CN107566210B - Intelligent estimation method for service flow in OTN (optical transport network) - Google Patents
Intelligent estimation method for service flow in OTN (optical transport network) Download PDFInfo
- Publication number
- CN107566210B CN107566210B CN201710721664.0A CN201710721664A CN107566210B CN 107566210 B CN107566210 B CN 107566210B CN 201710721664 A CN201710721664 A CN 201710721664A CN 107566210 B CN107566210 B CN 107566210B
- Authority
- CN
- China
- Prior art keywords
- flow
- equation
- estimation
- matrix
- value
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 57
- 230000003287 optical effect Effects 0.000 title description 3
- 239000011159 matrix material Substances 0.000 claims abstract description 68
- 238000002922 simulated annealing Methods 0.000 claims abstract description 19
- 230000008569 process Effects 0.000 claims abstract description 18
- 238000005457 optimization Methods 0.000 claims abstract description 16
- 238000003325 tomography Methods 0.000 claims abstract description 16
- 238000005259 measurement Methods 0.000 claims description 13
- 230000009286 beneficial effect Effects 0.000 claims description 3
- 230000008859 change Effects 0.000 claims description 3
- 230000001186 cumulative effect Effects 0.000 claims description 3
- 238000012804 iterative process Methods 0.000 abstract description 2
- ZEFNOZRLAWVAQF-UHFFFAOYSA-N Dinitolmide Chemical compound CC1=C(C(N)=O)C=C([N+]([O-])=O)C=C1[N+]([O-])=O ZEFNOZRLAWVAQF-UHFFFAOYSA-N 0.000 description 6
- 230000006872 improvement Effects 0.000 description 5
- 230000000875 corresponding effect Effects 0.000 description 4
- 238000001514 detection method Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 230000000737 periodic effect Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
Images
Abstract
The invention belongs to the field of OTN networks, and particularly relates to an intelligent estimation method for service flow in an OTN network. Comprises the following steps of 1: considering the space-time correlation OD flow, an optimization model is constructed through a covariance matrix C, and the like, so that a primary estimation value of the OD flow is intelligently obtained; step 2: taking the obtained OD flow preliminary estimation value as an initial value of an SA iteration process of a simulated annealing algorithm, estimating deviation by using an OD flow matrix, and intelligently obtaining a globally optimal OD flow estimation value by using iteration optimization from high temperature to low temperature through the simulated annealing process; and step 3: and based on the estimation result of the OD flow in the last step, obtaining a more accurate estimation result of the OD flow by utilizing the constraint condition of the network tomography of the OD flow matrix through an iterative proportional fitting process IPFP. The estimation method can quickly select the direction of the SA iterative process to the global optimal solution, the modified SA method is simple, and the intelligent estimation of the service flow in the OTN can be quickly executed and the dynamic state of the service flow can be tracked.
Description
The technical field is as follows:
the invention belongs to the field of OTN networks, and particularly provides an intelligent estimation method for service flow in an OTN network.
Background art:
with the urgent need for network management and maintenance brought by exponential growth of IP network scale, network operators are forced to know the forwarding conditions of data packets between different nodes in the network, so as to better perform network activity traffic matrix estimation such as load balancing, traffic detection, routing optimization, network maintenance, network design and network planning. The traffic matrix represents the network traffic flowing between source and destination (OD) nodes in the network (i.e., the size of the OD flows), and the dimension of the traffic matrix is equal to the number of all OD flows in the network, which describes the data flow situation of the whole network from a global viewpoint and is an important basis for the decision of the network operator. Therefore, accurately and quickly obtaining the flow of the large-scale IP communication service flow and quickly taking correct response measures are important preconditions for ensuring the safe and efficient operation of the network, and become the leading-edge scientific problem which is commonly concerned by the academic circles and the industrial circles at home and abroad at present. However, it is difficult to directly measure network traffic. Vardi first introduced the problem of network tomography to study the indirect measurement of the traffic matrix on the network. Since then, many researchers have studied this problem and proposed many solutions. However, it is difficult, sometimes even impossible, to obtain the traffic matrix by direct measurement, and new detection techniques, methods and mechanisms must be adopted to accurately estimate the traffic in the IP in a large scale from the relatively large and constantly changing normal traffic.
To date, the SA method has been extensively studied and there are many successful solutions. The method is a simple and globally optimal non-numerical algorithm, and is suitable for solving a large-scale combinatorial optimization problem. However, due to the complexity of large-scale traffic flow estimation, it is very difficult or even impractical to directly estimate with the SA method. The solution space of the traffic matrix is a continuous real subspace. Therefore, since the conventional SA method generates a new solution with randomness, it is difficult to quickly find a globally optimal solution.
The invention content is as follows:
in order to solve the problem of large-scale flow measurement, a new traffic flow intelligent estimation method in an OTN (optical transport network) is provided from a new view, and the method is also called a simulated annealing algorithm and an iterative proportional fitting procedure (SAIPFP) method. The new solution is based on a Simulated Annealing (SA) algorithm, the time-space correlated OD flow is fully considered by the SAIPFP method, partial flow measurement is combined, and then an iterative proportion fitting process is used for satisfying the fault constraint estimation of the flow matrix, namely an accurate estimation of the flow matrix is achieved.
The invention provides an intelligent estimation method of service flow in an OTN network, which comprises the following steps,
step 1: obtaining an OD flow covariance matrix C, constructing an OD flow iterative estimation equation, constructing an optimization model through the covariance matrix C based on the consideration of the time-space correlation characteristics of the OD flow, and intelligently obtaining a primary estimation value of the OD flow by using the optimization model and the iterative estimation equation;
step 2: based on the OD flow matrix network tomography constraint condition, limiting the range of the OD flow estimation value by using the constraint condition, thereby being beneficial to obtaining an accurate OD flow estimation value, taking the OD flow preliminary estimation value obtained in the step 1 as an initial value of an SA iteration process of a simulated annealing algorithm, estimating deviation by using an OD flow matrix, and intelligently obtaining a globally optimal OD flow estimation result by using iteration optimization from high temperature to low temperature through the simulated annealing process;
and step 3: and (3) based on the OD flow estimation result obtained in the step (2), obtaining a more accurate estimation result of the OD flow through an iterative ratio fitting process IPFP by utilizing an OD flow matrix network tomography constraint condition.
Further, the covariance matrix C of the OD stream in step 1 is obtained by the following equation:
wherein C represents a covariance matrix of the OD streams;
further, the OD flow measurement analysis in step 1 is performed to obtain a preliminary estimation value of the OD flow, and the following steps are adopted,
step 11: using an OD flow measurement sample, and calculating by using a covariance matrix C equation of the OD flow to obtain a covariance matrix C;
step 12: with the network tomography equation y (t) ax (t), where y (t) is known and represents the link load at time t, a is known and represents the routing matrix, and x (t) is unknown and represents the OD traffic matrix at time t of the network, but generally since the number of network links is much smaller than the number of OD flows, the equation network tomography equation y (t) ax (t) represents the underdetermined equation of x (t) which is unknown, so there are finite number of solutions that satisfy the equation y (t) ax (t), and the purpose of the present invention is to find the optimal solution needed to satisfy the equation, and to build the following optimization model:
min||(y(t)-Axk+1(t))||2+λ|||C×Δxk+1||2
step 13: the OD flow iteration equation was constructed as above:
xk+1(t)=xk(t)+Δxk+1
wherein, Δ xk+1To satisfy the equation min | (y: (t)-Axk+1(t))||2+λ|||C×Δxk+1||2Of OD flow estimate, i.e. Δ xk+1=xk+1(t)-xk(t)
Step 14: let k equal to 0, then
x1(t)=x0(t)+Δx1
Let the current measurement be x0(t) to obtain a preliminary estimate of the network OD traffic.
Furthermore, the OD flow estimated value obtained by the network tomography constraint and the simulated annealing intelligent iteration process in the step 2 adopts the following steps,
step 21, setting the error as the scaling factor α and the initial temperature T0Minimum temperature TminMaximum number of iteration steps K and maximum constant time M, and initializing a temperature variable T ═ T0The iteration variable K is 0, the variable m of the cumulative number of times the cost function value does not change is 0, and an initial flow matrix value x _0 is given,
by the equation f (x (t) | | y (t) -ax (t) | charging |)
Wherein, x (t) represents a traffic matrix, y (t) represents a link load, a represents a routing matrix, and a cost function value f (x-0) is calculated, and x _ opt is x _ cur is x _0, and f _ min is f _ cur is f (x _ 0);
step 22: let xk(t) ═ x _ cur, and arbitrary values from 0 to 1 are generated; by solving equations
Obtaining a new estimated value x _ g of the OD flow, and calculating to obtain a cost function value f (x _ g) through an equation f (x (t)) | | y (t)) -ax (t)) |;
step 23: let xk+1(t) ═ x _ g, by equation Δ fk+1=f(xk+1(t))-f(xk(t)), calculating the deviation of the cost function value as score Δ fk+1If Δ fk+1When m is 0, m is m + 1;
step 24: by generating a random number p _ rand and then using p _ rand and Δ fk+1To decide whether to receive a new estimate x _ g of the OD stream;
step 25: letting x _ cur ═ x _ g and f _ cur ═ f (x _ g) if a new estimate value x _ g is accepted;
step 26: if f _ cur < f _ min, let x _ opt ═ x _ cur and f _ min ═ f _ cur;
step 27: if K is less than K and M is less than M, making K equal to K +1, and then returning to step 22;
step 28, reducing the current temperature to enable T to be α T, and if T is larger than or equal to Tm i nIf k is 0, m is 0, and x _ cur is x _ optf _ cur is f _ min, then the procedure returns to step 22;
step 29: let x _ sa be x _ opt, adjust x _ sa using an iterative scale fitting process to arrive at an estimate
Further, in step 24, the specific decision determining method is as follows: if Δ fk+1Less than or equal to 0, orA new estimate of OD flow is accepted and the new estimate of OD flow is discarded.
Further, the coefficient p _ rand in step 24 takes any random number within a range of 0-1.
Further, the determination method of step 3 is ifOutput result OD flow estimation valueThis is achieved byThe value is a more accurate estimate of the OD flow, otherwise the OD flow is measured directly and returned to step 21.
The intelligent estimation method for the service flow in the OTN network provided by the invention has the following advantages:
the estimation method of the invention adopts the covariance matrix OD flow, fully considers the space-time related OD flow by combining partial flow measurement, and can quickly select the direction of the SA iterative process to the global optimal solution through the covariance matrix. The method can quickly execute the intelligent estimation of the service flow in the OTN network and can track the dynamic state of the service flow.
Description of the drawings:
the invention is described in further detail below with reference to the following figures and embodiments:
FIG. 1 is a block flow diagram of the present invention;
FIG. 2 is a flow chart of a method for intelligently estimating traffic flow in an OTN network;
FIG. 3 is a graph showing the results of a two-week estimation of OD flows 29, 56, and 78 in Abilene networks;
FIG. 4 is a graph showing two-week estimates for OD, 89th, 116th and 137th flows in an Abilene network, with true values represented by black lines and estimated values for SAIPFP represented by light gray lines;
FIG. 5 is a graph corresponding to FIG. 3, from time slots 3000 to 3200;
FIG. 6 is a diagram corresponding to FIG. 4, from time slot 3000 to 3200;
FIG. 7 is a view corresponding to FIG. 3, from time slot 5000 to 5200 true black, with light colors used to estimate SAIPFP;
fig. 8 corresponds to fig. 4, from time slot 5000 to 5200, true black, with light color to estimate the SAIPFP.
The specific implementation mode is as follows:
the invention will be further explained with reference to specific embodiments, without limiting the invention.
Referring to fig. 1-2, the present invention provides an intelligent estimation method for traffic flow in an OTN network, comprising the following steps:
step 1: obtaining an OD flow covariance matrix C, constructing an OD flow iterative estimation equation, constructing an optimization model through the covariance matrix C based on the consideration of the time-space correlation characteristics of the OD flow, and intelligently obtaining a primary estimation value of the OD flow by using the optimization model and the iterative estimation equation;
step 2: based on the OD flow matrix network tomography constraint condition, limiting the range of the OD flow estimation value by using the constraint condition, thereby being beneficial to obtaining an accurate OD flow estimation value, taking the OD flow preliminary estimation value obtained in the step 1 as an initial value of an SA iteration process of a simulated annealing algorithm, estimating deviation by using an OD flow matrix, and intelligently obtaining a globally optimal OD flow estimation result by using iteration optimization from high temperature to low temperature through the simulated annealing process;
and step 3: and (3) based on the OD flow estimation result obtained in the step (2), obtaining a more accurate estimation result of the OD flow through an iterative ratio fitting process IPFP by utilizing an OD flow matrix network tomography constraint condition.
As an improvement of the scheme, the covariance matrix C of the OD stream in step 1 is obtained by the following equation:
wherein C represents a covariance matrix of the OD streams;
as an improvement of the scheme, the OD flow measurement analysis in step 1 is performed to obtain a preliminary estimate of the OD flow, and the following steps are adopted:
step 11: using an OD flow measurement sample, and calculating by using a covariance matrix C equation of the OD flow to obtain a covariance matrix C;
step 12: with the network tomography equation y (t) ax (t), where y (t) is known and represents the link load at time t, a is known and represents the routing matrix, and x (t) is unknown and represents the OD traffic matrix at time t of the network, but generally since the number of network links is much smaller than the number of OD flows, the equation network tomography equation y (t) ax (t) represents the underdetermined equation of x (t) which is unknown, so there are finite number of solutions that satisfy the equation y (t) ax (t), and the purpose of the present invention is to find the optimal solution needed to satisfy the equation, and to build the following optimization model:
min||(y(t)-Axk+1(t))||2+λ|||C×Δxk+1||2
step 13: the OD flow iteration equation was constructed as above:
xk+1(t)=xk(t)+Δxk+1
wherein,Δxk+1To satisfy the equation min | (y (t) -Axk+1(t))||2+λ|||C×Δxk+1||2Of OD flow estimate, i.e. Δ xk+1=xk+1(t)-xk(t)
Step 14: let k equal to 0, then
x1(t)=x0(t)+Δx1
Let the current measurement be x0(t) to obtain a preliminary estimate of the network OD traffic.
As an improvement of the scheme, the OD flow estimated value is obtained by network tomography constraint in the step 2 and a simulated annealing intelligent iteration process by adopting the following steps,
step 21, setting the error as the scaling factor α and the initial temperature T0Minimum temperature TminMaximum number of iteration steps K and maximum constant time M, and initializing a temperature variable T ═ T0The iteration variable K is 0, the variable m of the cumulative number of times the cost function value does not change is 0, and an initial flow matrix value x _0 is given,
by the equation f (x (t) | | y (t) -ax (t) | charging |)
Wherein, x (t) represents a traffic matrix, y (t) represents a link load, a represents a routing matrix, and a cost function value f (x-0) is calculated, and x _ opt is x _ cur is x _0, and f _ min is f _ cur is f (x _ 0);
step 22: let xk(t) ═ x _ cur, and arbitrary values from 0 to 1 are generated; by solving equations
Obtaining a new estimated value x _ g of the OD flow, and calculating to obtain a cost function value f (x _ g) through an equation f (x (t)) | | y (t)) -ax (t)) |;
step 23: let xk+1(t) ═ x _ g, by equation Δ fk+1=f(xk+1(t))-f(xk(t)), calculating the deviation of the cost function value as score Δ fk+1If Δ fk+1When m is equal to 0, m is equal to m + 1.
Step 24: by raw materialsForming a random number p _ rand, and then using p _ rand and Δ fk+1To decide whether to receive a new estimate x _ g of the OD stream;
step 25: letting x _ cur ═ x _ g and f _ cur ═ f (x _ g) if a new estimate value x _ g is accepted;
step 26: if f _ cur < f _ min, let x _ opt ═ x _ cur and f _ min ═ f _ cur;
step 27: if K is less than K and M is less than M, making K equal to K +1, and then returning to step 22;
step 28, reducing the current temperature to enable T to be α T, and if T is larger than or equal to TminIf k is 0, m is 0, and x _ cur is x _ optf _ cur is f _ min, then the procedure returns to step 22;
step 29: let x _ sa be x _ opt, adjust x _ sa using an iterative scale fitting process to arrive at an estimate
As an improvement of the scheme, in step 24, the specific decision determining method is as follows: if Δ fk+1Less than or equal to 0, orA new estimate of OD flow is accepted and the new estimate of OD flow is discarded.
As an improvement of the scheme, the coefficient p _ rand in step 24 takes any random number within the range of 0-1.
As a modification of the scheme, the determination method of the step 3 is that ifOutput result OD flow estimation valueThis is achieved byThe value is a more accurate estimation result of the OD flow; otherwise the OD flow is measured directly and returned to step 21.
The data of the Abilene network is used below to validate the estimation method and trace the analysis traffic matrix. Since the Abilene backbone network has 12 nodes, 144 ODs and 54 links loaded. In an Abilene network, according to the above-mentioned method, in combination with the mimo model shown in fig. 2, which is the algorithm of the entire estimation method, we use 54 link load inputs and 144 OD stream outputs at a time to estimate the traffic matrix at the same time. Simulation results show that the method can well track the flow matrix SAIPFP.
And predicting the traffic matrix value of two weeks in the Abilene backbone network. Fig. 3 to 8 reflect the traffic matrix trajectory, and in fig. 3 to 8, black represents the true value and light gray represents the estimated value. Fig. 3 and 4 show estimates of the flow matrix for two consecutive weeks. As can be seen from the figure, this method not only tracks the dynamics of the OD flow well, but also estimates approach the true values. The method shows that the characteristics of the OD flow of the SAIPFP large-scale IP network can be well captured through the realization of the covariance matrix C, and the OD flow value of other time slots can be accurately predicted. In addition, it can be seen from fig. 3 and 4 that although the method can accurately predict the flow rate of the SAIPFP OD flow in the case of a burst flow rate of the OD flow, the SAIPFP can also obtain an accurate estimate of the OD flow in the case of dynamic periodic changes in the OD flow, for example twenty-nine, fifty-six, and eighty-nine days of the OD flow. On the other hand, the method obtains the covariance matrix using the time of one week, but can accurately predict the traffic matrix of 2 weeks. This indicates that SAIPFP has strong prediction capability to predict large-scale IP traffic matrix
Corresponding to fig. 3 and 4, fig. 5 and 6 show OD flow estimation from slots 3000 to 3200 using SAIPFP, while fig. 7 and 8 show OD flow estimation from slots 5000 to 5200. These pictures show the estimation results in detail. Fig. 5 to 8, to a smaller extent, reflect that the SAIPFP can not only accurately track the trajectory of the traffic matrix, but also track the dynamic sum of the periodic changes of the traffic matrix.
In summary, it can be said that: SAIPFP can be used to predict large IP flow matrix, namely an intelligent estimation method of service flow in OTN network is practical and feasible.
The embodiments of the present invention have been described in detail with reference to the drawings, but the present invention is not limited to the above embodiments, and various changes can be made without departing from the spirit of the present invention within the knowledge of those skilled in the art.
Claims (1)
1. An intelligent estimation method for service flow in OTN network is characterized in that: comprises the following steps of (a) carrying out,
step 1: obtaining an OD flow covariance matrix C, constructing an OD flow iterative estimation equation, constructing an optimization model through the covariance matrix C based on the consideration of the time-space correlation characteristics of the OD flow, and intelligently obtaining a primary estimation value of the OD flow by using the optimization model and the iterative estimation equation;
the covariance matrix C is obtained by the following equation:
wherein C represents a covariance matrix of the OD streams;
the initial estimation value of the OD flow adopts the following specific steps:
step 11: using an OD flow measurement sample, and calculating by using a covariance matrix C equation of the OD flow to obtain a covariance matrix C;
step 12: by the network tomography equation y (t) ax (t), where y (t) represents the link load at time t, a represents the routing matrix, and x (t) represents the OD traffic matrix at time t of the network, but generally because the number of network links is much smaller than the number of OD flows, the equation network tomography equation y (t) ax (t) represents x (t) unknown underdetermined equation, so there are infinite solutions satisfying the equation y (t) ax (t), and the following optimization model is established:
min||(y(t)-Axk+1(t))||2+λ||C×Δxk+1||2
step 13: the OD flow iteration equation was constructed as above:
xk+1(t)=xk(t)+Δxk+1
wherein, Δ xk+1To satisfy the equation min | (y (t) -Axk+1(t))||2+λ||C×Δxk+1||2Of OD flow estimate, i.e. Δ xk+1=xk+1(t)-xk(t);
Step 14: let k equal 0, then x is obtained1(t)=x0(t)+Δx1
Let the current measurement be x0(t) obtaining a preliminary estimate of the network OD traffic;
step 2: based on the OD flow matrix network tomography constraint condition, limiting the range of the OD flow estimation value by using the constraint condition, thereby being beneficial to obtaining an accurate OD flow estimation value, taking the OD flow preliminary estimation value obtained in the step 1 as an initial value of an SA iteration process of a simulated annealing algorithm, estimating deviation by using an OD flow matrix, and intelligently obtaining a globally optimal OD flow estimation result by using iteration optimization from high temperature to low temperature through the simulated annealing process;
in this step the OD flow estimation takes the following steps,
step 21, setting the error as the scaling factor α and the initial temperature T0Minimum temperature TminMaximum number of iteration steps K and maximum constant time M, and initializing a temperature variable T ═ T0The iteration variable K is 0, the variable m of the cumulative number of times the cost function value does not change is 0, and an initial flow matrix value x _0 is given,
by the equation f (x (t) | | y (t) -ax (t) | charging |)
Wherein, x (t) represents a traffic matrix, y (t) represents a link load, a represents a routing matrix, and a cost function value f (x-0) is calculated, and x _ opt is x _ cur is x _0, and f _ min is f _ cur is f (x _ 0);
step 22: let xk(t) ═ x _ cur, and arbitrary values from 0 to 1 are generated; by solving equations
Obtaining a new estimated value x _ g of the OD flow, and calculating to obtain a cost function value f (x _ g) through an equation f (x (t)) | | y (t)) -ax (t)) |;
step 23: let xk+1(t) ═ x _ g, by equation Δ fk+1=f(xk+1(t))-f(xk(t)), calculating the deviation of the cost function value as score Δ fk+1If Δ fk+1When m is 0, m is m + 1;
step 24: by generating a random number p _ rand and then using p _ rand and Δ fk+1To decide whether to receive a new estimate x _ g of the OD stream; the specific decision determining method comprises the following steps: if Δ fk+1Less than or equal to 0, orReceiving a new OD flow estimation value, and otherwise, abandoning the new OD flow estimation value; the coefficient p _ rand takes any random number within the range of 0-1;
step 25: letting x _ cur ═ x _ g and f _ cur ═ f (x _ g) if a new estimate value x _ g is accepted;
step 26: if f _ cur < f _ min, let x _ opt ═ x _ cur and f _ min ═ f _ cur;
step 27: if K is less than K and M is less than M, making K equal to K +1, and then returning to step 22;
step 28, reducing the current temperature to enable T to be α T, and if T is larger than or equal to TminIf k is 0, m is 0, and x _ cur is x _ opt f _ cur is f _ min, then the procedure returns to step 22;
step 29: let x _ sa be x _ opt, adjust x _ sa using an iterative scale fitting process to arrive at an estimate
And step 3: based on the OD flow estimation result obtained in the step 2, obtaining a more accurate estimation result of the OD flow through an iterative ratio fitting process IPFP by using an OD flow matrix network tomography constraint condition; if it is notOutput result OD flow estimation valueThis is achieved byThe value is a more accurate estimate of the OD flow, otherwise the OD flow is measured directly and returned to step 21.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710721664.0A CN107566210B (en) | 2017-08-22 | 2017-08-22 | Intelligent estimation method for service flow in OTN (optical transport network) |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710721664.0A CN107566210B (en) | 2017-08-22 | 2017-08-22 | Intelligent estimation method for service flow in OTN (optical transport network) |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107566210A CN107566210A (en) | 2018-01-09 |
CN107566210B true CN107566210B (en) | 2020-07-10 |
Family
ID=60975845
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710721664.0A Active CN107566210B (en) | 2017-08-22 | 2017-08-22 | Intelligent estimation method for service flow in OTN (optical transport network) |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107566210B (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102769806A (en) * | 2012-07-06 | 2012-11-07 | 中国联合网络通信集团有限公司 | Resource assignment method and device of optical transmission net |
CN105207859A (en) * | 2014-06-16 | 2015-12-30 | 国家电网公司 | OTN network planning setting method in power communication network |
CN107770084A (en) * | 2016-08-19 | 2018-03-06 | 华为技术有限公司 | The management method and device of a kind of data traffic |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7957266B2 (en) * | 2004-05-28 | 2011-06-07 | Alcatel-Lucent Usa Inc. | Efficient and robust routing independent of traffic pattern variability |
-
2017
- 2017-08-22 CN CN201710721664.0A patent/CN107566210B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102769806A (en) * | 2012-07-06 | 2012-11-07 | 中国联合网络通信集团有限公司 | Resource assignment method and device of optical transmission net |
CN105207859A (en) * | 2014-06-16 | 2015-12-30 | 国家电网公司 | OTN network planning setting method in power communication network |
CN107770084A (en) * | 2016-08-19 | 2018-03-06 | 华为技术有限公司 | The management method and device of a kind of data traffic |
Also Published As
Publication number | Publication date |
---|---|
CN107566210A (en) | 2018-01-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107483251B (en) | Network service abnormity detection method based on distributed probe monitoring | |
US8503320B2 (en) | Available bandwidth estimation in a packet-switched communication network | |
CN110209148B (en) | Fault estimation method of networked system based on description system observer | |
Jiang et al. | An approximation method of origin–destination flow traffic from link load counts | |
Tian et al. | Predictive compensation for variable network delays and packet losses in networked control systems | |
US20220156574A1 (en) | Methods and systems for remote training of a machine learning model | |
US7525929B2 (en) | Fast simulated annealing for traffic matrix estimation | |
CN107566210B (en) | Intelligent estimation method for service flow in OTN (optical transport network) | |
Lunze et al. | Introduction to networked control systems | |
Gao et al. | Airgnns: Graph neural networks over the air | |
Li et al. | Estimating SDN traffic matrix based on online informative flow measurement method | |
Lu et al. | Parameter estimation of software reliability growth models by a modified whale optimization algorithm | |
Fei et al. | A fast model identification method for networked control system | |
CN115022191A (en) | Quick inversion method for end-to-end flow in IPv6 network | |
Haghrah et al. | Analysis of incremental LMS adaptive algorithm over wireless sensor networks with delayed-links | |
CN114254734A (en) | Traffic matrix modeling method supporting deterministic application | |
CN113379092B (en) | Backbone network multi-service traffic estimation method and system facing big data | |
Karakuş et al. | Estimation of the nonlinearity degree for polynomial autoregressive processes with RJMCMC | |
Okokpujie et al. | A unique mathematical queuing model for wired and wireless networks | |
Harrison | A semi-product-form for a pair of queues with finite batches: Equilibrium state probabilities and response time densities | |
CN108599834B (en) | Method and system for analyzing utilization rate of satellite communication network link | |
Chen et al. | Structure-enhanced deep reinforcement learning for optimal transmission scheduling | |
Chen et al. | Semantic-aware transmission scheduling: A monotonicity-driven deep reinforcement learning approach | |
Bazzi et al. | A quality-aware incremental lms algorithm for distributed adaptive estimation | |
CN107645418A (en) | A kind of converged communication network traffic forecast method towards intelligent power |
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 |