CN107360046A - The modeling method and system of end to end network flow based on bayesian theory - Google Patents
The modeling method and system of end to end network flow based on bayesian theory Download PDFInfo
- Publication number
- CN107360046A CN107360046A CN201710813389.5A CN201710813389A CN107360046A CN 107360046 A CN107360046 A CN 107360046A CN 201710813389 A CN201710813389 A CN 201710813389A CN 107360046 A CN107360046 A CN 107360046A
- Authority
- CN
- China
- Prior art keywords
- time slots
- network flow
- end network
- current time
- continuous multiple
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/147—Network analysis or design for predicting network behaviour
-
- 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
Abstract
This application provides a kind of modeling method of the end to end network flow based on bayesian theory, including:In network topology structure, the initial value of the end to end network flow at default continuous multiple time slots is obtained;By the random process that the end to end network traffic transformation at default continuous multiple time slots is Normal Distribution, the parameter and probability of the end to end network flow at each time slot are obtained;Based on bayesian theory, according to the end to end network discharge model at the parameter of the end to end network flow at each time slot before current time slots in default continuous multiple time slots and probability structure current time slots;Definition Model departure function, the estimate of the end to end network flow at current time slots and the deviation of actual value are minimized, and according to the optimal models of the end to end network flow at the end to end network discharge model structure current time slots at the model bias function and the current time slots.
Description
Technical field
The present invention relates to large-scale network environment lower network to predict field, more particularly to a kind of end based on bayesian theory
To the modeling method and system of end network traffics.
Background technology
With the fast development of network technology and new opplication, network traffics illustrate new characteristic.This is network engineering band
Challenge is newly carried out.Characterize the characteristic of network traffics exactly and to improve network performance be extremely important to Modeling Network Traffic
's.The feature of network traffics, such as self-similarity, auto-correlation, heavytailed distribution etc., there is important shadow to the network optimization and routing optimality
Ring.End to end network flow is represented from the network range behavior from the point of view of global angle.Therefore, the modeling of end to end network flow has been
It is subjected to the extensive concern from researcher all over the world, operator and developer.
End-to-end flux behavior includes source node and destination node, embodies the path-level feature in network, can be used for
Network state and property, such as path load, handling capacity, network utilization etc. are described.Represented in the prior art using statistical method
Model of network traffic from source node to destination node, utilize Gravity Models, general evolution method, mixed method and compressed sensing
Method captures the attribute of end to end network flow.These methods can be carried out more by performing modeling process to end-to-end flux
Good prediction and estimation.However, these methods need to obtain additional information from link load or on end to end network stream
The prior information of amount, add the complexity and expense of model parameter calculation.And other methods, e.g., Time-Frequency Analysis and god
Through network, model can be built to represent end to end network flow, but be difficult accurate capture and obtain end to end network flow
Feature, it is impossible to for traffic engineering structure Model of network traffic accurately and suitably.
The content of the invention
In view of this, the invention provides a kind of modeling method of end to end network flow based on bayesian theory and it is
System, the network traffics of current time slots are represented using the network traffics before current time slots, accurate expression end to end network flow
Feature.
In order to realize foregoing invention purpose, concrete technical scheme provided by the invention is as follows:
A kind of modeling method of the end to end network flow based on bayesian theory, including:
In network topology structure, the initial value of the end to end network flow at default continuous multiple time slots is obtained;
Random process by the end to end network traffic transformation at default continuous multiple time slots for Normal Distribution,
Obtain the parameter and probability of the end to end network flow at each time slot;
Based on bayesian theory, at each time slot before current time slots in default continuous multiple time slots
End to end network flow parameter and probability build end to end network discharge model at the current time slots, when described current
Gap is last time slot in default continuous multiple time slots;
Definition Model departure function, make end to end network flow at the current time slots estimate and actual value it is inclined
Difference minimize, and according at the model bias function and the current time slots end to end network discharge model structure it is current when
The optimal models of end to end network flow at gap.
Preferably, the end to end network traffic transformation by default continuous multiple time slots is Normal Distribution
Random process, obtain the parameter and probability of the end to end network flow at each time slot, including:
The initial value of end to end network flow at default continuous multiple time slots is converted into Normal Distribution
Random process, obtain the probability of the end to end network flow at each time slot;
Normal state point is obeyed by the way that the initial value of the end to end network flow at default continuous multiple time slots is converted into
The random process of cloth, utilize the end-to-end net at each time slot before current time slots in default continuous multiple time slots
Network flow represents the end to end network flow at current time slots, obtains the ginseng of the end to end network flow at each time slot
Number, the current time slots are last time slot in default continuous multiple time slots.
Preferably, the Definition Model departure function, the estimate of the end to end network flow at the current time slots is made
And the deviation of actual value minimizes, and according to the end to end network flow mould at the model bias function and the current time slots
The optimal models of end to end network flow at type structure current time slots, including:
According to the estimate of the end to end network flow at current time slots and the deviation definition model bias function of actual value;
The optimal models of the end to end network flow at current time slots is built, the optimal models is with the current time slots
The estimate of end to end network flow and the deviation of actual value at place are minimised as object function, and including the model bias letter
The probability letter of the end to end network flow at end to end network discharge model and the current time slots at several, described current time slots
Several constraint single object optimization models.
Preferably, default continuous multiple time slots are continuous 4 time slots.
A kind of modeling of the end to end network flow based on bayesian theory, including:
Acquiring unit, in network topology structure, obtaining the end to end network flow at default continuous multiple time slots
Initial value;
Unit is represented, for the end to end network flow at default continuous multiple time slots to be expressed as obeying normal state point
The random process of cloth, obtain the parameter and probability of the end to end network flow at each time slot;
Construction unit, for based on bayesian theory, being preset according to described in continuous multiple time slots before current time slots
The parameter and probability of end to end network flow at each time slot build the end to end network flow at the current time slots
Model, the current time slots are last time slot in default continuous multiple time slots;
Definition unit, for Definition Model departure function, make the estimation of the end to end network flow at the current time slots
Value and the deviation of actual value minimize, and according to the end to end network flow at the model bias function and the current time slots
The optimal models of end to end network flow at model construction current time slots.
Preferably, the conversion unit includes:
Transforming subunit, for the initial value of the end to end network flow at default continuous multiple time slots to be converted into
The random process of Normal Distribution, obtain the probability of the end to end network flow at each time slot;
Subelement is represented, for by the way that the initial value of the end to end network flow at default continuous multiple time slots is turned
The random process of Normal Distribution is turned to, during using each described before current time slots in default continuous multiple time slots
End to end network flow at gap represents the end to end network flow at current time slots, obtains end-to-end at each time slot
The parameter of network traffics, the current time slots are last time slot in default continuous multiple time slots.
Preferably, the definition unit includes:
Subelement is defined, for determining according to the estimate of end to end network flow at current time slots and the deviation of actual value
Adopted model bias function;
Subelement is built, for building the optimal models of the end to end network flow at current time slots, the optimal models
To be minimised as object function with the estimate of end to end network flow at the current time slots and the deviation of actual value, and wrap
Include end-to-end at the model bias function, the end to end network discharge model at the current time slots and the current time slots
The constraint single object optimization model of the probability function of network traffics.
Preferably, default continuous multiple time slots are continuous 4 time slots.
It is as follows relative to prior art, beneficial effects of the present invention:
The modeling method and system of a kind of end to end network flow based on bayesian theory provided by the invention, first,
After the initial value for obtaining the end to end network flow at default continuous multiple time slots, at default continuous multiple time slots
End to end network traffic transformation is the random process of Normal Distribution, obtains the end to end network flow at each time slot
Parameter and probability, the statistical property of accurate description end to end network flow.Then, based on bayesian theory, according to described pre-
If the parameter and probability structure of the end to end network flow at each time slot in continuous multiple time slots before current time slots
End to end network discharge model at the current time slots, and by Definition Model departure function, make at the current time slots
The estimate of end to end network flow and the deviation of actual value minimize, and finally obtain the end to end network flow at current time slots
Optimal models, the correct optimal models for building end to end network flow, the accurate feature for obtaining end to end network flow.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is the required accompanying drawing used in technology description to be briefly described, it should be apparent that, drawings in the following description are only this
The embodiment of invention, for those of ordinary skill in the art, on the premise of not paying creative work, can also basis
The accompanying drawing of offer obtains other accompanying drawings.
Fig. 1 is a kind of modeling method stream of the end to end network flow based on bayesian theory disclosed in the embodiment of the present invention
Cheng Tu;
Fig. 2 is a kind of modeling of the end to end network flow based on bayesian theory disclosed in the embodiment of the present invention
Structural representation.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, rather than whole embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained every other under the premise of creative work is not made
Embodiment, belong to the scope of protection of the invention.
Present embodiment discloses a kind of modeling method of the end to end network flow based on bayesian theory, referring to Fig. 1,
Fig. 1 is the modeling method flow chart of the end to end network flow, is specifically included:
S101:In network topology structure, the initial value of the end to end network flow at default continuous multiple time slots is obtained;
The initial value of end to end network flow at default continuous multiple time slots can be expressed as:X={ x (1), x
(2) ... }, wherein, x (i) represent i time slots at end to end network flow x value, i=1,2 ....
S102:It is the random of Normal Distribution by the end to end network traffic transformation at default continuous multiple time slots
Process, obtain the parameter and probability of the end to end network flow at each time slot;
The initial value of end to end network flow at default continuous multiple time slots is converted into Normal Distribution
Random process, obtain the probability of the end to end network flow at each time slot;
Specifically, assume that end-to-end flux follows independent identically distributed normal stochastic process, then it is described default continuous multiple
The initial value of end to end network flow at time slot is converted into the random process X of Normal Distribution, then end-to-end flux X can table
It is shown as:
X~N (μ, δ) (1)
Wherein, μ and δ represents the average and variance of normal distribution respectively.Therefore, the end to end network at each time slot
The probability of flow can be indicated with below equation:
Wherein, probability during P (X=x) expressions end to end network flow X value x;μ parameters and δ describe end-to-end flux
Characteristic.However, equation (1) and (2) only represent end-to-end flux X statistical property.
Normal state point is obeyed by the way that the initial value of the end to end network flow at default continuous multiple time slots is converted into
The random process of cloth, utilize the end-to-end net at each time slot before current time slots in default continuous multiple time slots
Network flow represents the end to end network flow at current time slots, obtains the ginseng of the end to end network flow at each time slot
Number, the current time slots are last time slot in default continuous multiple time slots.
Specifically, using end-to-end at each time slot before current time slots in default continuous multiple time slots
Network traffics represent the end to end network flow x (k) at current time slots:
X (k)=a1x(k-1)+a2x(k-2)+...+anx(k-n)+ε (3)
Wherein, aiFor parameter, i=1,2 ..., n;ε is the normal stochastic process for obeying N (α, β), and in expression (3)
The deviation of model;The quantity of time slot before n expression current time slots.
Formula (3) shows to give end-to-end network traffics x (k-1), x (k-2) ..., x (k-n), in time slot k-1, k-
2 ..., k-n ε deviation profiles, we can obtain end-to-end flux x (k) in k time slots.
S103:Based on bayesian theory, according to each before current time slots in default continuous multiple time slots
The parameter and probability of end to end network flow at time slot build the end to end network discharge model at the current time slots, described
Current time slots are last time slot in default continuous multiple time slots;
Proper network data have certain similitude, and data below can be estimated with above one piece of data, according to
End-to-end network traffics are given in k-1, k-2 ..., k-n time slots and deviation profile, we are estimated that obtains end in k time slots
To end flow.In each time slot, end to end network flow is considered random process.Therefore, built according to bayesian theory
Formwork erection type, equation (3) can be expressed as:
Experiment shows, as n=3, when can accurately capture the characteristic of end to end network flow.I.e., it is preferred that described pre-
If continuous multiple time slots are continuous 4 time slots.It is described in this case, equation (3) and (4) can be expressed simply as:
X (k)=a1x(k-1)+a2x(k-2)+a3x(k-3)+ε (5)
The characteristic of end to end network flow can be accurately captured by bayesian theory.
S104:Definition Model departure function, make the estimate of end to end network flow at the current time slots with it is true
The deviation of value minimizes, and is built according to the end to end network discharge model at the model bias function and the current time slots
The optimal models of end to end network flow at current time slots.
According to the estimate of the end to end network flow at current time slots and the deviation definition model bias function of actual value;
Definition Model departure function:
Wherein,The end to end network flow estimation value obtained according to formula (5) is represented, Δ x (k) represents actual value x (k)
With estimateDeviation, and Δ x (k) is and parameter a1,a2,a3, α, β, random process related μ.Therefore, obtain following
Equation:
E(Δx(k)|a1,a2,a3, α, β, μ, δ) and=∫ ∫ Δ x (k) p (x, ε) dxd ε (8)
Wherein, p (x, ε) represents the joint probability density function of normal stochastic process, in general, essence is calculated in order to improve
Degree, it is desirable to which formula (8) is equal to zero.
The optimal models of the end to end network flow at current time slots is built, the optimal models is with the current time slots
The estimate of end to end network flow and the deviation of actual value at place are minimised as object function, and including the model bias letter
The probability letter of the end to end network flow at end to end network discharge model and the current time slots at several, described current time slots
Several constraint single object optimization models.
Due to the time-varying characteristics and correlation of end to end network flow, some deviations are constantly present in formula (8).Only when
During deviation minimum, the optimal models of end to end network flow could be obtained:
Formula (9) represents multiple constraint single-object problem, wherein first function representation object function, it makes equation (5)
Middle model bias minimizes;Second function representation obtains time slot k end to end network flow estimation value according to formula (5);3rd
Individual function stand network traffics X distribution;The estimated bias of 4th function representation model.
The model in equation (9) is trained and solved using sample data, and we can correctly obtain model parameter.At this
In the case of kind, we can correctly describe end to end network flow.
A kind of modeling method for end to end network flow based on bayesian theory that the present embodiment provides, first, is being obtained
After the initial value for taking the end to end network flow at default continuous multiple time slots, the end at default continuous multiple time slots is arrived
End network traffics are converted into the random process of Normal Distribution, obtain the ginseng of the end to end network flow at each time slot
Number and probability, the statistical property of accurate description end to end network flow.Then, based on bayesian theory, according to the default company
Continue described in parameter and the probability structure of the end to end network flow at each time slot in multiple time slots before current time slots
End to end network discharge model at current time slots, and by Definition Model departure function, arrive the end at the current time slots
Hold the estimate of network traffics and the deviation of actual value to minimize, finally obtain end to end network flow at current time slots most
Excellent model, the correct optimal models for building end to end network flow, the accurate feature for obtaining end to end network flow.
Based on a kind of modeling method of the end to end network flow based on bayesian theory, this reality disclosed in above-described embodiment
Apply example and correspondingly disclose a kind of modeling of the end to end network flow based on bayesian theory, referring to Fig. 2, the end is arrived
The modeling of end network traffics includes:
Acquiring unit 101, in network topology structure, obtaining the end to end network stream at default continuous multiple time slots
The initial value of amount;
Preferably, default continuous multiple time slots are continuous 4 time slots.
Unit 102 is represented, for the end to end network flow at default continuous multiple time slots to be expressed as obeying just
The random process of state distribution, obtains the parameter and probability of the end to end network flow at each time slot;
Construction unit 103, for based on bayesian theory, being preset according to described in continuous multiple time slots before current time slots
Each time slot at end to end network flow parameter and probability build end to end network stream at the current time slots
Model is measured, the current time slots are last time slot in default continuous multiple time slots;
Definition unit 104, for Definition Model departure function, make estimating for the end to end network flow at the current time slots
The deviation of evaluation and actual value minimizes, and according to the end to end network stream at the model bias function and the current time slots
Measure the optimal models of the end to end network flow at model construction current time slots.
Preferably, the conversion unit 102 includes:
Transforming subunit, for the initial value of the end to end network flow at default continuous multiple time slots to be converted into
The random process of Normal Distribution, obtain the probability of the end to end network flow at each time slot;
Subelement is represented, for by the way that the initial value of the end to end network flow at default continuous multiple time slots is turned
The random process of Normal Distribution is turned to, during using each described before current time slots in default continuous multiple time slots
End to end network flow at gap represents the end to end network flow at current time slots, obtains end-to-end at each time slot
The parameter of network traffics, the current time slots are last time slot in default continuous multiple time slots.
Preferably, the definition unit 104 includes:
Subelement is defined, for determining according to the estimate of end to end network flow at current time slots and the deviation of actual value
Adopted model bias function;
Subelement is built, for building the optimal models of the end to end network flow at current time slots, the optimal models
To be minimised as object function with the estimate of end to end network flow at the current time slots and the deviation of actual value, and wrap
Include end-to-end at the model bias function, the end to end network discharge model at the current time slots and the current time slots
The constraint single object optimization model of the probability function of network traffics.
A kind of modeling for end to end network flow based on bayesian theory that the present embodiment provides, first, is being obtained
After the initial value for taking the end to end network flow at default continuous multiple time slots, the end at default continuous multiple time slots is arrived
End network traffics are converted into the random process of Normal Distribution, obtain the ginseng of the end to end network flow at each time slot
Number and probability, the statistical property of accurate description end to end network flow.Then, based on bayesian theory, according to the default company
Continue described in parameter and the probability structure of the end to end network flow at each time slot in multiple time slots before current time slots
End to end network discharge model at current time slots, and by Definition Model departure function, arrive the end at the current time slots
Hold the estimate of network traffics and the deviation of actual value to minimize, finally obtain end to end network flow at current time slots most
Excellent model, the correct optimal models for building end to end network flow, the accurate feature for obtaining end to end network flow.
The foregoing description of the disclosed embodiments, professional and technical personnel in the field are enable to realize or using the present invention.
A variety of modifications to these embodiments will be apparent for those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the present invention.Therefore, it is of the invention
The embodiments shown herein is not intended to be limited to, and is to fit to and principles disclosed herein and features of novelty phase one
The most wide scope caused.
Claims (8)
- A kind of 1. modeling method of the end to end network flow based on bayesian theory, it is characterised in that including:In network topology structure, the initial value of the end to end network flow at default continuous multiple time slots is obtained;By the random process that the end to end network traffic transformation at default continuous multiple time slots is Normal Distribution, obtain The parameter and probability of end to end network flow at each time slot;Based on bayesian theory, according to the end at each time slot before current time slots in default continuous multiple time slots End to end network discharge model at the parameter and the probability structure current time slots of end network traffics, the current time slots are Last time slot in default continuous multiple time slots;Definition Model departure function, make end to end network flow at the current time slots estimate and actual value deviation most Smallization, and built according to the end to end network discharge model at the model bias function and the current time slots at current time slots End to end network flow optimal models.
- 2. according to the method for claim 1, it is characterised in that it is described will be at default continuous multiple time slots it is end-to-end Network traffics are converted into the random process of Normal Distribution, obtain the parameter of the end to end network flow at each time slot And probability, including:The initial value of end to end network flow at default continuous multiple time slots is converted into the random of Normal Distribution Process, obtain the probability of the end to end network flow at each time slot;By the way that the initial value of the end to end network flow at default continuous multiple time slots is converted into Normal Distribution Random process, utilize the end to end network stream at each time slot before current time slots in default continuous multiple time slots Amount represents the end to end network flow at current time slots, obtains the parameter of the end to end network flow at each time slot, institute Current time slots are stated as last time slot in default continuous multiple time slots.
- 3. according to the method for claim 1, it is characterised in that the Definition Model departure function, make the current time slots The estimate of the end to end network flow at place and the deviation of actual value minimize, and according to the model bias function and described work as The optimal models of the end to end network flow at end to end network discharge model structure current time slots at preceding time slot, including:According to the estimate of the end to end network flow at current time slots and the deviation definition model bias function of actual value;Build current time slots at end to end network flow optimal models, the optimal models be with the current time slots at The estimate of end to end network flow and the deviation of actual value are minimised as object function, and including the model bias function, The probability function of the end to end network flow at end to end network discharge model and the current time slots at the current time slots Constraint single object optimization model.
- 4. according to the method for claim 1, it is characterised in that default continuous multiple time slots are continuous 4 time slots.
- A kind of 5. modeling of the end to end network flow based on bayesian theory, it is characterised in that including:Acquiring unit, in network topology structure, obtaining the first of the end to end network flow at default continuous multiple time slots Initial value;Unit is represented, for the end to end network flow at default continuous multiple time slots to be expressed as into Normal Distribution Random process, obtain the parameter and probability of the end to end network flow at each time slot;Construction unit, for based on bayesian theory, according to each before current time slots in default continuous multiple time slots The parameter and probability of end to end network flow at the time slot build the end to end network discharge model at the current time slots, The current time slots are last time slot in default continuous multiple time slots;Definition unit, for Definition Model departure function, make the estimate of end to end network flow at the current time slots with The deviation of actual value minimizes, and according to the end to end network discharge model at the model bias function and the current time slots Build the optimal models of the end to end network flow at current time slots.
- 6. system according to claim 5, it is characterised in that the conversion unit includes:Transforming subunit, for the initial value of the end to end network flow at default continuous multiple time slots to be converted into obedience The random process of normal distribution, obtain the probability of the end to end network flow at each time slot;Subelement is represented, for by the way that the initial value of the end to end network flow at default continuous multiple time slots is converted into The random process of Normal Distribution, at each time slot before current time slots in default continuous multiple time slots End to end network flow represent current time slots at end to end network flow, obtain the end to end network at each time slot The parameter of flow, the current time slots are last time slot in default continuous multiple time slots.
- 7. system according to claim 5, it is characterised in that the definition unit includes:Subelement is defined, for the estimate and the deviation definition mould of actual value according to the end to end network flow at current time slots Type departure function;Build subelement, for building the optimal models of the end to end network flow at current time slots, the optimal models be with The estimate of end to end network flow and the deviation of actual value at the current time slots are minimised as object function, and including institute State model bias function, the end to end network discharge model at the current time slots and the end to end network at the current time slots The constraint single object optimization model of the probability function of flow.
- 8. system according to claim 5, it is characterised in that default continuous multiple time slots are continuous 4 time slots.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710813389.5A CN107360046A (en) | 2017-09-11 | 2017-09-11 | The modeling method and system of end to end network flow based on bayesian theory |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710813389.5A CN107360046A (en) | 2017-09-11 | 2017-09-11 | The modeling method and system of end to end network flow based on bayesian theory |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107360046A true CN107360046A (en) | 2017-11-17 |
Family
ID=60290963
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710813389.5A Pending CN107360046A (en) | 2017-09-11 | 2017-09-11 | The modeling method and system of end to end network flow based on bayesian theory |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107360046A (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101707559A (en) * | 2009-10-30 | 2010-05-12 | 北京邮电大学 | System and method for diagnosing and quantitatively ensuring end-to-end quality of service |
CN102325090A (en) * | 2011-09-21 | 2012-01-18 | 电子科技大学 | Network flow estimating method |
US20160033536A1 (en) * | 2010-02-10 | 2016-02-04 | Rules-Based Medicine, Inc. | Novel groups of biomarkers for diagnosing alzheimer's disease |
-
2017
- 2017-09-11 CN CN201710813389.5A patent/CN107360046A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101707559A (en) * | 2009-10-30 | 2010-05-12 | 北京邮电大学 | System and method for diagnosing and quantitatively ensuring end-to-end quality of service |
US20160033536A1 (en) * | 2010-02-10 | 2016-02-04 | Rules-Based Medicine, Inc. | Novel groups of biomarkers for diagnosing alzheimer's disease |
CN102325090A (en) * | 2011-09-21 | 2012-01-18 | 电子科技大学 | Network flow estimating method |
Non-Patent Citations (1)
Title |
---|
HONG-HAO ZHAO,FAN-BO MENG,SI-WEN ZHAO,SI-HANG ZHAO,YI LU: "A Bayes Theory-Based Modeling Algorithm to End-to-end Network Traffic", 《ITM WEB OF CONFERENCES》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103516508B (en) | A kind of unidirectional and two-way delay variation revises the method and system of clock drift in measuring | |
CN104065398B (en) | A kind of electric power communication network network merges method for designing and the system of test platform | |
JP2007128121A (en) | Method for precisely estimating congested traffic volume generated by zone and od traffic amount | |
CN106161102B (en) | A kind of IP RAN network optimization emulation mode and system | |
Zhang et al. | Enhancing network transmission capacity by efficiently allocating node capability | |
CN109102075A (en) | Gradient updating method and relevant device during a kind of distribution is trained | |
CN106302012B (en) | A kind of PTN network simulation-optimization method and system | |
CN104518899A (en) | Network routing traffic simulation method and device | |
EP1798894A1 (en) | A fast simulated annealing for traffic matrix estimation | |
CN109766642A (en) | One kind is from evolution traffic network topological modelling approach | |
CN105071963A (en) | Multi-service traffic estimation method facing big data Internet | |
CN102571423B (en) | Generalized stochastic high-level Petri net (GSHLPN)-based network data transmission modeling and performance analysis method | |
CN103024799B (en) | The method of wireless sense network delay analysis on a large scale | |
CN102739527A (en) | Network packet loss rate detection method | |
CN104270217B (en) | A method of realizing Enhanced time synchronizing process link delay fault-tolerance | |
CN107360046A (en) | The modeling method and system of end to end network flow based on bayesian theory | |
CN103338472A (en) | Wireless network link quality estimation method | |
Li et al. | A maximum algebraic connectivity increment edge-based strategy for capacity enhancement in scale-free networks | |
JP4872702B2 (en) | Distributed workflow simulation system, method, and program | |
JP4972649B2 (en) | System and method for estimating network tomography | |
CN104968047B (en) | A kind of Forecasting Methodology of mobile network-oriented interior joint network distance | |
CN109981307A (en) | Method and system for network simulation | |
CN108650141A (en) | A kind of access model of large scale network being connected to base based on car networking | |
CN106850348A (en) | A kind of traffic matrix On-line Estimation method of the data center internet based on SDN | |
Niu et al. | Study on a new model for network traffic matrix estimation |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20171117 |
|
WD01 | Invention patent application deemed withdrawn after publication |