CN105873105A - Method for anomaly detection and positioning of mobile communication network based on network experience quality - Google Patents
Method for anomaly detection and positioning of mobile communication network based on network experience quality Download PDFInfo
- Publication number
- CN105873105A CN105873105A CN201610262009.9A CN201610262009A CN105873105A CN 105873105 A CN105873105 A CN 105873105A CN 201610262009 A CN201610262009 A CN 201610262009A CN 105873105 A CN105873105 A CN 105873105A
- Authority
- CN
- China
- Prior art keywords
- network
- point
- anomaly
- abnormal
- key performance
- 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.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/02—Arrangements for optimising operational condition
Abstract
The invention discloses a method for anomaly detection and positioning of a mobile communication network based on network experience quality. The method is characterized in that real-time anomaly monitoring and anomaly positioning are performed on a network through an anomaly detection subsystem and an anomaly root cause positioning subsystem which are based on network experience quality; at an anomaly diagnosis part, three key performance indexes are selected from the angle of network synthesis user experience to serve as network diagnosis characteristics, and a method of coarse-grained threshold division and fine-grained clustering is adopted to perform abnormal classification; at an anomaly root cause positioning part, a mode of cumulative distribution function matching is adopted to obtain anomaly symptom characteristics, clustering analysis is performed in each type, and root cause positioning of different anomalies can be performed; and finally a cellular network anomaly detection and anomaly root cause positioning subsystem is formed. According to the method, not only can obvious anomalies be detected, but also potential anomalies can be detected, and anomaly root cause positioning can be performed according to different anomaly types; independent study can be performed, and the anomaly detection root cause positioning accuracy is improved continuously.
Description
Technical field
The invention belongs to mobile communications network monitoring and optimisation technique field, be specifically related to cellular mobile communications networks is experienced matter
Amount (Quality of Experience, QoE) is monitored and to abnormal root in real time because of the method and system positioned,
And monitoring Long Term Evolution (LTE) existing network of 4G universal mobile communications technology and future mobile communications network operation quality and
The processing method of service level.
Background technology
The network optimization for cellular mobile communications networks is a system engineering through whole network Development overall process, also
It is that mobile communication network operator is in order to improve the important technical of network investment benefit, running quality and service level.
Abnormality detection mechanism is an important research content in network optimisation techniques, can be used for finding and positioning honeycomb mobile communication
Deviant Behavior in network.Mobile communication network operator have come into effect abnormal inspection to cellular mobile communications networks at present
Surveying, Key Performance Indicator (Key Performance Indicator, KPI) is mainly monitored by the method used,
Then use empirical value that Key Performance Indicator KPI is carried out hard decision.Being limited in that only when entirely of this method
When portion or part KPI index interested fall outside normality threshold scope, this mechanism just assert network occur abnormal the most also
Report to the police.Analysis shows, this method for detecting abnormality based on KPI threshold value, and its detection probability sets relatedness with threshold value
Greatly, and be there is bigger false-alarm and false dismissal probability in the Deviant Behavior fallen at threshold skirt, be the most often difficult to real
Reflection network quality.Existing method for detecting abnormality based on KPI threshold value is general lack of to be sent out other potential Deviant Behavioies
Existing ability, also cannot accurately and be automatically positioned exception root because of.This mode often relies on the experience of skilled engineer and determines
The root of Deviant Behavior that position has detected because of.
Summary of the invention
The purpose of the present invention is to propose to a kind of mobile radio communication abnormality detection based on network Quality of experience and localization method, logical
Cross and Key Performance Indicator (KPI) is carried out cluster analysis, it is achieved Deviant Behavior based on network Quality of experience diagnoses and divides
Class, be automatically completed different abnormal root because of location.
Present invention mobile radio communication based on network Quality of experience abnormality detection and localization method, from network parameter, sentence
Whether abnormal circuit network is, it is characterised in that: use abnormality detection subsystem based on network Quality of experience and based on network body
The abnormal root of the amount of checking the quality, because of positioning subsystem, utilizes threshold value to divide and feature clustering carries out the anomaly classification that network is overall, with
Time remaining Key Performance Indicator of network is carried out cluster analysis；Concrete operations are:
Described abnormality detection subsystem based on network Quality of experience, first inputs from communication network radio resource controller
The network key performance indications (KPI) that (Radio Network Controller, RNC) gathers, then from access property,
Integrity, three aspects of retentivity select corresponding network key performance indications as network overall experience quality, are used for anti-
Answer the average user perception of all users in network；Use the mode that threshold value divides again, according to above-mentioned three classes being previously set
The decision threshold of Key Performance Indicator, marks off obvious abnormal, slight abnormality and the most normal 3 class points；Use from group
Knit neural mapping (Self-Organizing Map, SOM) and two kinds of clustering algorithms pair of K central point algorithm (K-mediods)
Slight abnormality point and obvious abnormity point carry out fine granularity analysis again: it is a kind of unsupervised that self-organizing nerve maps SOM
Practise algorithm, comprise input layer and output layer: the input vector of the corresponding higher-dimension of input layer, output layer is by a series of tissues
M on two-dimensional grid × N number of ordered nodes is constituted, and input node is connected by weight vectors with output node；Learn
During habit, find the output layer unit i.e. winning unit that distance is the shortest therewith, be updated；Meanwhile, by adjacent domain
Right value update, make output node keep input vector topological characteristic, ultimately form M × N number of group, the most right
Should different types of exception；And K central point algorithm is a kind of Classic Clustering Algorithms, first randomly select one group of cluster sample
This is as center point set, corresponding one bunch of each central point；Then each sample point distance to each central point is calculated, will
Sample point is put in that bunch that distance center point is the shortest, and in calculating each bunch of middle-range bunch, the absolutely degree error of each sample point distance is
Little point is as new central point；If the newest center point set is identical with former center point set, then algorithm terminates；Use
Self-organizing nerve maps to enter, with K central point algorithm algorithm, the abnormal data set that slight abnormality point is corresponding with obvious abnormity point
Row cluster, finally gives K Key Performance Indicator set, it is simply that Exception Type；
Described abnormal root based on network Quality of experience is because of positioning subsystem, crucial by finding between normal point and abnormity point
The difference of performance indications parameter, the possible cause of location abnormity point；First the abnormity point class obtained from abnormality detection subsystem
Type sets out, the cumulative distribution function (Cumulative of remaining Key Performance Indicator between contrast normal point and abnormity point
Distribution Function, CDF) curve chart, extracts corresponding abnormal symptom Key Performance Indicator, builds abnormality disease
Shape Key Performance Indicator storehouse；In each type, it is again with K central point algorithm key to the symptom of above-mentioned acquisition
Can cluster by index feature, and then obtain different abnormal symptom lists, realize root because of location by searching this list.
Present invention mobile radio communication based on network Quality of experience abnormality detection and localization method take by based on network
The abnormality detection subsystem of Quality of experience and abnormal root based on network Quality of experience are because of positioning subsystem, real to real-time performance
Time exception monitoring and root cause analysis；In abnormity diagnosis part, due to the angle from network Quality of experience, select three classes crucial
Performance indications, as network holistic diagnosis feature, use coarseness threshold value to divide and the method for fine granularity cluster carry out exception
Classification, can effectively avoid false-alarm and false dismissal, it is possible to detect the most abnormal and potential exception；At abnormal Gen Yinding
Bit position, owing to normal point and remaining Key Performance Indicator of abnormity point using the mode of cumulative distribution function coupling, builds
Abnormal symptom Key Performance Indicator storehouse, and in each Exception Type, carry out cluster analysis, it is possible to achieve different abnormal roots
Because of location.The system taking the inventive method to be built can be with autonomic learning such that it is able to constantly improve abnormality detection
With root because of the accuracy of location.
Accompanying drawing explanation
Fig. 1 is mobile radio communication abnormality detection based on network Quality of experience and the overall structure schematic diagram of localization method.
Fig. 2 is the structural representation of abnormality detection subsystem B based on network Quality of experience.
Fig. 3 is that abnormal root based on network Quality of experience is because of the structural representation of positioning subsystem C.
Detailed description of the invention
Embodiment 1:
The present embodiment mobile radio communication based on network Quality of experience abnormality detection and localization method, including using based on network
The abnormality detection subsystem of Quality of experience and abnormal root based on network Quality of experience, because of positioning subsystem, utilize threshold value to divide
With the anomaly classification that feature clustering carries out network overall experience quality, remaining symptom Key Performance Indicator of network is carried out simultaneously
Cluster analysis；
Fig. 1 gives the overall structure signal of mobile radio communication abnormality detection based on network Quality of experience and localization method
Figure.The present embodiment mobile radio communication based on network Quality of experience abnormality detection and localization method specifically include following steps:
Key Performance Indicator (the Key gathered from communication network radio resource controller (Radio Network Controller, RNC)
Performance Indicator, KPI), form Key Performance Indicator storehouse A, as the input of whole system.In this enforcement
In example 1, Key Performance Indicator storehouse A is 27 days that 2659 communities that 3G RNC gathers are moved in Chengde City, Hebei Province
62 Key Performance Indicators.First in terms of access property, integrity, retentivity three, select corresponding key performance
Index, as the Quality of experience (Quality of Experience, QoE) that network is overall, useful for institute in reaction network
The average user perception at family；Select in the present embodiment 1 is wireless access, voice quality and Traffic Channel cutting off rate three
Class Key Performance Indicator A1.Then, this three class Key Performance Indicator A1 is inputted abnormality detection subsystem B, obtains
Different Exception Type B4；Finally, by these Exception Types B4 and its in addition to above-mentioned three class Key Performance Indicator A1
The remaining abnormal root of Key Performance Indicator A2 input, because of positioning subsystem C, obtains abnormal symptom list D1, by searching this
A little lists can be obtained by abnormal root because of D.
Fig. 2 gives the structural representation of abnormality detection subsystem B based on network Quality of experience: at the present embodiment 1
In, abnormality detection subsystem B input is Key Performance Indicator storehouse A, first from access property, integrity, retentivity three
Individual aspect, selects wireless access, voice quality and Traffic Channel cutting off rate three class Key Performance Indicator A1 conduct respectively
The Quality of experience that network is overall, for the average user perception of users all in reaction network.Owing to abnormal degree is different,
The most abnormal counts seldom, is easily submerged in normal point, therefore data is carried out chromatographic analysis.First with threshold value
Divide B1 and 10000 data points before in the A of Key Performance Indicator storehouse are carried out coarseness classification.So-called coarseness refers to
The levels of precision that problem analysis is reached, fine granularity below also means this.Coarseness analyzes basis specifically
The decision threshold of the above-mentioned three class Key Performance Indicator A1 being previously set, obtains obvious abnormal, slight abnormality and substantially
Normal 3 class points.This coarseness analysis is similar with the hard decision of operator, when some Key Performance Indicator index less than or
During higher than the threshold value set, it is assumed that be that network exception occurs and reports to the police.The present embodiment 1 sort out obvious and the most different
Often point adds up to 4079.Then, slight and obvious two class abnormity point B2 are carried out fine granularity analysis again.This step master
Self-organizing nerve to be used maps mixing of (Self-Organizing Map, SOM) and K central point algorithm (K-mediods)
Data set is clustered by hop algorithm, obtains different Exception Types.So-called cluster refers to enter the group of unknown classification
The method of row classification, is a class important method of data mining, uses self-organizing nerve here and maps (SOM) and K
Two kinds of clustering algorithms of central point algorithm (K-mediods).SOM is a kind of unsupervised learning algorithm, comprise input layer and
Output layer.The input vector of the corresponding higher-dimension of input layer, output layer is by a series of M × N being organized on two-dimensional grid
Individual ordered nodes is constituted, and input node is connected by weight vectors with output node；In learning process, find distance therewith
The shortest output layer unit i.e. winning unit, is updated；Meanwhile, by the right value update of adjacent domain, output node is made
Keep the topological characteristic of input vector, ultimately form M × N number of group, respectively corresponding different types of exception.And K
Central point algorithm K-mediods algorithm is a kind of Classic Clustering Algorithms, first randomly selects one group of cluster sample as center
Point set, corresponding one bunch of each central point；Then calculate each sample point to each central point distance (as euclidean away from
From), sample point is put in that bunch that distance center point is the shortest, by the exhausted degree of each sample point distance in each bunch of middle-range bunch
Minimum the putting as new central point of error；If the newest center point set is identical with former center point set, then algorithm terminates.
Use SOM with K-mediods algorithm that the abnormal data set that slight abnormality point is corresponding with obvious abnormity point is clustered,
Finally give K class Key Performance Indicator set, here it is Exception Type B4.Associating SOM's and K-mediods is mixed
Close clustering algorithm B2, high dimensional data can be made to be transformed on two dimensional surface, there is typical effect of visualization, the most also
Reduce dimension and the data volume of data.The present embodiment 1 uses the recommendation 10 × 10 of SOM algorithm, to SOM
Type center after cluster, selects suitable K value (most suitable according to operating experience 10 to 20), uses K-mediods
Algorithm, merges the type with bigger similarity, and (above-mentioned three classes are key to finally give K big type
The combination of the different values of energy index A1), namely Exception Type B4.Above-mentioned three classes of new subzone network are key
Index A1 can input this module, by the way of similarity mates, i.e. first pass through coarseness and divide, then with corresponding
Exception Type central point coupling, affiliated type may finally be found from Exception Type B4.
Fig. 3 is that abnormal root based on network Quality of experience is because of the structural representation of positioning subsystem.
At root abnormal shown in Fig. 3 because of in positioning subsystem structure chart, it can be seen that whole abnormal root is because of the knot of positioning subsystem
Structure.First, to the most normal and substantially exception the point obtained by threshold value division B1 in abnormality detection subsystem B
Remaining Key Performance Indicator A2 (i.e. 59 KPI of residue) carries out abnormal symptom and extracts C1.Method is to calculate normal point
With the Cumulative Distribution Function (CDF) of remaining Key Performance Indicator A2 of abnormity point, compare difference between the two.When
When the CDF of certain class Key Performance Indicator (KPI) exists notable difference, then it is believed that this KPI is probably abnormal product
Raw reason.Utilize the mode that this CDF compares, build abnormal symptom KPI storehouse.Then, to abnormality detection subsystem
Each Exception Type B4 that system B obtains, uses K-mediods algorithm to occur in it in abnormal symptom KPI storehouse
KPI carries out abnormal symptom cluster C2, and then obtains the different abnormal symptom list D1 of such exception, i.e. causes such different
The possible KPI set often produced.Abnormal root is can be obtained by because of D by searching these lists.In embodiment 1,
First the wireless access of each community, voice quality, Traffic Channel call drop three class KPI in detection test data, by it
Mate with the output abnormality type B 4 of abnormality detection subsystem, obtain affiliated Exception Type.Extract again community its
Remaining Key Performance Indicator A2, searches the abnormal symptom list D1 under this abnormal class, and final analysis goes out abnormal root because of D.
Knowable to the description and the process of analysis of example above operating process, the input of the present invention is to gather to join from the network of network element
Number, calculating process is to realize on the hardware such as server, and output then can export different in modes such as text or screen show
Whether normal, Exception Type and abnormal root are because of these results.
Present invention mobile radio communication based on network Quality of experience abnormality detection and localization method, by checking the quality based on network body
The abnormality detection subsystem of amount and abnormal root based on network Quality of experience are because of positioning subsystem, exception real-time to real-time performance
Monitoring；In abnormity diagnosis part, from the angle of network synthesis Consumer's Experience, select three class Key Performance Indicators as network
Diagnostic characteristic, uses coarseness threshold value to divide and the method for fine granularity cluster carries out anomaly classification；Network Abnormal root because of
Position portion, uses the mode of Cumulative Distribution Function coupling to build abnormal symptom KPI storehouse, enters in each Exception Type
Row cluster analysis, obtains abnormal symptom list, for position different abnormal roots because of, finally constitute Cellular Networks abnormality detection
With abnormal root because of positioning subsystem.Owing to the inventive method takes, Exception Type is carried out thick fine-grained chromatographic analysis,
Be capable of detecting when substantially abnormal and potential exception, and for different Exception Types carry out abnormal root because of location.It addition,
The clustering algorithm that the system taking the inventive method to be built uses with autonomic learning, thus can constantly improve abnormal inspection
Survey and root is because of the accuracy of location.
Claims (1)
1. mobile radio communication abnormality detection based on network Quality of experience and a localization method, from network parameter,
Whether judge Network Abnormal, it is characterised in that: use abnormality detection subsystem based on network Quality of experience and based on network
The abnormal root of Quality of experience, because of positioning subsystem, utilizes threshold value to divide and feature clustering carries out the anomaly classification that network is overall,
Remaining Key Performance Indicator of network is carried out cluster analysis simultaneously；Concrete operations are:
Described abnormality detection subsystem based on network Quality of experience, first inputs and adopts from communication network radio resource controller
The network key performance indications of collection, then select corresponding network key in terms of access property, integrity, retentivity three
Performance indications are as network overall experience quality, for the average user perception of users all in reaction network；Use threshold again
Value divide mode, according to the decision threshold of the above-mentioned three class Key Performance Indicators being previously set, mark off obvious abnormal,
Slight abnormality and the most normal 3 class points；Self-organizing nerve is used to map and two kinds of clustering algorithms pair of K central point algorithm
Slight abnormality point and obvious abnormity point carry out fine granularity analysis again: it is that a kind of unsupervised study is calculated that self-organizing nerve maps
Method, comprises input layer and output layer: the input vector of the corresponding higher-dimension of input layer, output layer is organized in two by a series of
M × N number of ordered nodes on dimension grid is constituted, and input node is connected by weight vectors with output node；Learnt
Cheng Zhong, finds the shortest output layer unit i.e. winning unit of distance to be therewith updated；Meanwhile, by the weights of adjacent domain
Update, make output node keep the topological characteristic of input vector, ultimately form M × N number of group, the most corresponding different
The exception of type；And K central point algorithm is a kind of Classic Clustering Algorithms, first randomly select the sample conduct of one group of cluster
Center point set, corresponding one bunch of each central point；Then each sample point distance to each central point is calculated, by sample point
Put in that bunch that distance center point is the shortest, the point that in calculating each bunch of middle-range bunch, the error of degree absolutely of each sample point distance is minimum
As new central point；If the newest center point set is identical with former center point set, then algorithm terminates；Use self-organizing
The abnormal data set that slight abnormality point is corresponding with obvious abnormity point is clustered by neural mapping with K central point algorithm algorithm,
Finally give K Key Performance Indicator set, it is simply that Exception Type；
Described abnormal root based on network Quality of experience is because of positioning subsystem, crucial by finding between normal point and abnormity point
The difference of performance indications parameter, the possible cause of location abnormity point；Then the abnormity point class obtained from abnormality detection subsystem
Type sets out, the cumulative distribution function curve chart of remaining Key Performance Indicator between contrast normal point and abnormity point, extracts correspondence
Abnormal symptom Key Performance Indicator, build abnormal symptom Key Performance Indicator storehouse；In each type, it is again with K
The abnormal symptom Key Performance Indicator feature of above-mentioned acquisition is clustered by central point algorithm, and then obtains different abnormality diseases
Shape list, realizes root because of location by searching this list.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610262009.9A CN105873105B (en) | 2016-04-22 | 2016-04-22 | A kind of mobile radio communication abnormality detection and localization method based on network Quality of experience |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610262009.9A CN105873105B (en) | 2016-04-22 | 2016-04-22 | A kind of mobile radio communication abnormality detection and localization method based on network Quality of experience |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105873105A true CN105873105A (en) | 2016-08-17 |
| CN105873105B CN105873105B (en) | 2018-07-03 |
Family
ID=56629062
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610262009.9A Active CN105873105B (en) | 2016-04-22 | 2016-04-22 | A kind of mobile radio communication abnormality detection and localization method based on network Quality of experience |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105873105B (en) |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106254153A (en) * | 2016-09-19 | 2016-12-21 | 腾讯科技（深圳）有限公司 | A kind of Network Abnormal monitoring method and apparatus |
| CN107105432A (en) * | 2017-06-12 | 2017-08-29 | 北京奇虎科技有限公司 | The detection method and device of WLAN |
| CN107395450A (en) * | 2017-06-15 | 2017-11-24 | 腾讯科技（深圳）有限公司 | Using the monitoring method and device, storage medium, electronic installation of the situation of login |
| CN107809780A (en) * | 2016-09-09 | 2018-03-16 | 中兴通讯股份有限公司 | It is a kind of to the active process method, apparatus of user equipment access failure and base station |
| CN108076197A (en) * | 2016-11-14 | 2018-05-25 | 中移(苏州)软件技术有限公司 | A kind of detection method and device of terminal network performance degradation |
| CN108260148A (en) * | 2016-12-28 | 2018-07-06 | 华为技术服务有限公司 | A kind of fault detection method and device |
| CN108271194A (en) * | 2016-12-31 | 2018-07-10 | 中国移动通信集团北京有限公司 | A kind of information processing method and device |
| CN109547248A (en) * | 2018-11-20 | 2019-03-29 | 上海航天测控通信研究所 | Based on artificial intelligence in orbit aerocraft ad hoc network method for diagnosing faults and device |
| CN109634819A (en) * | 2018-10-26 | 2019-04-16 | 阿里巴巴集团控股有限公司 | Alarm root is because of localization method and device, electronic equipment |
| CN109981326A (en) * | 2017-12-28 | 2019-07-05 | 中国移动通信集团山东有限公司 | The method and device of home broadband perception fault location |
| CN110574338A (en) * | 2017-04-20 | 2019-12-13 | 思科技术公司 | Root cause discovery engine |
| CN111064614A (en) * | 2019-12-17 | 2020-04-24 | 腾讯科技（深圳）有限公司 | Fault root cause positioning method, device, equipment and storage medium |
| CN111200530A (en) * | 2019-12-31 | 2020-05-26 | 畅捷通信息技术股份有限公司 | Method and device for performing root cause analysis based on KPI (Key performance indicator) |
| CN111327435A (en) * | 2018-12-13 | 2020-06-23 | 中兴通讯股份有限公司 | Root cause positioning method, server and storage medium |
| CN111405605A (en) * | 2020-03-24 | 2020-07-10 | 东南大学 | Wireless network interruption detection method based on self-organizing mapping |
| CN111628901A (en) * | 2019-02-28 | 2020-09-04 | 华为技术有限公司 | Index abnormality detection method and related device |
| CN111698700A (en) * | 2019-03-15 | 2020-09-22 | 大唐移动通信设备有限公司 | Method and device for judging working state of cell |
| CN111935761A (en) * | 2020-07-15 | 2020-11-13 | 山东建筑大学 | Method and system for automatically positioning quality difference root cause of mobile communication network |
| CN112543465A (en) * | 2019-09-23 | 2021-03-23 | 中兴通讯股份有限公司 | Abnormity detection method, abnormity detection device, terminal and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102821002A (en) * | 2011-06-09 | 2012-12-12 | 中国移动通信集团河南有限公司信阳分公司 | Method and system for network flow anomaly detection |
| CN103581971A (en) * | 2012-07-23 | 2014-02-12 | 中兴通讯股份有限公司 | Method, device and access points for detecting abnormity of centralized wireless local area network |
| CN103945442A (en) * | 2014-05-07 | 2014-07-23 | 东南大学 | System anomaly detection method based on linear prediction principle in mobile communication system |
| CN104735710A (en) * | 2015-03-18 | 2015-06-24 | 大连理工大学 | Mobile network performance early warning pre-judging method based on trend extrapolation clustering |
-
2016
- 2016-04-22 CN CN201610262009.9A patent/CN105873105B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102821002A (en) * | 2011-06-09 | 2012-12-12 | 中国移动通信集团河南有限公司信阳分公司 | Method and system for network flow anomaly detection |
| CN103581971A (en) * | 2012-07-23 | 2014-02-12 | 中兴通讯股份有限公司 | Method, device and access points for detecting abnormity of centralized wireless local area network |
| CN103945442A (en) * | 2014-05-07 | 2014-07-23 | 东南大学 | System anomaly detection method based on linear prediction principle in mobile communication system |
| CN104735710A (en) * | 2015-03-18 | 2015-06-24 | 大连理工大学 | Mobile network performance early warning pre-judging method based on trend extrapolation clustering |
Non-Patent Citations (1)
| Title |
|---|
| PICHANUN SUKKHAWATCHANI等: "Performance evaluation of anomaly detection in cellular core networks using self-organizing map", 《ENGINEERING/ELECTRONICS, COMPUTER, TELECOMMUNICATIONS AND INFORMATION TECHNOLOGY，ECTI-CON》 * |
Cited By (30)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107809780A (en) * | 2016-09-09 | 2018-03-16 | 中兴通讯股份有限公司 | It is a kind of to the active process method, apparatus of user equipment access failure and base station |
| CN106254153A (en) * | 2016-09-19 | 2016-12-21 | 腾讯科技（深圳）有限公司 | A kind of Network Abnormal monitoring method and apparatus |
| CN108076197A (en) * | 2016-11-14 | 2018-05-25 | 中移(苏州)软件技术有限公司 | A kind of detection method and device of terminal network performance degradation |
| CN108260148B (en) * | 2016-12-28 | 2021-02-09 | 华为技术服务有限公司 | Fault detection method and device |
| CN108260148A (en) * | 2016-12-28 | 2018-07-06 | 华为技术服务有限公司 | A kind of fault detection method and device |
| CN108271194B (en) * | 2016-12-31 | 2021-09-03 | 中国移动通信集团北京有限公司 | Information processing method and device |
| CN108271194A (en) * | 2016-12-31 | 2018-07-10 | 中国移动通信集团北京有限公司 | A kind of information processing method and device |
| CN110574338B (en) * | 2017-04-20 | 2022-06-24 | 思科技术公司 | Root cause discovery method and system |
| CN110574338A (en) * | 2017-04-20 | 2019-12-13 | 思科技术公司 | Root cause discovery engine |
| US11954568B2 (en) | 2017-04-20 | 2024-04-09 | Cisco Technology, Inc. | Root cause discovery engine |
| CN107105432A (en) * | 2017-06-12 | 2017-08-29 | 北京奇虎科技有限公司 | The detection method and device of WLAN |
| CN107395450A (en) * | 2017-06-15 | 2017-11-24 | 腾讯科技（深圳）有限公司 | Using the monitoring method and device, storage medium, electronic installation of the situation of login |
| CN109981326A (en) * | 2017-12-28 | 2019-07-05 | 中国移动通信集团山东有限公司 | The method and device of home broadband perception fault location |
| CN109981326B (en) * | 2017-12-28 | 2022-01-25 | 中国移动通信集团山东有限公司 | Method and device for positioning household broadband sensing fault |
| CN109634819A (en) * | 2018-10-26 | 2019-04-16 | 阿里巴巴集团控股有限公司 | Alarm root is because of localization method and device, electronic equipment |
| CN109634819B (en) * | 2018-10-26 | 2022-02-01 | 创新先进技术有限公司 | Alarm root cause positioning method and device and electronic equipment |
| CN109547248A (en) * | 2018-11-20 | 2019-03-29 | 上海航天测控通信研究所 | Based on artificial intelligence in orbit aerocraft ad hoc network method for diagnosing faults and device |
| CN111327435B (en) * | 2018-12-13 | 2022-07-05 | 中兴通讯股份有限公司 | Root cause positioning method, server and storage medium |
| CN111327435A (en) * | 2018-12-13 | 2020-06-23 | 中兴通讯股份有限公司 | Root cause positioning method, server and storage medium |
| CN111628901A (en) * | 2019-02-28 | 2020-09-04 | 华为技术有限公司 | Index abnormality detection method and related device |
| CN111628901B (en) * | 2019-02-28 | 2022-11-18 | 华为技术有限公司 | Index anomaly detection method and related device |
| CN111698700A (en) * | 2019-03-15 | 2020-09-22 | 大唐移动通信设备有限公司 | Method and device for judging working state of cell |
| CN111698700B (en) * | 2019-03-15 | 2021-08-27 | 大唐移动通信设备有限公司 | Method and device for judging working state of cell |
| CN112543465A (en) * | 2019-09-23 | 2021-03-23 | 中兴通讯股份有限公司 | Abnormity detection method, abnormity detection device, terminal and storage medium |
| CN112543465B (en) * | 2019-09-23 | 2022-04-29 | 中兴通讯股份有限公司 | Abnormity detection method, abnormity detection device, terminal and storage medium |
| CN111064614A (en) * | 2019-12-17 | 2020-04-24 | 腾讯科技（深圳）有限公司 | Fault root cause positioning method, device, equipment and storage medium |
| CN111200530A (en) * | 2019-12-31 | 2020-05-26 | 畅捷通信息技术股份有限公司 | Method and device for performing root cause analysis based on KPI (Key performance indicator) |
| CN111405605A (en) * | 2020-03-24 | 2020-07-10 | 东南大学 | Wireless network interruption detection method based on self-organizing mapping |
| CN111935761A (en) * | 2020-07-15 | 2020-11-13 | 山东建筑大学 | Method and system for automatically positioning quality difference root cause of mobile communication network |
| CN111935761B (en) * | 2020-07-15 | 2024-04-09 | 山东建筑大学 | Automatic positioning method and system for quality difference root cause of mobile communication network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105873105B (en) | 2018-07-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105873105A (en) | Method for anomaly detection and positioning of mobile communication network based on network experience quality | |
| WO2019233189A1 (en) | Method for detecting sensor network abnormal data | |
| CN108322347B (en) | Data detection method, device, detection server and storage medium | |
| CN108366386B (en) | Method for realizing wireless network fault detection by using neural network | |
| Sugiarto et al. | Data classification for air quality on wireless sensor network monitoring system using decision tree algorithm | |
| CN106021771A (en) | Method and device for diagnosing faults | |
| CN105335752A (en) | Principal component analysis multivariable decision-making tree-based connection manner identification method | |
| CN110430224B (en) | Communication network abnormal behavior detection method based on random block model | |
| CN109273096A (en) | A kind of risk management grading evaluation method based on machine learning | |
| CN106330624B (en) | A kind of Power Information Network Traffic anomaly detection method | |
| CN107483487B (en) | TOPSIS-based multi-dimensional network security measurement method | |
| CN105893637A (en) | Link prediction method in large-scale microblog heterogeneous information network | |
| CN104268375A (en) | Ship electric power station fault diagnosing method based on knowledge petri network | |
| CN110188834A (en) | A kind of method for diagnosing faults of power telecom network, device and equipment | |
| CN108734359B (en) | Wind power prediction data preprocessing method | |
| CN112598199A (en) | Monitoring and early warning method based on decision tree algorithm | |
| CN109145951B (en) | Bayesian network-based tunnel lining structure service condition evaluation method | |
| CN103905276A (en) | Fast network topology detecting method | |
| CN105049286A (en) | Cloud platform speed measurement data judging method based on hierarchical clustering | |
| Chiang et al. | BTCI: A new framework for identifying congestion cascades using bus trajectory data | |
| CN116862081B (en) | Operation and maintenance method and system for pollution treatment equipment | |
| CN110084423A (en) | A kind of link prediction method based on local similarity | |
| CN107454089A (en) | A kind of network safety situation diagnostic method based on multinode relevance | |
| CN111405605B (en) | Wireless network interruption detection method based on self-organizing mapping | |
| CN110751201B (en) | SAR equipment task failure cause reasoning method based on textural feature transformation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |