CN107135126A - Flow on-line identification method based on subflow fractal index - Google Patents
Flow on-line identification method based on subflow fractal index Download PDFInfo
- Publication number
- CN107135126A CN107135126A CN201710362094.0A CN201710362094A CN107135126A CN 107135126 A CN107135126 A CN 107135126A CN 201710362094 A CN201710362094 A CN 201710362094A CN 107135126 A CN107135126 A CN 107135126A
- Authority
- CN
- China
- Prior art keywords
- subflow
- fractal index
- mrow
- fractal
- index
- 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
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0876—Network utilisation, e.g. volume of load or congestion level
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/16—Threshold monitoring
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L63/00—Network architectures or network communication protocols for network security
- H04L63/14—Network architectures or network communication protocols for network security for detecting or protecting against malicious traffic
- H04L63/1408—Network architectures or network communication protocols for network security for detecting or protecting against malicious traffic by monitoring network traffic
- H04L63/1425—Traffic logging, e.g. anomaly detection
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Environmental & Geological Engineering (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Complex Calculations (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention is applied to communication technique field there is provided a kind of flow on-line identification method based on subflow fractal index, and this method comprises the following steps：From point of penetration time t0Start, intercept subflow fragment and sampled, obtain subflow sequenceSubflow sequence is normalized；Estimate the fractal index of subflow；Calculate subflow fractal index and respectively have determined that the measures of dispersion of m random fractal index of discharge pattern；Threshold value T will be less than*Minimum difference amount corresponding have determined that discharge pattern as the discharge pattern for treating measurement of discharge.The embodiment of the present invention carries out the ONLINE RECOGNITION of discharge pattern by the fractal index collection for having determined that type in fractal index and system based on subflow, statistical nature need not be extracted, therefore the contradiction that accuracy and real-time can not get both is avoided, it also avoid lasting observed volume influences the unfavorable factor of real-time to obtain statistical nature data.
Description
Technical field
The invention belongs to the art field that communicates, more particularly to a kind of flow on-line identification method based on subflow fractal index.
Background technology
Accurately and effectively Classification and Identification is carried out to network traffics, the abnormal flow and attack stream in monitoring network can be helped
Amount, to ensure network security, Logistics networks even running；Operator is may also help in be implemented with the service of differentiation, realize flow
Charging, allows teenager away from the bad video data information such as yellow, violence, rubbish video flow on limitation network etc.；In addition,
Maximally utilizing, ensureing efficient end-to-end QoS for Internet resources, all be unable to do without the become more meticulous classification, therefore, net of network traffics
The Classification and Identification of network flow turns into the study hotspot of the communications field.
Researchers are started based on machine learning, the research of online classification technology are carried out using the method for feature extraction, this
The method of kind has outstanding performance in terms of accuracy rate, stability, flexibility, the research heat as current online classification technology
Point, and produce series of studies achievement, such as NB Algorithm, the pattra leaves based on association fast filtering mechanism, kernel estimates technology
This algorithm, decision Tree algorithms, algorithm of support vector machine, neural network algorithm and clustering algorithm etc., in this class method,
The selection of feature directly determines the accuracy and real-time of traffic classification method, in general, and increase feature can be obtained more preferably
Classifying quality, but computing cost and storage overhead exponential can also increase so that the real-time of classification is greatly reduced, because
This, it is difficult to take into account accuracy and real-time.
The content of the invention
The embodiment of the present invention provides a kind of flow on-line identification method based on subflow fractal index, it is intended to solve to be based on machine
Device learns, carries out the problem of online classification technology is difficult to take into account accuracy and real-time using the method for feature extraction.
The embodiments of the invention provide a kind of flow on-line identification method based on subflow fractal index, this method is included such as
Lower step：
S1, measurement of discharge X (t) is treated, from point of penetration time t0Start, intercept subflow fragment and sampled, obtain subflow
Sequence
S2, the subflow sequence is normalized；
S3, the fractal index D for estimating subflowH；
S4, calculating subflow fractal index DHWith the measures of dispersion for m random fractal index for respectively having determined that discharge pattern；
S5, threshold value T will be less than*Minimum difference amount corresponding have determined that discharge pattern as the flow for treating measurement of discharge
Type.
The embodiments of the invention provide another flow on-line identification method based on subflow fractal index, this method includes
Following steps：
S1, measurement of discharge X (t) is treated, from point of penetration time t0Start, intercept subflow fragment and sampled, obtain subflow
Sequence
Subflow sequence is normalized described in S2, antithetical phrase；
S3, the fractal index D for estimating subflowH；
S6, by the fractal index of subflow with having determined that the fractal index interval of discharge pattern is compared, obtain subflow
Discharge pattern is estimated, it is every kind of to have determined that discharge pattern all corresponds to a fractal index interval；
The measures of dispersion of the m random fractal index of S7, the fractal index for calculating subflow with estimating discharge pattern, it is described to estimate
Discharge pattern is that the two fractal index interval corresponding two nearest away from subflow fractal index has determined that discharge pattern, and subflow point shape refers to
The distance for counting to fractal index interval is subflow fractal index between the interval end value of fractal index for closing on subflow fractal index
Distance；
S8, threshold value T will be less than*The corresponding discharge pattern of estimating of minimum difference amount as the class of traffic for treating measurement of discharge
Type.
The embodiment of the present invention is entered by the fractal index collection for having determined that type in fractal index and system based on subflow
The ONLINE RECOGNITION of row discharge pattern, it is not necessary to extract statistical nature, therefore avoid the lance that accuracy and real-time can not get both
Shield, it also avoid lasting observed volume influences the unfavorable factor of real-time to obtain statistical nature data.
Brief description of the drawings
Fig. 1 implements the flow chart of the flow on-line identification method based on subflow fractal index of an offer for the present invention；
Fig. 2 implements the flow chart of the flow on-line identification method based on subflow fractal index of two offers for the present invention；
Fig. 3 is the flow chart of subflow fragment fractal index computational methods provided in an embodiment of the present invention.
Embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, it is right below in conjunction with drawings and Examples
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.
Fig. 1 is the flow chart for the flow on-line identification method based on subflow fractal index that the embodiment of the present invention one is provided,
This method comprises the following steps：
S1, measurement of discharge X (t) is treated, from point of penetration time t0Start, intercept subflow fragment and sampled, obtain subflow
Sequence
In embodiments of the present invention, complete flow is represented with X (t), and subflow is exactly to be intercepted at any time from X (t)
It is a bit of ----for example, from X (t0) start interception；Then sampled, obtain subflow sequence
Wherein, t0For the point of penetration time, N is number of samples, N concrete numerical value during system testing by training regulation to obtain,
There is floating for specific training dataset, value general N is 50, sampling interval 2ms.
S2, subflow sequence is normalized；
In embodiments of the present invention, in order to eliminate the amplitude difference influence that heterogeneous networks data band comes, to the subflow of acquisition
Sequence is normalized, and normalized processing formula is as follows：
S3, the fractal index D for estimating subflowH；
In embodiments of the present invention, fractal index is the parameter for quantitatively portraying fractal characteristic, using Hausdorff dimensions as
Example, definition：
DHThat is fractal index, N (r) represents quantity of the object to be measured under certain yardstick, with measurement scale diminish without
Disconnected increase, but fractal index DHKeep constant, this parameter can be used for the general characteristic of sign system, therefore use subflow fragment
Fractal index realize the online classification of flow.
According to formula (2), require that stream series infinite is long in theory, therefore, the method that can only use numerical analysis in reality
The estimate of fractal index is obtained, referring to fractal index estimation steps S31-S33.
S4, the measures of dispersion for calculating subflow fractal index and respectively having determined that m random fractal index of discharge pattern；
Because fractal index is obtained using numerical analysis method estimation, so even the flowmeter point counting shape to same type refers to
, always inevitably there is floatability in number.Therefore, random take the m bars determination discharge pattern fractal index, calculate subflow point shape and refer to
Total measures of dispersion, can so ensure classified calculating result between number and m random fractal index for respectively having determined that discharge pattern
It is more stable.
In embodiments of the present invention, the calculation formula of measures of dispersion is as follows：
Wherein, Δ H is measures of dispersion, DHFor the fractal index of subflow fragment, DHKTo have determined that the fractal index of discharge pattern
Collection, m refers to that fractal index concentrates the random number for obtaining fractal index.
In embodiments of the present invention, categorizing system uses cross-validation method in advance, that is, data set is divided into training
Collection and test set, training set data and test set data are all artificially to have carried out the mark of type in advance, anti-with training set data
Refreshment is practiced and regulating system parameters, including number of samples N, threshold value T*And m values, after system is stable, determined with test set
Classifying quality, in test process, Δ H measuress of dispersion just start to float very big, but with m increase, Δ H will gradually tend to be steady
Fixed, until the Δ H coefficient of variation is below 0.1, m values are determined.
S5, threshold value T will be less than*Minimum difference amount corresponding have determined that discharge pattern as the class of traffic for treating measurement of discharge
Type.
In embodiments of the present invention, if subflow fractal index has determined that m random fractal index of discharge pattern with some
Measures of dispersion be less than threshold value T*And be minimum, then this has determined that discharge pattern is to treat the discharge pattern of measurement of discharge, if the difference
Different amount is less than threshold value T*But and it is non-minimum, or measures of dispersion be more than or equal to threshold value T*, then this have determined that discharge pattern is not to treat flow measurement
The discharge pattern of amount；If the measures of dispersion of m random fractal index of the fractal index of some subflow with respectively having determined that type is big
In threshold value T*, then it is assumed that this treats that the discharge pattern of measurement of discharge is new, and the system can not be identified.
In embodiments of the present invention, threshold value T*Debugged using training dataset, when reaching highest recognition accuracy
When, threshold value T*It is determined, otherwise continues to adjust threshold value T*。
The embodiment of the present invention is entered by the fractal index collection for having determined that type in fractal index and system based on subflow
The ONLINE RECOGNITION of row discharge pattern, it is not necessary to extract statistical nature, therefore avoid the lance that accuracy and real-time can not get both
Shield, it also avoid lasting observed volume influences the unfavorable factor of real-time to obtain statistical nature data；
In addition, the object of data processing is subflow atomic in time scale, and the acquisition of subflow can be flow life
Any time in life cycle, therefore while classification accuracy is ensured, can greatly improve the real-time of online classification.
For each type of flow, its fractal index is floated in the range of some, according to the fractal index of subflow, roughly
Estimate which kind of flow type it may belong to, then calculate the fractal index of this subflow and estimate individual random point of the m of discharge pattern
The difference total amount of shape index, refers to without calculating the m random fractal of fractal index and all determination discharge patterns of this subflow
Several difference total amounts, reduces amount of calculation, improves the ageing of flow ONLINE RECOGNITION.
Fig. 2 is the flow chart for the flow on-line identification method based on subflow fractal index that the embodiment of the present invention two is provided,
Embodiment two is that step S4 and step S5 are optimized on the basis of embodiment one, after step s 3 including following step
Suddenly：
S6, by the fractal index of subflow with having determined that the fractal index interval of discharge pattern is compared, determine subflow
Estimate discharge pattern；
In embodiments of the present invention, it is every kind of to have determined that discharge pattern all corresponds to a fractal index interval, two points closed on
Shape interval index may have that small part region is overlapping in interface, estimate discharge pattern be away from subflow fractal index it is nearest two
Fractal index interval corresponding two has determined that discharge pattern, and subflow fractal index to the interval distance of fractal index is subflow point shape
The distance between the fractal index interval end value of index with closing on subflow fractal index；In embodiments of the present invention, it has been determined that stream
Amount type is generally comprised：GAME, QQ, Tudou, BT and FTP, the fractal index of each type of flow are floated in the range of some
Dynamic, such as GAME fractal index interval range is between 0.7852 to 0.7855, and QQ fractal index interval ranges are arrived 0.9134
Between 0.9136, Tudou fractal index interval ranges are between 0.8816 to 0.8819, and BT fractal index interval ranges exist
Between 0.8395 to 0.8415, FTP fractal index interval ranges are between 0.8689 to 0.8697.
The measures of dispersion of the m random fractal index of S7, the fractal index for calculating subflow with estimating discharge pattern；
In embodiments of the present invention, the calculation formula of measures of dispersion is as follows：
Wherein, Δ H is measures of dispersion, DHFor the fractal index of subflow fragment, DHKTo have determined that the fractal index of discharge pattern
Collection, m refers to that fractal index concentrates the random number for obtaining fractal index.
In embodiments of the present invention, categorizing system uses cross-validation method in advance, that is, data set is divided into training
Collection and test set, training set data and test set data are all artificially to have carried out the mark of type in advance, anti-with training set data
Refreshment is practiced and regulating system parameters, including number of samples N, threshold value T*And m values, after system is stable, determined with test set
Classifying quality, in test process, Δ H measuress of dispersion just start to float very big, but with m increase, Δ H will gradually tend to be steady
Fixed, until the Δ H coefficient of variation is below 0.1, m values are determined.
S8, threshold value T will be less than*The corresponding discharge pattern of estimating of minimum difference amount as the discharge pattern for treating measurement of discharge.
In embodiments of the present invention, if subflow fractal index estimates m random fractal index of discharge pattern with some
Measures of dispersion is less than threshold value T*And be minimum, then it is to treat the discharge pattern of measurement of discharge that this, which estimates discharge pattern, if the measures of dispersion
Less than threshold value T*But and it is non-minimum, or measures of dispersion be more than or equal to threshold value T*, then it is not to treat the stream of measurement of discharge that this, which estimates discharge pattern,
Measure type.
If it is 0.8818 to obtain certain subflow fractal index for treating measurement of discharge, discharge pattern is estimated for Tudou discharge patterns
And QQ discharge patterns, then calculate the m random fractal index of the subflow fractal index and Tudou discharge patterns and QQ discharge patterns
Measures of dispersion, if Tudou discharge patterns and the corresponding measures of dispersion of QQ discharge patterns are respectively less than threshold value T*, then it is assumed that minimum difference amount
Corresponding discharge pattern treats the discharge pattern of measurement of discharge for this, if only the corresponding measures of dispersion of Tudou discharge patterns is less than threshold value
T*, then the discharge pattern for treating measurement of discharge is Tudou discharge patterns, if measures of dispersion is all higher than being equal to threshold value T*, then it is assumed that estimate flow
Type is not the discharge pattern that this treats measurement of discharge；If certain treats that the subflow fractal index of measurement of discharge is 0.8515, class of traffic is estimated
Type is BT discharge patterns or ftp flow amount type, then calculates the m of the subflow fractal index and BT discharge patterns and ftp flow amount type
The measures of dispersion of individual random fractal index, if measures of dispersion is all higher than being equal to threshold value T*, then it is assumed that this treats the discharge pattern of measurement of discharge not
It is BT or FTP, if subflow fractal index is only less than threshold value T with the measures of dispersion of m random fractal index of BT discharge patterns*,
Then think that this treats that the discharge pattern of measurement of discharge is BT, if the m of subflow fractal index and BT discharge patterns and ftp flow amount type is individual
The measures of dispersion of random fractal index is respectively less than threshold value T*, and the former measures of dispersion is less than the latter's measures of dispersion, then it is assumed that this treats measurement of discharge
Discharge pattern be BT.
The embodiment of the present invention is entered by the fractal index collection for having determined that type in fractal index and system based on subflow
The ONLINE RECOGNITION of row discharge pattern, it is not necessary to extract statistical nature, therefore avoid the lance that accuracy and real-time can not get both
Shield, it also avoid lasting observed volume influences the unfavorable factor of real-time to obtain statistical nature data, in addition, the present invention is real
Example is applied by the fractal index for calculating subflow and the difference total amount for m random fractal index for estimating discharge pattern, without calculating
The fractal index of subflow and the difference total amount of m random fractal index of all determination discharge patterns, reduce amount of calculation, improve stream
Measure the ageing of ONLINE RECOGNITION.
Fig. 3 is the flow chart of subflow fragment fractal index computational methods provided in an embodiment of the present invention, and step S3 is specifically wrapped
Include following steps：
S31, DFT is carried out to subflow sequence, then by Paasche Wa Er Relation acquisition energy density spectrums；
In embodiments of the present invention, discrete Fourier transform is carried out to subflow sequence, calculation formula is as follows：
Further according to Paasche Wa Er Relation acquisition energy density spectrums, the calculation formula of energy density spectrum is as follows：
S (w)=| X (ejw)|2 (5)
S32, according to energy density spectrum S (w), calculate different SjIt is worth corresponding Aj, wherein SjIt is the random value of energy, j=
1,2 ... n, AjRefer to be more than energy value S on energy density spectral curvejPart and S=SjThe area surrounded；
In embodiments of the present invention, EDF S (w) represents the Energy distribution situation of different frequency, SjIt is energy
Random value, j=1,2 ... n, number of sampling n can be adjusted as the case may be, in test process, appoint in random take
Meaning height stream is calculated, and fractal index can be in metastable scope, i.e. the coefficient of variation is below 0.1, and n is to determine
Get off.
S33, to SjAnd AjTake the logarithm simultaneously, obtain double logarithmic curve, straight slope is obtained by LSM fitting a straight lines, it is taken
Negative value is fractal index DH。
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention
Any modifications, equivalent substitutions and improvements made within refreshing and principle etc., should be included in the scope of the protection.
Claims (6)
1. a kind of flow on-line identification method based on subflow fractal index, it is characterised in that methods described comprises the following steps：
S1, measurement of discharge X (t) is treated, from point of penetration time t0Start, intercept subflow fragment and sampled, obtain subflow sequence
S2, the subflow sequence is normalized；
S3, the fractal index D for estimating subflowH；
S4, calculating subflow fractal index DHWith the measures of dispersion for m random fractal index for respectively having determined that discharge pattern；
S5, threshold value T will be less than*Minimum difference amount corresponding have determined that discharge pattern as the discharge pattern for treating measurement of discharge.
2. the flow on-line identification method as claimed in claim 1 based on subflow fractal index, it is characterised in that the step
S3 specifically includes following steps：
S31, DFT is carried out to subflow sequence, then by Paasche Wa Er Relation acquisition energy density spectrum S (w)；
S32, according to the energy density spectrum S (w), calculate different SjIt is worth corresponding Aj, wherein the SjIt is that the random of energy takes
Value, j=1,2 ... n, the AjRefer to be more than energy value S on energy spectral density spectral curvejPart and S=SjThe area surrounded；
S33, to the SjWith the AjTake the logarithm simultaneously, obtain double logarithmic curve, straight slope is obtained by LSM fitting a straight lines,
It is fractal index D to take its negative valueH。
3. the flow on-line identification method as claimed in claim 1 based on subflow fractal index, it is characterised in that the difference
Amount is calculated using equation below：
<mrow>
<mi>&Delta;</mi>
<mi>H</mi>
<mo>=</mo>
<mo>&lsqb;</mo>
<mfrac>
<mn>1</mn>
<msup>
<mi>m</mi>
<mn>2</mn>
</msup>
</mfrac>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>K</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>m</mi>
</munderover>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>D</mi>
<mi>H</mi>
</msub>
<mo>-</mo>
<msub>
<mi>D</mi>
<mrow>
<mi>H</mi>
<mi>K</mi>
</mrow>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>&rsqb;</mo>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>3</mn>
<mo>)</mo>
</mrow>
</mrow>
Wherein, Δ H is measures of dispersion, DHFor the fractal index of subflow fragment, DHKTo have determined that the fractal index collection of discharge pattern, m
Refer to that fractal index concentrates the random number for obtaining fractal index.
4. a kind of flow on-line identification method based on subflow fractal index, it is characterised in that methods described comprises the following steps：
S1, measurement of discharge X (t) is treated, from point of penetration time t0Start, intercept subflow fragment and sampled, obtain subflow sequence
Subflow sequence is normalized described in S2, antithetical phrase；
S3, the fractal index D for estimating subflowH；
S6, by the fractal index of subflow with having determined that the fractal index interval of discharge pattern is compared, obtain subflow and estimate
Discharge pattern, it is every kind of to have determined that discharge pattern all corresponds to a fractal index interval；
The measures of dispersion of the m random fractal index of S7, the fractal index for calculating subflow with estimating discharge pattern, it is described to estimate flow
Type is that the two fractal index interval corresponding two nearest away from subflow fractal index has determined that discharge pattern, and subflow fractal index is arrived
The interval distance of fractal index arrived for subflow fractal index between the fractal index interval end value for closing on subflow fractal index away from
From；
S8, using the corresponding discharge pattern of estimating of the minimum difference amount less than threshold value T* as the discharge pattern for treating measurement of discharge.
5. the flow on-line identification method as claimed in claim 4 based on subflow fractal index, it is characterised in that the step
S3 specifically includes following steps：
S31, to subflow sequence carry out DFT, then by Paasche Wa Er Relation acquisitions energy spectral density compose S (w)；
S32, according to the energy density spectrum S (w), calculate different SjIt is worth corresponding Aj, wherein the SjIt is that the random of energy takes
Value, j=1,2 ... n, the AjRefer to be more than energy value S on energy density spectral curvejPart and S=SjThe area surrounded；
S33, to the SjWith the AjTake the logarithm simultaneously, obtain double logarithmic curve, straight slope is obtained by LSM fitting a straight lines,
It is fractal index D to take its negative valueH。
6. the flow on-line identification method as claimed in claim 4 based on subflow fractal index, it is characterised in that the difference
Amount is calculated using equation below：
<mrow>
<mi>&Delta;</mi>
<mi>H</mi>
<mo>=</mo>
<mo>&lsqb;</mo>
<mfrac>
<mn>1</mn>
<msup>
<mi>m</mi>
<mn>2</mn>
</msup>
</mfrac>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>K</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>m</mi>
</munderover>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>D</mi>
<mi>H</mi>
</msub>
<mo>-</mo>
<msub>
<mi>D</mi>
<mrow>
<mi>H</mi>
<mi>K</mi>
</mrow>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
<mo>&rsqb;</mo>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>3</mn>
<mo>)</mo>
</mrow>
</mrow>
Wherein, Δ H is measures of dispersion, DHFor the fractal index of subflow fragment, DHKTo have determined that the fractal index collection of discharge pattern, m
Refer to that fractal index concentrates the random number for obtaining fractal index.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710362094.0A CN107135126B (en) | 2017-05-22 | 2017-05-22 | Flow online identification method based on sub-flow fractal index |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710362094.0A CN107135126B (en) | 2017-05-22 | 2017-05-22 | Flow online identification method based on sub-flow fractal index |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107135126A true CN107135126A (en) | 2017-09-05 |
| CN107135126B CN107135126B (en) | 2020-03-24 |
Family
ID=59731832
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710362094.0A Expired - Fee Related CN107135126B (en) | 2017-05-22 | 2017-05-22 | Flow online identification method based on sub-flow fractal index |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107135126B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110784381A (en) * | 2019-11-05 | 2020-02-11 | 安徽师范大学 | Flow classification method based on particle calculation |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103747477A (en) * | 2014-01-15 | 2014-04-23 | 广州杰赛科技股份有限公司 | Network flow analysis and prediction method and device |
| CN105809190A (en) * | 2016-03-03 | 2016-07-27 | 南京邮电大学 | Characteristic selection based SVM cascade classifier method |
| CN106411829A (en) * | 2015-12-14 | 2017-02-15 | 中国民航大学 | LDoS attack detection method based on wavelet energy spectrum and combined neural network |
-
2017
- 2017-05-22 CN CN201710362094.0A patent/CN107135126B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103747477A (en) * | 2014-01-15 | 2014-04-23 | 广州杰赛科技股份有限公司 | Network flow analysis and prediction method and device |
| CN106411829A (en) * | 2015-12-14 | 2017-02-15 | 中国民航大学 | LDoS attack detection method based on wavelet energy spectrum and combined neural network |
| CN105809190A (en) * | 2016-03-03 | 2016-07-27 | 南京邮电大学 | Characteristic selection based SVM cascade classifier method |
Non-Patent Citations (2)
| Title |
|---|
| 常慧君，单洪，满毅: "基于分段、聚类和时序关联分析的用户行为分析_", 《计算机应用研究》 * |
| 李锐，朱洪亮，辛阳，王枞，杨义先: "基于分形理论P2P流量行为的自相似性", 《北京邮电大学学报》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110784381A (en) * | 2019-11-05 | 2020-02-11 | 安徽师范大学 | Flow classification method based on particle calculation |
| CN110784381B (en) * | 2019-11-05 | 2021-04-13 | 安徽师范大学 | Flow classification method based on particle calculation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107135126B (en) | 2020-03-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105425779B (en) | ICA-PCA multi-state method for diagnosing faults based on local neighborhood standardization and Bayesian inference | |
| CN106533750B (en) | The forecasting system and method for non-stationary type application user concurrent amount under a kind of cloud environment | |
| CN108932535B (en) | Edge computing clone node identification method based on machine learning | |
| CN108540451A (en) | A method of classification and Detection being carried out to attack with machine learning techniques | |
| CN107026763B (en) | A kind of data communication network method for predicting decomposed based on flow | |
| WO2014141344A1 (en) | Data prediction device | |
| CN109657613A (en) | Large scale electric network abnormal load recognition methods based on power method and parallel computing | |
| CN115580445B (en) | Unknown attack intrusion detection method, unknown attack intrusion detection device and computer readable storage medium | |
| CN108460486A (en) | A kind of voltage deviation prediction technique based on improvement clustering algorithm and neural network | |
| CN110119948A (en) | Based on when variable weight dynamic combined power consumer credit assessment method and system | |
| WO2021129464A1 (en) | Channel identification method and apparatus, transmission method, transmission device, base station, and medium | |
| CN111310833A (en) | Travel mode identification method based on Bayesian neural network | |
| CN113422695B (en) | Optimization method for improving robustness of topological structure of Internet of things | |
| CN113011888A (en) | Method, device, equipment and medium for detecting abnormal transaction behaviors of digital currency | |
| CN103778341B (en) | Method for estimating probabilistic load flow of harmonic pollution user | |
| CN106162720A (en) | A kind of cognition wireless self-organized network nodes stability assessment method based on multiple attribute decision making (MADM) | |
| CN110351303B (en) | DDoS feature extraction method and device | |
| CN107135126A (en) | Flow on-line identification method based on subflow fractal index | |
| CN103810482A (en) | Multi-information fusion classification and identification method | |
| CN106408571B (en) | A kind of variable class remote sensing image segmentation method based on the selection of optimal fuzzy factor | |
| CN112541634B (en) | Method and device for predicting top-layer oil temperature and discriminating false alarm and storage medium | |
| CN107247962A (en) | A kind of real-time electrical appliance recognition and system based on sliding window | |
| CN107274025B (en) | System and method for realizing intelligent identification and management of power consumption mode | |
| CN109039797A (en) | Big stream detection method based on intensified learning | |
| CN107770813A (en) | LTE uplink interference sorting techniques based on PCA Yu two-dimentional degree of bias feature |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20200324 Termination date: 20210522 |