determination module Module 404. The load balancing judging device here may be the above-mentioned server itself, or a chip or an integrated circuit that implements the function of the server. It should be noted here that the division of the acquisition module 401 , the input module 402 , the first determination module 403 and the second determination module 404 is only a division of logical functions, and the two may be physically integrated or independent.

Wherein, the acquiring module acquires current network perception data of all users in the cell to be determined, wherein the current network perception data includes current voice service data and current data service data;

The input module is used to input the current network perception data into the preset network perception discrimination model, and obtain the perception weights of all users in the preset network area according to the output of the preset network perception discrimination model;

a first determining module, configured to determine the number of users with poor perception in the cell to be determined according to a preset network perception threshold and a perception weight;

The second determining module is configured to determine whether the cell to be determined needs to perform load balancing according to the number of perceived poor users and the total number of users in the cell to be determined.

Optionally, before the input module inputs the current network perception data into the preset network perception discrimination model, the above device further includes:

The training module is used to obtain the historical network perception data of users in the cell to be judged and the historical perception weight corresponding to the historical network perception data, wherein the historical network perception data includes historical voice service data and historical data service data; according to the historical network perception data and Model training is performed on the historical perception weights corresponding to the historical network perception data to obtain a preset network perception discrimination model.

Optionally, the training module is specifically used to:

Perform data preprocessing on the historical network perception data and the perception weights corresponding to the historical network perception data to obtain the processed historical network perception data and the processed historical perception weights;

Taking the processed historical network perception data as input, and using the processed historical perception weight as output, model training is performed to obtain a preset network perception discrimination model.

Optionally, the first determining module is specifically used for:

Determine the current network perception interval of all users according to the preset network perception threshold and perception weight, wherein the current network perception interval includes the perception difference interval;

It is determined that the number of users whose perception weight is within the range of the perception difference interval is the number of perception difference users in the cell to be judged.

Optionally, the second determining module is specifically used for:

According to the number of users with poor perception and the total number of users in the cell to be judged, determine the proportion of users with poor perception in the cell to be judged;

Determine the load status of the cell to be judged according to the proportion of users with poor perception;

According to the load status, it is determined whether the cell to be judged needs to perform load balancing.

Optionally, the second determining module is specifically used for:

If the proportion of users with poor perception is greater than the first preset proportion of users, it is determined that the load status of the cell to be determined is high load;

If the ratio of perceived poor users is less than or equal to the first preset user ratio and greater than or equal to the second preset user ratio, it is determined that the load status of the cell to be judged is normal load;

If the ratio of perceived poor users is less than the second preset ratio of users, it is determined that the load status of the cell to be determined is low load.

FIG. 5 is a schematic structural diagram of a load balancing judging device provided by an embodiment of the present application, and the load balancing judging device may be a server. The components shown herein, their connections and relationships, and their functions are by way of example only, and do not limit implementations of the application described and/or claimed herein.

As shown in FIG. 5 , the load balancing judging device includes: a processor 501 and a memory 502, each of which is connected to each other using different buses, and can be installed on a common motherboard or installed in other ways as required. Processor 501 may process instructions executed within the load balancing determination device, including instructions stored in or on memory for displaying graphical information on external input/output devices, such as a display device coupled to the interface. In other embodiments, multiple processors and/or multiple buses may be used with multiple memories and multiple memories, if desired. A processor 501 is taken as an example in FIG. 5 .

As a non-transitory computer-readable storage medium, the memory 502 can be used to store non-transitory software programs, non-transitory computer-executable programs and modules, such as program instructions/modules ( For example, as shown in FIG. 4, an acquisition module 401, an input module 402, a first determination module 403 and a second determination module 404). The processor 501 executes various functional applications and data processing of the authentication platform by running the non-transitory software programs, instructions and modules stored in the memory 502, that is, the method for implementing the load balancing judging device in the above method embodiments.

The load balancing judging device may further include: an input device 503 and an output device 504 . The processor 501 , the memory 502 , the input device 503 and the output device 504 may be connected by a bus or in other ways, and the connection by a bus is taken as an example in FIG. 5 .

The input device 503 can receive input numerical or character information, and generate key signal input related to user settings and function control of the load balancing judging device, such as a touch screen, a keypad, a mouse, or multiple mouse buttons, a trackball, a joystick and other input devices. The output device 504 may be an output device such as a display device of a load balance determination device. The display device may include, but is not limited to, a liquid crystal display (LCD), a light emitting diode (LED) display, and a plasma display. In some implementations, the display device may be a touch screen.

The load balancing judging device of the embodiment of the present application can be used to execute the technical solutions in the above method embodiments of the present application, and the implementation principle and technical effect thereof are similar, and are not repeated here.

Embodiments of the present application further provide a computer-readable storage medium, where computer-executable instructions are stored in the computer-readable storage medium, and when the computer-executable instructions are executed by a processor, are used to implement any of the load balancing determination methods described above.

Embodiments of the present application further provide a computer program product, including a computer program, which, when executed by a processor, is used to implement any of the load balancing determination methods described above.

In the several embodiments provided in this application, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the apparatus embodiments described above are only illustrative. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods, for example, multiple units or components may be combined or integrated. to another system, or some features can be ignored, or not implemented. On the other hand, the shown or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, indirect coupling or communication connection of devices or units, and may be in electrical, mechanical or other forms.

In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit. The above-mentioned integrated units may be implemented in the form of hardware, or may be implemented in the form of software functional units.

Other embodiments of the present disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or techniques in the technical field not disclosed by the present disclosure . The specification and examples are to be regarded as exemplary only, with the true scope and spirit of the disclosure being indicated by the following claims.

It is to be understood that the present disclosure is not limited to the precise structures described above and illustrated in the accompanying drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims (8)

1. A method for judging load balance is characterized by comprising the following steps:

acquiring current network sensing data of all users in a cell to be judged, wherein the current network sensing data comprises current voice service data and current data service data;

inputting the current network perception data into a preset network perception discrimination model, and obtaining perception weights of all users in the preset network area according to the output of the preset network perception discrimination model;

determining the number of the perception difference users in the cell to be judged according to a preset network perception threshold value and the perception weight;

determining whether the cell to be judged needs to carry out load balancing according to the number of the sensing difference users and the total number of the users of the cell to be judged;

determining whether the cell to be judged needs to perform load balancing according to the number of the perception difference users and the total number of the users of the cell to be judged, including:

determining the perception difference user proportion of the cell to be judged according to the perception difference user number and the total user number of the cell to be judged;

determining the load state of the cell to be judged according to the perception difference user proportion;

and determining whether the cell to be judged needs to carry out load balancing or not according to the load state.

2. The method of claim 1, wherein prior to said inputting the current network-aware data into a preset network-aware discriminant model, further comprising:

acquiring historical network perception data of users in the cell to be judged and historical perception weights corresponding to the historical network perception data, wherein the historical network perception data comprises historical voice service data and historical data service data;

and performing model training according to the historical network perception data and historical perception weights corresponding to the historical network perception data to obtain a preset network perception discrimination model.

3. The method according to claim 2, wherein the performing model training according to the historical network perception data and the historical perception weights corresponding to the historical network perception data comprises:

performing data preprocessing on the historical network perception data and the perception weight corresponding to the historical network perception data to obtain processed historical network perception data and processed historical perception weight;

and performing model training by taking the processed historical network perception data as input and the processed historical perception weight as output to obtain the preset network perception discrimination model.

4. The method according to any one of claims 1 to 3, wherein the determining the number of the poor-sensing users in the cell to be determined according to a preset network sensing threshold and the sensing weight includes:

determining current network perception intervals of all the users according to a preset network perception threshold and the perception weight, wherein the current network perception intervals comprise perception difference intervals;

and determining the number of the users with the perception weight in the range of the perception difference region as the number of the perception difference users in the cell to be judged.

5. The method according to claim 1, wherein the determining the load status of the cell to be determined according to the poor-perception user ratio comprises:

if the perception difference user proportion is larger than a first preset user proportion, determining that the load state of the cell to be judged is high;

if the perception difference user proportion is smaller than or equal to a first preset user proportion and larger than or equal to a second preset user proportion, determining that the load state of the cell to be judged is a common load;

and if the perception difference user proportion is smaller than a second preset user proportion, determining that the load state of the cell to be judged is low.

6. A load balancing determination device, comprising:

the system comprises an acquisition module, a judging module and a judging module, wherein the acquisition module is used for acquiring current network sensing data of all users in a cell to be judged, and the current network sensing data comprises current voice service data and current data service data;

the input module is used for inputting the current network perception data into a preset network perception discrimination model and obtaining the perception weights of all the users in the preset network area according to the output of the preset network perception discrimination model;

the first determining module is used for determining the number of the perception difference users in the cell to be judged according to a preset network perception threshold value and the perception weight;

a second determining module, configured to determine whether the cell to be determined needs to perform load balancing according to the number of the poor sensing users and the total number of users of the cell to be determined;

the second determining module is specifically configured to:

determining the perception difference user proportion of the cell to be judged according to the perception difference user number and the total user number of the cell to be judged;

determining the load state of the cell to be judged according to the perception difference user proportion;

and determining whether the cell to be judged needs to carry out load balancing or not according to the load state.

7. A load balancing determination device, comprising:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of load balancing determination as claimed in any one of claims 1 to 5.

8. A computer-readable storage medium, wherein a computer-executable instruction is stored in the computer-readable storage medium, and when the computer-executable instruction is executed by a processor, the computer-readable storage medium is configured to implement the load balancing determination method according to any one of claims 1 to 5.

Images