TWI724784B - Method for focusing on problem area of mobile user - Google Patents

Abstract

The invention discloses a method for focusing on problem area of mobile user, which retrieves signaling data of an access network component and a core network component, and then uses a KPI analysis statistical method for user communication base station level, a grid position algorithm for dynamic user communication and an environment attribute algorithm for user communication according to the signaling data to focus on a problem area of user communication, wherein, the KPI analysis statistical method for user communication base station level analyzes or counts KPI of a user. The grid position algorithm for dynamic user communication dynamically maps a position of the user to a relative position of grids according to coverage and number of users of the base station. The environment attribute algorithm for user communication uses a feature identification algorithm and an artificial intelligence method to obtain the user's indoor environment or outdoor environment.

Description

Ways to focus on the problem areas of mobile users

The present invention relates to a technology for improving the problem area of mobile users, in particular to a method of focusing on the problem area of mobile users.

The general network management KPI (Key Performance Index; key performance index) can only achieve the level of detail at the base station, that is, the telecom operator only knows that a certain KPI (key performance index) in the base station is abnormal. However, the coverage of a base station ranges from a radius of 100 meters to a radius of 500 to 600 meters. Therefore, it is difficult to know exactly which users have concentrated anomalies in which area in the coverage of the base station. Therefore, engineers often Unable to propose a suitable improvement plan.

Therefore, how to provide a novel and innovative technology focusing on the problem area of mobile users to more clearly grasp the problem area or abnormal area of user communication, or to narrow down the abnormal search problem area of user communication, has become one of those skilled in the art. A major research topic.

The present invention provides a novel and innovative method of focusing on the problem area of mobile users. For example, it can more clearly grasp the problem area or abnormal area of user communication, and it can also reduce the use of The abnormal search of user communication can also accurately analyze and quickly focus on the target area that needs improvement.

The method of focusing on the problem area of the mobile user in the present invention includes: providing or retrieving the signaling data of the access network component and the core network component; and based on the signaling data of the access network component and the core network component, according to Use or go through a KPI (key performance indicator) analysis and statistics method related to user communication base station level, a grid location algorithm related to dynamic user communication, and an environment attribute algorithm related to user communication to focus or find users Communication problem area or abnormal area, among them, the KPI analysis and statistical method of the user communication base station level analysis or statistical user KPI (key performance indicators) and the KPI is the base station level, and the grid position calculation of dynamic user communication The law system dynamically corresponds the location of the user to the relative location of different grids based on the coverage of the base station used by the user and the number of users of the base station, and the environment attribute algorithm related to user communication includes or uses sequentially Feature recognition algorithms and artificial intelligence methods are used to obtain user attributes or discrimination results in indoor or outdoor environments.

In order to make the above-mentioned features and advantages of the present invention more comprehensible, embodiments are specifically described below in conjunction with the accompanying drawings. In the following description, the additional features and advantages of the present invention will be partially explained, and these features and advantages will be partly known from the description, or can be learned by practicing the present invention. The features and advantages of the present invention are realized and achieved by means of the elements and combinations specifically pointed out in the scope of the patent application. It should be understood that both the foregoing general description and the following detailed description are only illustrative and explanatory, and are not intended to limit the scope of the present invention.

11: Access network components

12: Core network components

21: Access to the signaling data of network components and core network components

22: KPI analysis and statistics on the user communication base station level

23: Grid location algorithm for dynamic user communication

24: Algorithm of environmental attributes related to user communication

25: Appropriate network improvement solutions

311: GPS longitude data

312: GPS latitude data

32: The coverage of the base station

331, 332: Urban Form

341, 342: Number of base station users

351: Suburban area with many users

352: Urban area and many users

361: The first grid area

362: second grid area

411: KPI data for users engaged in indoor activities

412: KPI data for users engaged in outdoor activities

413: User S1/X2 interface data

42: Feature recognition algorithm

421: Communication Specificity Analysis

422: Network event specific analysis

423: User behavior specific analysis

431: User signal strength

432: User Handover Event

433: User continuous use time

434: User movement range

441: Maximum information gain analysis

442: Maximize multiple divisions of the objective function

443: Analysis of the number of inverted split layers

45: Judgment criteria for environmental attributes related to user communication

461: Indoor Environment

462: outdoor environment

Figure 1 is a schematic diagram of the capture of signaling data related to accessing network components and core network components in the method of focusing on the problem areas of mobile users of the present invention;

Figure 2 is a flowchart of input data, algorithm and output data in the method of focusing on the problem area of mobile users of the present invention;

Figure 3 is a flowchart of the grid location algorithm for dynamic user communication in the method of focusing on the problem area of mobile users of the present invention; and

Figure 4 is a flow chart of analyzing the environmental attributes of user communication in the method of focusing on the problem area of mobile users of the present invention, including feature recognition algorithms and artificial intelligence methods.

The following describes the implementation of the present invention with specific specific embodiments. Those familiar with this technology can understand the other advantages and effects of the present invention from the contents disclosed in this specification, and can also implement other different specific equivalent embodiments. Or apply.

FIG. 1 is a schematic diagram of capturing the signaling data of the network component 11 and the core network component 12 in the method of focusing on the problem area of the mobile user of the present invention. For example, the access network element 11 may be a base station such as an evolved node B (abbreviated as eNB or eNodeB), and the core network element 12 may be a Long Term Evolution (LTE) or a mobility management entity (Mobility Management Entity; Core network components such as MME) or System Architecture Evolution (SAE), but the present invention is not limited to this.

As shown in Figure 1, in the present invention, a data capture module (such as a data capture device or a data processor) can capture at least two (such as three) signaling data of the X2 interface between the network components 11, and Access the signaling data of the S1 interface between the network component 11 and the core network component 12, so The signaling data of the access network component 11 and the core network component 12 can be obtained, and the signaling data of the X2 interface or the S1 interface can be as in the relevant specifications of the 3rd Generation Partnership Project (3rd Generation Partnership Project; 3GPP), respectively Defined.

Figure 2 is a flow chart of input data, algorithm and output data in the method of focusing on the problem area of mobile users of the present invention, and refer to Figure 1 for explanation.

As shown in Figure 2, input data such as the signaling data 21 (such as the signaling data of the X2 interface and the S1 interface) of the access network component 11 and the core network component 12 shown in Figure 1 are provided or retrieved. Then, according to the signaling data 21 of the access network component 11 and the core network component 12, use it in sequence or go through a KPI analysis and statistics method 22 (KPI analysis and statistics module) related to the user communication base station level, and a related dynamic user Communication grid location algorithm 23 (grid location calculation module) and an environmental attribute algorithm 24 (environmental attribute calculation module) related to user communication to focus or find (precise) user communication problem areas or Abnormal area. Then, a feasible or (more appropriate/most appropriate) appropriate network improvement solution 25 is selected based on the environmental characteristics of different areas (user communication problem areas or abnormal areas).

KPI analysis statistical method 22 (KPI analysis and statistics module) related to the user communication base station level can calculate, analyze or count the KPI (key performance indicator) of the user (individual user), and the KPI (key performance indicator) is the base station level It can also be KPI (key performance indicator) values such as signal strength, signal quality, throughput (Throughput), eRAB (radio terminal resource) interruption rate, and moving speed. The grid location algorithm for dynamic user communication 23 (grid location calculation module) can dynamically correspond the user's location to the grid of different sizes according to the coverage of the base station used by the user and the number of users of the base station. Relative location, and the environmental attribute algorithm 24 (environmental attribute calculation module) related to user communication can include or sequentially use feature recognition algorithms and artificial intelligence methods In order to obtain the user's attributes or judgment results in an indoor environment or an outdoor environment.

In addition, the grid location algorithm 23 (grid location calculation module) related to dynamic user communication and the environment attribute algorithm 24 (environment attribute calculation module) related to user communication are shown in Figures 3 to 4 below, respectively. Further instructions in.

Figure 3 is a flowchart of the grid location algorithm 23 (grid location calculation module) for dynamic user communication in the method of focusing on the problem area of mobile users of the present invention, and refer to Figures 1 to 2 for details Description.

As shown in Figure 3, the user’s location data or input data includes the user’s GPS (Global Positioning System) longitude data 311 (such as 121.123456) and GPS latitude data 312 (such as 23.123456), and the user’s GPS longitude data The two input values of 311 and GPS latitude data 312 are obtained by calculating the signaling data of the access network element 11 and the core network element 12 shown in Figure 1 plus the mathematical model used by the base station's map data. In addition, because the two input values of GPS longitude data 311 and GPS latitude data 312 obtained by calculation often have a certain distance error from the actual user location, the grid location algorithm for dynamic user communication shown in Figure 2 is used. Grid position calculation module), dynamically corresponding to the user’s position (such as GPS position or GPS absolute position) according to the coverage of the base station (service base station or user service base station) used by the user and the number of users. The relative position of a grid of different sizes.

For example, the grid location algorithm 23 (grid location calculation module) related to dynamic user communication first compares the user’s GPS location (GPS longitude data 311 and GPS latitude data 312) with the coverage area 32 of the base station. And the base station (service base station or user service base station) is obtained from user S1/X2 interface data.

Next, the grid location algorithm 23 (grid location calculation module) for dynamic user communication uses the geographic location data of the base station (service base station) to analyze and determine the (large) city form 331 or (small) city form 332 , And then import KPI (key performance indicator) data to get or help get the number of base station users 341 or the number of base station users 342. If the user uses the mobile network in the urban area and the base station (service base station) used by the user has a large number of users (that is, the urban area and the number of users is 352), the analysis must be performed in a smaller grid area, such as output The data is the second grid area 362 of 50 meters * 50 meters. In addition, although the user uses the mobile network in the suburbs, the base station (serving base station) used by the user has a large number of users (that is, the number of users in the suburbs is 351 more). At this time, it can be used less than the base station (serving base station). ) But larger than the range of the second grid area 362 (such as 50 meters * 50 meters). For example, the output data is the first grid area 361 of 100 meters * 100 meters. Other different conditions can also set different sizes of grids. Grid area.

Figure 4 is a flow chart of analyzing the environmental attributes of user communication in the method of focusing on the problem area of mobile users of the present invention, including feature recognition algorithm 42 and artificial intelligence methods. That is, the environmental attribute algorithm 24 (environmental attribute calculation module) related to user communication shown in Figure 2 sequentially passes or includes the feature recognition algorithm 42 and the artificial intelligence method. Finally, the user's indoor environment 461 can be obtained. Or the attributes or discrimination results of the outdoor environment 462.

Knowing that users use communication services in an indoor environment 461 or an outdoor environment 462, the network improvement plan adopted for the telecom operator will be completely different. For example, there will be input data including KPI (key performance indicator) data 411 for users engaged in indoor activities, KPI (key performance indicator) data 412 for users engaged in outdoor activities, and user interface data (such as user S1/X2 interface data 413) Input to the feature recognition algorithm 42 to communicate the signal quality and signal strength of the input data through the communication specific analysis 421 (such as RF/radio specific analysis) of the feature recognition algorithm 42 Information (such as RF) KPIs (key performance indicators) screen out specific user signal strength 431 that can distinguish indoor or outdoor, and generate specific analysis 422 through the network event of the feature recognition algorithm 42 for specific analysis 422 Screen out specific user handover events 432 from the event KPIs (key performance indicators) such as handover or call disconnection of the input data, and analyze the specificity of user behavior through the feature recognition algorithm 42 423 Filter the user's continuous use time 433 that can distinguish the specific indoor or outdoor use from the various user behaviors of the user's S1/X2 interface data such as the continuous use time of the input data, the use of different application types (APP) types, etc., so through the feature recognition The algorithm 42 can generate three characteristics: the user signal strength 431, the user handover event 432, and the user's continuous use time 433.

Then, use artificial intelligence methods for analysis. For example, after the feature recognition algorithm 42 captures the KPI (key performance indicator) parameters generated by the access network component 11 (such as eNB) and the core network component 12 (such as MME), use The artificial intelligence method executes the principal component analysis of KPI (key performance indicator) parameters. This artificial intelligence method calculates the variance of each feature such as user signal strength 431, user handover event 432, and user continuous use time 433. In order to prevent the artificial intelligence method from being executed for a long time, it will be less than the threshold value of variance (such as 100). However, in order to preserve the readability of the original data, the output of the feature recognition algorithm 42 still retains the original data without the linear transformation of the feature space, and only deletes those with the smaller variance (that is, the lower variance is retained). Larger ones) to adopt artificial intelligence methods.

In addition, this artificial intelligence method belongs to the category of supervised learning, and captures the output of the feature recognition algorithm 42, and includes the user signal strength 431, the user movement range 434, and also includes the user's continuous use time 433 as the input feature, and also takes The user handover event 432 serves as a label for the data. At the same time, this artificial intelligence method can first find the user's signal strength 431, user interaction Pass event 432, user continuous use time 433, user movement range 434 and other input characteristics of at least one of the segmentation conditions (such as optimal segmentation conditions) to perform the procedure of maximum message gain analysis 441, and the message in the maximum message gain analysis The gain is the difference between the information entropy before and after the segmentation. Once the foregoing procedure is executed, one segmentation can be obtained, and multiple segmentations that maximize the objective function can be performed for the segmentation result 442. In addition, due to the significant disparity in the types of labeling, the accuracy of the objective function in the objective function is maximized, and the recall and precision are adopted to analyze the number of reverse segmentation layers. 443 , That is, the best model is taken out before the inverted segmentation layer number is generated according to the test error. Through the analysis process of the above artificial intelligence method, it can be concluded that the environmental attribute judgment criterion 45 of the user communication is [1] if the user's signal strength is less than the strength threshold, or [2] the user's moving speed is less than the speed threshold, or [3] the user If the continuous use time is greater than the time threshold and the user handover event is equal to the event threshold, the artificial intelligence method determines that the user is performing the communication service in the indoor environment 461; otherwise, the artificial intelligence method determines that the user is performing the communication service in the outdoor environment 462.

As shown in Figure 2, after using the KPI analysis statistical method 22 (KPI analysis and statistics module) related to the user communication base station level to obtain the KPI (key performance indicator) analysis statistics of the user communication base station level , Use the grid location algorithm 23 (grid location calculation module) related to dynamic user communication to reduce the base station level to the grid level. The coverage of a base station can be cut into at least two grid areas. The number of grids is calculated dynamically.

As shown in Figure 3, the grid location algorithm 23 (grid location calculation module) related to dynamic user communication in Figure 2 is based on the base station (service base station) used by the user and the coverage of the base station, The base station is located in the urban or suburban area, the number of users of the base station, and other judgment and analysis criteria, to cut the coverage of the base station to obtain multiple small-area grids and large-area networks. Grids or grids of different sizes. For example, basically, the closer to the urban area and the greater the number of users (users) of the base station, the finer or smaller the grid should be cut, and the closer to the suburbs, the fewer the number of users (users) of the base station. , You can do coarser or larger mesh cutting.

As shown in Figure 4, for engineers, although the grid area has effectively reduced the scope of the problem area, knowing that the user is in an indoor or outdoor environment will be more helpful to the development of improvement plans. For example, first use the feature recognition algorithm 42 to select KPIs (key performance indicators) that can clearly identify indoor or outdoor. In order to avoid too long implementation of artificial intelligence methods, the KPIs (key performance indicators) of users who are mainly engaged in indoor activities are respectively selected. Data 411 and KPI (key performance indicator) data 412 of users mainly engaged in outdoor activities are used as input, and user S1/X2 interface data 413 is selected as the input data for the user behavior pattern, and then the KPI selected by the feature recognition algorithm 42 (Key Performance Indicator) project, and then use or import artificial intelligence algorithm (such as decision tree algorithm or its decision tree cart 4.5 algorithm), can summarize the KPI (key performance indicator) items selected by the feature recognition algorithm 42 A set of appropriate threshold values can be used to use the threshold values of these KPI (Key Performance Indicators) items to identify and analyze indoor or outdoor for each user at different points in time. At this time, the abnormal communication area is removed from The size of the grid area is then reduced to indoor or outdoor.

In other words, as shown in FIGS. 1 to 4, the method of focusing on the problem area of the mobile user of the present invention may include the following steps S11 to S13.

In step S11, the grid location algorithm 23 (grid location calculation module) related to dynamic user communication shown in Figure 2 receives the location data of the user shown in Figure 3 or GPS location data (such as GPS longitude data 311). , GPS latitude data 312) and user interface data shown in Figure 4 (such as user S1/X2 interface data). For example, the user interface data shown in Figure 1 is the interface captured or analyzed by the data capture module. Obtain the signaling data between the network element 11 and the core network element 12 (including access to the signaling data between the network element 11 and access to the signaling data between the network element 11 and the core network element 12), and the user The GPS location data is user interface data (such as user S1/X2 interface data) calculated through mathematical models or obtained from the user (user terminal).

In step S12, the grid location algorithm 23 (grid location calculation module) related to dynamic user communication shown in Figure 2 receives the geographic location data of the base station (serving base station) used by the user. For example, the grid location algorithm 23 (grid location calculation module) related to dynamic user communication can know the base station used by the user through the user interface data shown in Figure 4 (such as user S1/X2 interface data 413). Service base station), and then compare the geographic location data of the base station (service base station) or the population data of the geographic administrative area to get the city form of the base station (user service base station), such as urban or suburban type .

In step S13, the grid location algorithm 23 (grid location calculation module) related to dynamic user communication shown in Fig. 2 receives data on the number of users of the base station (serving base station). For example, the grid location algorithm 23 (grid location calculation module) related to dynamic user communication can use different KPIs (key performance indicators) such as the number of users (number of users), data flow, data speed, or the number of links established on the base station. ) In order to obtain the user number data of the base station, and then integrate the user number data and the analysis information of the city form, the cutting size of the required grid can be dynamically adjusted.

In addition, the method of focusing on the problem area of the mobile user of the present invention may further include the following steps S21 to S22.

In step S21, the environment attribute algorithm 24 (environment attribute calculation module) related to user communication shown in Fig. 2 receives the data of the environment mode used by the user. For example, first compare the KPI (Key Performance Indicators) data 411 of users engaged in indoor activities and the KPI (Key Performance Indicators) of users engaged in outdoor activities as shown in Figure 4. Key performance indicators) data 412, user interface data (such as user S1/X2 interface data 413) and other relevant information about the user’s environment are input to the feature recognition algorithm 42, and then through the communication specificity analysis of the feature recognition algorithm 42 421 ( Programs such as RF/radio specific analysis), network event specific analysis 422, user behavior specific analysis 423, etc., will be able to separate the specific factors (such as user signal strength 431, user The handover event 432, the user's continuous use time 433, and the user's moving range 434) are analyzed.

In step S22, the environment attribute algorithm 24 (environment attribute calculation module) related to user communication shown in FIG. 2 receives the specific factor data of the environment mode of the user. For example, the environmental attribute algorithm 24 (environmental attribute calculation module) related to user communication can incorporate the specific factors of the user's usage environment pattern into the usage environment analysis model, and the usage environment analysis model can use artificial intelligence methods or various statistical models , Analyze the specific factors of the user's usage environment in a way of weighting, ranking of importance, or setting thresholds to analyze the state of the user's usage environment.

In summary, the method of focusing on the problem areas of mobile users in the present invention can at least have the following features, advantages or technical effects.

1. The present invention provides a smart method for narrowing down user communication abnormalities in mobile networks to find problem areas. By accessing the signaling data of network components and core network components as analysis data, it eliminates the labor-intensive and cost-intensive drive test sampling. Partially general question. At the same time, the data method used in the intelligent analysis of the present invention can greatly reduce the demand for manpower or time, and can also eliminate the cognitive deviation caused by different engineers in the analysis.

2. The present invention provides an intelligent method for narrowing the abnormal search problem area of user communication in a fast mobile network. It first dynamically analyzes the grid location of user communication, and then uses the environment attribute algorithm related to user communication (including feature recognition algorithm). Method and artificial intelligence method) For the indoor or outdoor attributes in the grid area where the user is located, the most suitable network improvement solutions are matched for different attributes, which greatly reduces the number of continuous attempts to various inappropriate network improvement solutions, and not only shortens the user's communication abnormal time , It is also conducive to reducing the number of customer complaints.

3. The present invention uses intelligent analysis algorithms (such as feature recognition algorithms and artificial intelligence methods, etc.) to assist telecom operators to quickly and accurately find specific poor-quality grids and preliminary causes within the coverage of the base station. Optimization of subsequent network construction and quality improvement.

4. The present invention creates grid data through signaling data and positioning data, and analyzes indoor or outdoor information through feature recognition algorithms and artificial intelligence methods, and overlays the user’s grid data with KPI (key performance indicator) data , The network quality status in each grid area can be obtained, and the indoor or outdoor analysis results can be added to more clearly grasp the problem area and propose appropriate improvement plans to help the telecom operator to quickly and accurately find out The grid covers specific poor network areas and optimization directions, and can also reduce the number of user complaints.

5. The present invention provides a smart method for reducing abnormal areas of user communication in mobile networks. In a real wireless network environment, the telecom operator needs to accurately analyze the environmental attributes of the abnormal area of user communication in order to provide a suitable quality improvement plan. The present invention uses the signaling data of the access network components and core network components obtained in the network management system, and the grid data created by matching the user's positioning data, and uses feature recognition algorithms and artificial intelligence methods to analyze that the user is indoors Or outdoor environmental information to facilitate accurate analysis and quickly focus on the target area that needs improvement, which can improve the efficiency of the telecommunication operation terminal to find abnormal areas of user communication.

6. The method of the present invention to focus on the problem area of mobile users uses the network management system to access the signaling data of the network components and core network components, and use the positioning data of the user and the grid location algorithm related to dynamic user communication ( Grid position calculation module), and use feature identification to perform Algorithms and artificial intelligence methods establish indoor or outdoor identification rules, and complete the development of intelligent analysis algorithms for narrowing user communication abnormal areas to find problem areas, so as to assist telecom operators to accurately identify the use environment attributes of user communication abnormal areas.

7. The present invention may be applied to, for example, mobile telecom operators, mobile network optimization and maintenance providers, and the products that may be applied are, for example, mobile network planning products, mobile network maintenance products, etc.

The above embodiments are only illustrative of the principles, features and effects of the present invention, and are not intended to limit the scope of implementation of the present invention. Anyone familiar with the art can comment on the above without departing from the spirit and scope of the present invention. Modifications and changes to the implementation form. Any equivalent changes and modifications made using the content disclosed in the present invention should still be covered by the scope of the patent application. Therefore, the protection scope of the present invention should be as listed in the scope of the patent application.

21: Access to the signaling data of network components and core network components

22: KPI analysis and statistics on the user communication base station level

23: Grid location algorithm for dynamic user communication

24: Algorithm of environmental attributes related to user communication

25: Appropriate network improvement solutions

Claims (12)

A method of focusing on the problem areas of mobile users, including: Provide or retrieve signaling data for access network components and core network components; and According to the signaling data of the access network component and the core network component, use or go through a KPI (key performance indicator) analysis and statistics method related to the user communication base station level, and a grid location related to dynamic user communication The algorithm and an environmental attribute algorithm related to user communication are used to focus or find out the problem area or abnormal area of user communication, Among them, the KPI analysis and statistical method related to the user communication base station level analyzes or counts the user's KPI (key performance indicators) and the KPI is the base station level, and the grid location algorithm related to dynamic user communication is based on the user The location dynamically corresponds to the relative location of the grid of different sizes based on the coverage of the base station used by the user and the number of users of the base station, and the environment attribute algorithm of the user’s communication includes or sequentially uses features Identification algorithms and artificial intelligence methods are used to obtain the attributes or discrimination results of the user in an indoor environment or an outdoor environment. For example, the method described in item 1 of the scope of patent application further includes enabling the data acquisition module to acquire at least two signaling data of the X2 interface between the access network components, and the access network components and the core network The signaling data of the S1 interface between the path components to obtain the signaling data of the access network component and the core network component. The method described in item 1 of the scope of patent application, wherein the user’s location data includes the user’s GPS (Global Positioning System) longitude data and GPS latitude data, and the user’s GPS longitude data and GPS latitude data are obtained through The signalling data of the access network element and the core network element is calculated by adding the mathematical model used by the map data of the base station. For example, the method described in item 1 of the scope of patent application further includes making the grid location algorithm related to dynamic user communication use the geographic location data of the base station to analyze and judge the city shape, and then use or import the user’s KPI (Key Performance Indicators) data to get the number of users of the base station. The method described in item 1 of the scope of patent application further includes inputting input data including KPI data of users engaged in indoor activities, KPI data of users engaged in outdoor activities, and user interface data into the feature recognition algorithm. For example, the method described in item 5 of the scope of patent application further includes the communication specific analysis of the feature recognition algorithm to filter the input data from the input data to distinguish the specific user signal strength indoor or outdoor, and to identify the specific user signal strength through the feature The algorithm’s network event generates a specific analysis, by which the input data can be used to screen out the specific indoor or outdoor user handover event, and the user behavior of the algorithm is identified through the feature The specificity analysis screens out the continuous use time of users who can distinguish the specific indoor or outdoor from the input data. For example, the method described in item 1 of the scope of the patent application further includes the artificial intelligence method to find at least one of the user signal strength, user handover event, user continuous use time, and user movement range to maximize the information gain. The analysis procedure, and the information gain in the maximum information gain analysis is the difference between the information entropy before and after the segmentation. For example, the method described in item 1 of the scope of patent application further includes making the grid location algorithm related to dynamic user communication pass through the base station used by the user, the coverage area of the base station, and whether the base station is located in an urban area or The judgment and analysis criterion of the number of users in the suburbs or base stations is to cut the coverage of the base station to obtain multiple small-area grids, large-area grids, or grids with different sizes. For example, the method described in item 1 of the scope of patent application further includes enabling the grid location algorithm related to dynamic user communication to learn the base station used by the user through user interface data, and compare the geographic location of the base station Data or population data of geographic administrative area to get what kind of city form the base station is in. For example, the method described in item 1 of the scope of patent application further includes making the grid location algorithm related to dynamic user communication use the number of users, data flow, data speed, or link establishment of the base station to obtain the number of the base station. The user quantity data is then integrated with the analysis information of the city form to dynamically adjust the cutting size of the grid. For example, the method described in item 1 of the scope of patent application further includes inputting KPI data of users engaged in indoor activities, KPI data of users engaged in outdoor activities, and user interface data into the feature recognition algorithm, and then through the feature recognition algorithm. Communication specific analysis, network event generation specific analysis and user behavior specific analysis procedures are used to analyze the specific factors of the user's different usage environment patterns. For example, the method described in item 11 of the scope of patent application further includes making the environment attribute algorithm related to user communication incorporate the specific factors of the user's usage environment pattern into the usage environment analysis model, and then the usage environment analysis model uses the The artificial intelligence method or statistical model analyzes the specific factors of the user's environment using weights, importance rankings, or thresholds to analyze the user's environment.

Images