CN102325342A - Cell interrupt detection and judgment method and device for self-recovery function of self-organization of network (SON) system - Google Patents
Cell interrupt detection and judgment method and device for self-recovery function of self-organization of network (SON) system Download PDFInfo
- Publication number
- CN102325342A CN102325342A CN201110336609A CN201110336609A CN102325342A CN 102325342 A CN102325342 A CN 102325342A CN 201110336609 A CN201110336609 A CN 201110336609A CN 201110336609 A CN201110336609 A CN 201110336609A CN 102325342 A CN102325342 A CN 102325342A
- Authority
- CN
- China
- Prior art keywords
- data
- network
- dimension
- interrupt
- cell
- 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
Images
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a cell interrupt detection and judgment method and a device for realizing interrupt detection and judgment. The device comprises a data acquirer and a data preprocessor, wherein the data acquirer collects the performance data of a network in real time; and the data preprocessor constructs high-dimensional observed data. A variance sigma of a Gaussian kernel function and a dimension number d=3 of target data are selected, dimension reduction is performed on high-dimensional data in a diffusion mapping way to obtain low-dimensional manifold characteristics Y={y1, y2, ..., yN} of the high-dimensional data, wherein yi belongs to Rd. The position of an interrupted cell, an interrupt type and factors for the interrupt are rapidly obtained by judgment criterions. By the scheme, real-time detection and judgment over the interrupted cell are relatively better realized, a network failure is ensured to be discovered timely, preparations are made for the rapid recovery and performance compensation of the interrupted cell, the capability of the network in coping with emergencies can be remarkably improved, and user experiences can be improved.
Description
Technical field
The present invention relates to wireless communication technology field, be specifically related to the detection and the judgment device of break detection arbitration schemes in sub-district in the SON system, realization such scheme.
Background technology
Wide-band mobile communication network IMT-Advanced of future generation adopts the flattening network configuration; Introduce Home eNodeB and relay station comprehensively; Advanced technologies such as the multiple-input and multiple-output (MIMO) of employing enhancement mode, multipoint cooperative (COMP), carrier aggregation; Thereby introduced a large amount of parameters and data processing, these will increase the complexity of network configuration optimization and cell outage judgement and location, the cost of labor of lifting network O&M.Under such background, the high-end operator of American-European main flow has proposed self-organization of network (SON) technology, hopes to reduce operation cost through SON, reduces O&M cost, improves operating efficiency, strengthens the performance and the stability of wireless access network.
The SON technology comprises multinomial functions such as planning, configuration, optimization, calculating, adjustment, test, guard against false, adjustment failure and self-recovery, to improve network performance, simplifies the wireless network design and reduces the network O&M cost.Through adopting advanced novel radio measuring technique and carrying out the network autonomous management, finally realize aspect of network self-configuration, self-optimizing and cure 3 big functions certainly that its main flow chart is as shown in Figure 1 from mounting strategy.
The SON system is as shown in Figure 2 from curing function, comprises that cell outage discovery, detection, judgement, location and single cell outage recover flow process, and multi-plot joint function such as restarts fast from curing and moving back the clothes sub-district.
For realizing SON from curing function, the first step is exactly through the type of the interruption of the sub-district of performance data detection generation interruption numerous and complicated in the network, judgement sub-district and the ID and the geographical position of interruption sub-district, location.Confirm further that according to the type of interrupting taking place the interruption degree is slight, the still interruption completely of moderate, the reason of interruption still is that transmission fault causes by software, hardware, electric power, and the cell recovery for next step provides strategy on this basis.
Yet the data extract Useful Information for magnanimity from network is not a nothing the matter, and traditional linear dimension reduction method such as principal component analysis have good dimensionality reduction effect for the high dimensional data that processing has linear relationship.But the inventor finds that various data degree of coupling are quite high in the highly developed modern network, is not simple linear relationship, but nonlinear relation.All data should be made full use of, Useful Information can be obtained fast again, just must utilization Nonlinear Dimension Reduction method.In recent years study awfully hot diffusion mapping dimension reduction method, can be good at non-linear high dimensional data is carried out effective dimensionality reduction the application of succeeding in a lot of fields.The Promethean SON of being introduced into of the inventor realizes finding fast, detecting and adjudicate the interruption of sub-district from curing in the functional module.
Diffusion dimensionality reduction thought is following:
To a given high dimensional data collection X={x
1, x
2..., x
N, x
i∈ R
D, (the stream shape characteristic Y of d<D), the performing step of diffusion mapping method is following therefrom to extract d dimension.
At first, according to given data point set X={x
1, x
2..., x
N, x
i∈ R
DSet up a corresponding with it figure.Use Gaussian function, the weights on limit among the definition figure obtain a weight matrix W, its element:
σ is the variance of gaussian kernel in the formula.σ is big more, and weights are big more.
Then through normalization technology, the every capable sum unit of matrix W is turned to 1, can obtain normalized weight matrix P
(1), its element is:
This matrix can be regarded the probability of corresponding data point random walk to other data points, then P as
(1)Be the step of one between any two data points transition probability matrix.Through t step migration, corresponding transition probability matrix is P
(t)=(P
(1))
i, 2 x
iAnd x
jBetween t step after diffusion length be defined as:
In the formula
described bigger this attribute of weights in figure middle-high density district.Can draw from following formula, the figure mid point is intensive more, and the diffusion length between the paired data point is more little.Diffusion length has considered to connect the contribution on 2 all limits, so stronger than some other Nonlinear Dimension Reduction method robustness.
Final step under the condition that keeps diffusion length, is extracted low dimension stream shape Y.The spectrogram on road is theoretical at random according to Markov, and Y is made up of the main characteristic vector of d non-trivial of following formula
P
iY=λY
Owing to be full connection layout, eigenvalue of maximum (is λ
1=1) be ordinary, its characteristic of correspondence vector v
1Should give up.Low dimension stream shape Y is provided by remaining d main characteristic vector.
Y={λ
2v
2,λ
3v
3,…λ
d+1v
d+1}
Summary of the invention
The invention provides a kind of SON system from curing sub-district break detection and arbitration schemes in the function; And fully excavating on the basis of the network information; Confirming of type judgement and geographical position and sub-district ID carried out in the interruption sub-district that well network is occurred; Guaranteed the timely discovery of network failure and the fast quick-recovery of interruption estate performance, high-quality network service is provided to the user.
The present invention utilized and used very successful diffusion mapping dimension reduction method in recent years in other field, made full use of the computational complexity that the mass data in the network can extremely reduce network again.
Be illustrated in figure 3 as SON of the present invention system from curing sub-district break detection in the function, judgement and positioning flow figure, its practical implementation process is following:
Stage one:
Step 1, network alarm trigger, and get into cell outage and detect inlet;
Step 2, parameter acquisition, with SNR, RSRP disturbs in the sub-district, the eNodeB presence of intercell interference, BLER, RLF, transmitting power, CQI, the access failure rate, handover failure rate, and wireless performance parameter such as cutting off rate is tieed up image data as D;
Step 3, data preliminary treatment utilize interpolation method that missing values is carried out polishing.Set that importance weight is unified various attributes and to data point each property value normalization handle.
Step 4, structure higher-dimension observation data X={x
1, x
2..., x
N, x
i∈ R
D
Stage two:
The variances sigma of step 1, selection gaussian kernel function, and the dimension d of target data, the present invention is taken as 3;
Step 2, higher-dimension observation data obtain low dimension stream shape characteristic Y={ y through the diffusion mapping
1, y
2..., y
N, y
i∈ R
d
Step 3, to resultant low dimension stream shape characteristic, classifying at lower dimensional space has following two kinds of scheme criterions.
Criterion one:
Each was half the before and after the tentation data sampled point came from the interruption early warning, if only there is single zone in stream shape feature space, judged that then the base station is in proper working order.If then judging the base station, plural zone breaks down.
Criterion two:
The tentation data sampled point all comes from the later performance data of base station early warning, and the statistic parameter of the data characteristics of low dimension has: average, standard deviation, peak value, root-mean-square value, peak-to-peak value.With the average is example, sets threshold value μ={ μ
1, μ
2, μ
3(according to the average of data flow shape characteristic before the early warning), then three-dimensional data can be divided into 8 types with it by the monodrome thresholding.Do not take place to interrupt this situation if remove, can interrupt type be divided into 7 kinds at the most, the cell outage rate is very low usually, and seven types enough, can some interrupt type be merged according to real network performance recovery demand.
According to such scheme; The present invention proposes a kind of cell outage based on diffusion mapping dimensionality reduction and detect and judgment device, said break detection device comprises: parameter acquisition module, parameter pretreatment module; Diffusion mapping dimensionality reduction module, break detection and type judging module.The parameter acquisition module is obtained the performance parameter related with cell outage from network; Data preprocessing module, the data that network is directly obtained are carried out processed and are obtained higher-dimension observation data collection; The diffusion mapping block, the low dimension of obtaining performance parameter flows the shape characteristic; Detect and judging module, confirm the generation and the interrupt type that interrupt.
This scheme can make full use of network performance parameter, and the real time monitoring network operation conditions can fast detecting go out the cell outage that happens suddenly in the network, and carries out the fault judgement, for the fast cell performance recovery is got ready.
Description of drawings
Fig. 1 is SON three big functional schematics.
Fig. 2 cures flow chart certainly for the SON cell outage.
Fig. 3 for SON of the present invention system from curing sub-district break detection in the function, judgement and positioning flow figure.
Fig. 4 is SON system cell break detection of the present invention and judgment device sketch map.
Claims (7)
1. a method of interrupting cell detection and judgement is characterized in that, comprising:
Confirm the object of parameter acquisition and the higher-dimension observation sample of structure;
Based on the higher-dimension observation sample, select the dimension of gaussian kernel function and target data, confirm the diffusion mapping algorithm;
Confirm the position and the type of cell outage through decision rule.
2. according to the said method of claim 1, it is characterized in that, from all network performance parameter set, choose D, make up higher-dimension observation data collection: X={x
1, x
2..., x
N, x
i∈ R
D
3. method according to claim 1 is characterized in that, selects suitable gaussian kernel function variances sigma, and the dimension d=3 of low dimension target data.
4. method according to claim 2 is characterized in that carrying out preliminary treatment for the network measure data, and data are carried out the completion missing value, and the unification of data attribute and normalization are handled.
5. method according to claim 3 is characterized in that, obtains the low dimension stream shape characteristic of data through the diffusion mapping algorithm.
6. method according to claim 1 is characterized in that, confirms to have or not in the network cell outage, the position of the sub-district of interrupt type and interruption according to following two decision rules.
Criterion one:
Each was half the before and after the tentation data sampled point came from the interruption early warning, if only there is single zone in stream shape feature space, judged that then the base station is in proper working order.If then judging the base station, plural zone breaks down;
Criterion two:
The tentation data sampled point all comes from the later performance data of base station early warning, sets threshold value μ={ μ with average
1, μ
2, μ
3, then three-dimensional data can be divided into 8 types with it by the monodrome thresholding.Do not take place to interrupt this situation if remove, can interrupt type be divided into 7 kinds at the most, the cell outage rate is very low usually, and seven types enough, can some interrupt type be merged according to real network performance recovery demand.
7. a SON is used for detecting the interruption sub-district that network occurs in real time from the detection of the interruption sub-district that cures and the device of judgement, and said checkout gear comprises:
The parameter acquisition module is obtained the performance parameter related with cell outage from network;
Data preprocessing module, the data that network is directly obtained are carried out processed and are obtained higher-dimension observation data collection;
The diffusion mapping block, the low dimension of obtaining performance parameter flows the shape characteristic;
Detect and judging module, confirm the generation and the interrupt type that interrupt.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201110336609A CN102325342A (en) | 2011-10-31 | 2011-10-31 | Cell interrupt detection and judgment method and device for self-recovery function of self-organization of network (SON) system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201110336609A CN102325342A (en) | 2011-10-31 | 2011-10-31 | Cell interrupt detection and judgment method and device for self-recovery function of self-organization of network (SON) system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN102325342A true CN102325342A (en) | 2012-01-18 |
Family
ID=45452996
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201110336609A Pending CN102325342A (en) | 2011-10-31 | 2011-10-31 | Cell interrupt detection and judgment method and device for self-recovery function of self-organization of network (SON) system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102325342A (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102843709A (en) * | 2012-05-30 | 2012-12-26 | 北京邮电大学 | Interruption compensation method and device for SON (self-organized network) systems |
CN102932826A (en) * | 2012-11-30 | 2013-02-13 | 北京邮电大学 | Cell interruption detection positioning method in self-organizing network of cell mobile communication system |
WO2013107035A1 (en) * | 2012-01-20 | 2013-07-25 | 富士通株式会社 | Method and device for detecting resumption of interrupted cell |
CN103269494A (en) * | 2013-04-24 | 2013-08-28 | 北京邮电大学 | Method and system for interruption compensating in wireless access network cell |
CN104782157A (en) * | 2012-11-01 | 2015-07-15 | 英特尔公司 | System and method of cell outage compensation in cellular systems |
CN105101265A (en) * | 2015-07-03 | 2015-11-25 | 北京邮电大学 | Cooperative compensation service method for solving cell service interruption |
CN105188080A (en) * | 2015-08-05 | 2015-12-23 | 东南大学 | Cell interruption detection method based on direct push confidence machine and hypothesis testing in mobile communication network |
CN105988885A (en) * | 2015-03-26 | 2016-10-05 | 朱怡安 | Compensation rollback-based operation system fault self-recovery method |
CN110309492A (en) * | 2019-06-29 | 2019-10-08 | 河北工业大学 | Wind power generating set health degree appraisal procedure based on scatter diagram Data Dimensionality Reduction technology |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101984697A (en) * | 2010-10-19 | 2011-03-09 | 中兴通讯股份有限公司 | Wireless data service troubleshooting method and system |
CN102111797A (en) * | 2011-02-15 | 2011-06-29 | 大唐移动通信设备有限公司 | Fault diagnosis method and fault diagnosis equipment |
-
2011
- 2011-10-31 CN CN201110336609A patent/CN102325342A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101984697A (en) * | 2010-10-19 | 2011-03-09 | 中兴通讯股份有限公司 | Wireless data service troubleshooting method and system |
CN102111797A (en) * | 2011-02-15 | 2011-06-29 | 大唐移动通信设备有限公司 | Fault diagnosis method and fault diagnosis equipment |
Non-Patent Citations (1)
Title |
---|
尚晓清: "一种基于扩散映射的非线性降维算法", 《西安电子科技大学学报(自然科学版)》 * |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2013107035A1 (en) * | 2012-01-20 | 2013-07-25 | 富士通株式会社 | Method and device for detecting resumption of interrupted cell |
CN102843709A (en) * | 2012-05-30 | 2012-12-26 | 北京邮电大学 | Interruption compensation method and device for SON (self-organized network) systems |
CN104782157A (en) * | 2012-11-01 | 2015-07-15 | 英特尔公司 | System and method of cell outage compensation in cellular systems |
CN102932826A (en) * | 2012-11-30 | 2013-02-13 | 北京邮电大学 | Cell interruption detection positioning method in self-organizing network of cell mobile communication system |
WO2014082437A1 (en) * | 2012-11-30 | 2014-06-05 | 北京邮电大学 | Method for detecting and positioning cell interruption in cellular mobile communication system self-organizing network |
CN102932826B (en) * | 2012-11-30 | 2015-01-14 | 北京邮电大学 | Cell interruption detection positioning method in self-organizing network of cell mobile communication system |
US9426771B2 (en) | 2012-11-30 | 2016-08-23 | Beijing University Of Posts And Telecommunications | Method for detecting cell disconnection and locating disconnected cell in SON of cellular mobile communication system |
CN103269494B (en) * | 2013-04-24 | 2016-04-06 | 北京邮电大学 | Radio Access Network cell interrupt compensation method and system |
CN103269494A (en) * | 2013-04-24 | 2013-08-28 | 北京邮电大学 | Method and system for interruption compensating in wireless access network cell |
CN105988885B (en) * | 2015-03-26 | 2019-01-29 | 朱怡安 | Operating system failure self-recovery method based on compensation rollback |
CN105988885A (en) * | 2015-03-26 | 2016-10-05 | 朱怡安 | Compensation rollback-based operation system fault self-recovery method |
CN105101265A (en) * | 2015-07-03 | 2015-11-25 | 北京邮电大学 | Cooperative compensation service method for solving cell service interruption |
CN105101265B (en) * | 2015-07-03 | 2020-05-22 | 北京邮电大学 | Cooperative compensation service method for solving cell service interruption |
CN105188080A (en) * | 2015-08-05 | 2015-12-23 | 东南大学 | Cell interruption detection method based on direct push confidence machine and hypothesis testing in mobile communication network |
CN105188080B (en) * | 2015-08-05 | 2019-04-23 | 东南大学 | Cell outage detection method in mobile communications network based on direct-push confidence machine and hypothesis testing |
CN110309492A (en) * | 2019-06-29 | 2019-10-08 | 河北工业大学 | Wind power generating set health degree appraisal procedure based on scatter diagram Data Dimensionality Reduction technology |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102325342A (en) | Cell interrupt detection and judgment method and device for self-recovery function of self-organization of network (SON) system | |
CN102932826B (en) | Cell interruption detection positioning method in self-organizing network of cell mobile communication system | |
CN116633978B (en) | Sewage treatment remote monitoring system based on Internet of Things | |
CN103200597B (en) | Wireless sensor network-oriented fault-tolerance event boundary detection method | |
CN109299698B (en) | Wireless channel scene recognition method based on support vector machine | |
CN107391369A (en) | A kind of spanned item mesh failure prediction method based on data screening and data oversampling | |
CN110830946B (en) | Mixed type online data anomaly detection method | |
Zhe et al. | DoS attack detection model of smart grid based on machine learning method | |
CN102594904A (en) | Method for detecting abnormal events of wireless sensor network in distributed way | |
CN108337645B (en) | Gridding radio signal monitoring system architecture and interference signal joint detection method | |
CN112243249A (en) | LTE new access anchor point cell parameter configuration method and device under 5G NSA networking | |
CN101257355A (en) | Method for detecting wireless sensor network node malfunction | |
CN112367400A (en) | Intelligent inspection method and system for power internet of things with edge cloud coordination | |
CN103729444A (en) | Abnormal data detection method based on potential relation among equipment monitoring data | |
Yu et al. | Self-organized cell outage detection architecture and approach for 5G H-CRAN | |
CN105632108A (en) | GPRS and ZigBee network-based debris flow monitoring and early warning system | |
CN103401626B (en) | Based on the collaborative spectrum sensing optimization method of genetic algorithm | |
CN105429913A (en) | Multi-level detection and identification method based on characteristic value | |
CN111405605B (en) | Wireless network interruption detection method based on self-organizing mapping | |
Wang et al. | Extracting cell patterns from high-dimensional radio network performance datasets using self-organizing maps and K-means clustering | |
CN104158604B (en) | A kind of distributed collaborative frequency spectrum sensing method based on average common recognition | |
CN113949163A (en) | Anomaly monitoring system applying big data to smart power grid | |
CN109167642B (en) | Low-overhead real-time spectrogram construction method based on historical decision data learning | |
CN113747441A (en) | Mobile network flow abnormity detection method and system based on feature dimension reduction | |
Zhang et al. | Clustering and decision tree based analysis of typical operation modes of power systems |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20120118 |