CN110636528A - Regional wireless perception problem feature identification method - Google Patents
Regional wireless perception problem feature identification method Download PDFInfo
- Publication number
- CN110636528A CN110636528A CN201910938584.XA CN201910938584A CN110636528A CN 110636528 A CN110636528 A CN 110636528A CN 201910938584 A CN201910938584 A CN 201910938584A CN 110636528 A CN110636528 A CN 110636528A
- Authority
- CN
- China
- Prior art keywords
- data
- index
- kqi
- success rate
- clustering
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 28
- 230000008447 perception Effects 0.000 title claims abstract description 26
- 230000002159 abnormal effect Effects 0.000 claims abstract description 12
- 238000003064 k means clustering Methods 0.000 claims abstract description 9
- 238000012545 processing Methods 0.000 claims abstract description 8
- 238000004140 cleaning Methods 0.000 claims abstract description 4
- 238000010276 construction Methods 0.000 claims description 2
- 238000001914 filtration Methods 0.000 abstract description 5
- 238000012216 screening Methods 0.000 abstract description 5
- 238000004891 communication Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B17/00—Monitoring; Testing
- H04B17/30—Monitoring; Testing of propagation channels
- H04B17/309—Measuring or estimating channel quality parameters
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/02—Arrangements for optimising operational condition
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Electromagnetism (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Quality & Reliability (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The invention discloses a method for identifying regional wireless perception problem features, which comprises the following steps: collecting data of a regional wireless perception key quality index KQI eight busy hours; carrying out data cleaning and null filling on the acquired KQI index data of the region, and constructing the worst index data of the region for one day; processing abnormal values of the constructed data; sending the data processed by the abnormal value into a K-means clustering model, outputting the center of each category and visually presenting the center of each category in a parallel coordinate system; analyzing the clustering center data of each category, and outputting the problem KQI index characteristics of various sample data. The method can be used for screening and discriminating without using a conventional index threshold value filtering method, intelligently classifying through feature engineering, structural data and feature clustering according to historical data, outputting regional problem KQI index features, enabling a solution to be formed in a subsequent quality improvement service process in a targeted mode, finding and solving problems before complaints of users, and improving user experience.
Description
Technical Field
The invention relates to the field of wireless communication, in particular to a method for identifying characteristics of regional wireless perception problems.
Background
In the present day that communication technology and internet technology are changing day by day, the demands of users on communication rate and wireless perception quality are increasing.
The wireless perceptibility is the only standard for measuring the wireless signal quality of a region. The screening and discrimination efficiency of the conventional index threshold value filtering method is low, so that inconvenience is caused for subsequent quality improvement service.
Disclosure of Invention
The invention aims to overcome the defect that the screening and discrimination efficiency of a conventional index threshold value filtering method in the prior art is low, and provides a regional wireless perception problem feature identification method which can quickly identify a wireless perception problem.
In order to solve the prior art problem, the invention discloses a method for identifying the characteristics of regional wireless perception problems, which comprises the following steps:
collecting data of a local wireless perception quality index KQI in eight busy hours;
carrying out data cleaning and null value filling on the collected data of the wireless perception quality index KQI in eight busy hours, and constructing the worst index data of one day in one area;
carrying out abnormal value processing on the constructed data to form a sample data set;
sending the sample data set subjected to abnormal value processing into a K-means clustering model, outputting the center of each category, and visually presenting the center of each category in a parallel coordinate system;
and analyzing the clustering center data of each category, and outputting the wireless perception quality index KQI index characteristics of various sample data.
Further, the regional wireless perceptual quality KQI index data includes: page response success rate, page response delay, page download rate, mobile video initial play success rate, mobile video per minute pause times, pause duration ratio, initial cache delay, streaming media rate, instant messaging response success rate, instant messaging response delay, mobile service game response success rate and mobile service game response delay; the eight busy hour data includes KQI index data of 8:00, 9:00, 10:00, 18:00, 19:00, 20:00, 21:00, 22:00 of the area in one day.
Further, the null padding comprises: and filling missing values in the data of the wireless sensing quality index KQI in eight busy hours by using median.
Further, the constructing worst index data of one area per day specifically includes: and taking the worst value in eight busy hours for each key quality index KQI eight busy hour data.
Further, the abnormal value processing is performed on the worst index data after the construction, specifically: deleting outliers in the worst indicator data, the outliers comprising: the index values of the page response success rate, the page display success rate, the page download rate, the video playing success rate, the video download rate, the instant messaging response success rate and the game response success rate are 0.
Further, the K-means clustering model algorithm comprises the following steps:
the first step is as follows: inputting a clustering number k and a sample data set n;
the second step is that: initializing k clustering centers;
the third step: distributing each data object to the class with the closest distance;
the fourth step: recalculating each clustering center;
the fifth step: judging whether convergence occurs or not, if yes, outputting a clustering result, and if not, returning to the second step for iterative computation;
and a sixth step: and visualizing the output clustering result in a parallel coordinate system.
The invention has the following beneficial effects:
according to the method for identifying the KQI index features of the regional wireless perception problem, screening and discrimination can be performed without using a conventional index threshold value filtering method, intelligent classification is performed through feature engineering, structural data and feature clustering according to a large amount of historical data, the KQI index features of the regional problem are output, a solution is formed in a subsequent quality improvement service process in a targeted mode, the problem is found and solved before complaints of users, and user experience is greatly improved.
Drawings
FIG. 1 is a schematic flow chart of the present invention.
Fig. 2 is a table of data of the wireless sensing eight busy hour indicator according to the present invention.
FIG. 3 is a data table of the present invention in which the abnormal value is constructed and deleted.
FIG. 4 is a flow chart of the steps of the K-means clustering method.
Fig. 5 is a visual result diagram of the clustering result in a parallel coordinate system.
FIG. 6 is a table of problem feature analysis for each cluster center.
Detailed Description
The invention is further described below with reference to the accompanying drawings. The following examples are only for illustrating the technical solutions of the present invention more clearly, and the protection scope of the present invention is not limited thereby.
As shown in fig. 1-3, a method for identifying characteristics of regional wireless perception problems includes the following steps:
collecting data of a regional wireless perception key quality index KQI eight busy hours; the regional wireless perception key quality KQI index data comprises the following data: page response success rate, page response delay, page download rate, mobile video initial play success rate, mobile video per minute pause times, pause duration ratio, initial cache delay, streaming media rate, instant messaging response success rate, instant messaging response delay, mobile service game response success rate and mobile service game response delay; the eight busy hour data includes KQI index data of 8:00, 9:00, 10:00, 18:00, 19:00, 20:00, 21:00, 22:00 of the area in one day.
Carrying out data cleaning and null filling on the acquired KQI index data of the region, and constructing the worst index data of the region for one day; and deleting the area data in less than eight busy hours according to the data integrity requirement. And meanwhile, missing values in the data of the eight busy hour area are filled by using a median. The method for constructing the worst index data comprises the following steps: the worst value in eight busy hours is taken for each index, namely, the positive index (the larger the index is, the better the minimum value is) is taken for the positive index, and the negative index (the smaller the index is, the better the maximum value is) is taken for the negative index. And finally, changing the data of an area in eight busy hours into data representing the worst index of the area in one day.
Processing abnormal values of the constructed data; data that is apparently abnormal in some indicators is deleted. Such as some samples with a rate class index of 0.
Sending the data processed by the abnormal value into a K-means clustering model, outputting the center of each category and visually presenting the center of each category in a parallel coordinate system; the K-means clustering model algorithm shown in fig. 4 includes the following steps:
the first step is as follows: inputting a clustering number k and a sample data set n; the method for selecting the clustering number K in the first step of the K-means clustering model algorithm comprises the following steps: comparing the Sum of Squared Errors (SSE) for different numbers of categories, the number of categories that makes the SSE after clustering smaller is selected among the number of categories, thereby determining the number of clusters. In the invention, when the number of clusters is 14, the SSE is small and the class characteristics presented by each class are most obvious.
The second step is that: initializing k clustering centers;
the third step: distributing each data object to the class with the closest distance;
the fourth step: recalculating each clustering center;
the fifth step: judging whether convergence occurs or not, if yes, outputting a clustering result, and if not, returning to the second step for iterative computation; the method for judging whether convergence occurs comprises the following steps: judging whether the recalculated clustering center changes after data distribution, if not, converging, and if the clustering center is updated, not converging, and needing to distribute data again;
and a sixth step: and visualizing the output clustering result in a parallel coordinate system. In this embodiment, the parallel coordinate system is a common method for visualizing high-dimensional sets and analyzing multivariate data. As shown in fig. 5, in the present invention, the number of clusters is 14, each cluster center is 14 dimensions, 14 broken lines are displayed in a parallel coordinate system, and each inflection point represents an index feature. And 14 broken lines in the parallel coordinate system respectively represent 14 clustering centers, the wave crests and the wave troughs appearing in the broken lines are the problem index characteristics of the category, and the KQI indexes corresponding to the wave crests and the wave troughs in each broken line are output.
Analyzing the cluster center data of each category, wherein the analysis result of the sample data set adopted in the invention is shown in fig. 6, and outputting the problem KQI index characteristics of various sample data, wherein the light-colored indexes in the graph are problem indexes, so as to locate the problem KQI index characteristics of the poor perception area.
Based on the above, the method for identifying the KQI index features of the regional wireless perception problem can be used for screening and discriminating without using a conventional index threshold value filtering method, intelligently classifying the KQI index features of the regional problem through feature engineering, structural data and feature clustering according to a large amount of historical data, outputting the KQI index features of the regional problem, forming a solution in a targeted manner, finding and solving the problem before complaint of a user, and greatly improving user experience.
The above description is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, several modifications and variations can be made without departing from the technical principle of the present invention, and these modifications and variations should also be regarded as the protection scope of the present invention.
Claims (6)
1. A regional wireless perception problem feature identification method is characterized by comprising the following steps:
collecting data of a local wireless perception quality index KQI in eight busy hours;
carrying out data cleaning and null value filling on the collected data of the wireless perception quality index KQI in eight busy hours, and constructing the worst index data of one day in one area;
abnormal value processing is carried out on the constructed worst index data to form a sample data set;
sending the sample data set subjected to abnormal value processing into a K-means clustering model, outputting the center of each category, and visually presenting the center of each category in a parallel coordinate system;
and analyzing the clustering center data of each category, and outputting the wireless perception quality index KQI index characteristics of various sample data.
2. The method as claimed in claim 1, wherein the local wireless perceptual quality KQI index data comprises: page response success rate, page response delay, page download rate, mobile video initial play success rate, mobile video per minute pause times, pause duration ratio, initial cache delay, streaming media rate, instant messaging response success rate, instant messaging response delay, mobile service game response success rate and mobile service game response delay; the eight busy hour data includes KQI index data of 8:00, 9:00, 10:00, 18:00, 19:00, 20:00, 21:00, 22:00 of the area in one day.
3. The method for identifying the regional wireless perception problem features according to claim 1, wherein the null padding specifically comprises: and filling missing values in the data of the wireless sensing quality index KQI in eight busy hours by using median.
4. The method for identifying the regional wireless perception problem features according to claim 1, wherein the constructing worst index data of a region in one day specifically includes: and taking the worst value in eight busy hours for each key quality index KQI eight busy hour data.
5. The method for identifying the regional wireless perception problem features according to claim 1, wherein the abnormal value processing is performed on the worst index data after the construction, specifically: deleting outliers in the worst indicator data, the outliers comprising: the index values of the page response success rate, the page display success rate, the page download rate, the video playing success rate, the video download rate, the instant messaging response success rate and the game response success rate are 0.
6. The method for identifying the regional wireless perception problem features according to claim 1, wherein the K-means clustering model algorithm comprises the following steps:
the first step is as follows: inputting a clustering number k and a sample data set n;
the second step is that: initializing k clustering centers;
the third step: distributing each data object to the class with the closest distance;
the fourth step: recalculating each clustering center;
the fifth step: judging whether convergence occurs or not, if yes, outputting a clustering result, and if not, returning to the second step for iterative computation;
and a sixth step: and visualizing the output clustering result in a parallel coordinate system.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910938584.XA CN110636528A (en) | 2019-09-30 | 2019-09-30 | Regional wireless perception problem feature identification method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910938584.XA CN110636528A (en) | 2019-09-30 | 2019-09-30 | Regional wireless perception problem feature identification method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110636528A true CN110636528A (en) | 2019-12-31 |
Family
ID=68973816
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910938584.XA Pending CN110636528A (en) | 2019-09-30 | 2019-09-30 | Regional wireless perception problem feature identification method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110636528A (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105873113A (en) * | 2015-01-21 | 2016-08-17 | 中国移动通信集团福建有限公司 | Method and system for positioning wireless quality problem |
CN108363810A (en) * | 2018-03-09 | 2018-08-03 | 南京工业大学 | A kind of file classification method and device |
CN109936857A (en) * | 2019-01-21 | 2019-06-25 | 南京邮电大学 | A kind of wireless aware degree intelligent identification Method |
-
2019
- 2019-09-30 CN CN201910938584.XA patent/CN110636528A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105873113A (en) * | 2015-01-21 | 2016-08-17 | 中国移动通信集团福建有限公司 | Method and system for positioning wireless quality problem |
CN108363810A (en) * | 2018-03-09 | 2018-08-03 | 南京工业大学 | A kind of file classification method and device |
CN109936857A (en) * | 2019-01-21 | 2019-06-25 | 南京邮电大学 | A kind of wireless aware degree intelligent identification Method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110909205B (en) | Video cover determination method and device, electronic equipment and readable storage medium | |
CN103793484B (en) | The fraud identifying system based on machine learning in classification information website | |
CN109936857B (en) | Intelligent identification method for wireless perceptibility | |
CN107483613B (en) | Information pushing method | |
CN103795612B (en) | Rubbish and illegal information detecting method in instant messaging | |
CN108038500B (en) | Clustering method, apparatus, computer device, storage medium, and program product | |
CN110830450A (en) | Abnormal flow monitoring method, device and equipment based on statistics and storage medium | |
CN111144941A (en) | Merchant score generation method, device, equipment and readable storage medium | |
CN110347888B (en) | Order data processing method and device and storage medium | |
CN105426441B (en) | A kind of automatic preprocess method of time series | |
CN113051317B (en) | Data mining model updating method, system, computer equipment and readable medium | |
WO2015124024A1 (en) | Method and device for promoting exposure rate of information, method and device for determining value of search word | |
CN113762377A (en) | Network traffic identification method, device, equipment and storage medium | |
CN110889036A (en) | Multi-dimensional information processing method and device and terminal equipment | |
CN111291071A (en) | Data processing method and device and electronic equipment | |
CN110910207A (en) | Method and system for improving commodity recommendation diversity | |
CN111144957A (en) | Information delivery method, device, server and storage medium | |
CN110348717A (en) | Base station value methods of marking and device based on grid granularity | |
CN110636528A (en) | Regional wireless perception problem feature identification method | |
CN108510298B (en) | Target user identification method and device | |
CN110968570A (en) | Distributed big data mining system facing E-commerce platform | |
CN114650239A (en) | Data brushing amount identification method, storage medium and electronic equipment | |
CN111581508A (en) | Service monitoring method, device, equipment and storage medium | |
CN112015726A (en) | User activity prediction method, system and readable storage medium | |
CN110852830A (en) | Target user information acquisition method and service subscription system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
CB02 | Change of applicant information | ||
CB02 | Change of applicant information |
Address after: No.66 Xinfan Road, Gulou District, Nanjing City, Jiangsu Province Applicant after: NANJING University OF POSTS AND TELECOMMUNICATIONS Address before: 210023 No.1 Xichun Road, Yuhuatai District, Nanjing City, Jiangsu Province Applicant before: NANJING University OF POSTS AND TELECOMMUNICATIONS |
|
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191231 |