CN110149226B - Improved particle swarm algorithm for multi-controller deployment problem in software defined network - Google Patents
Improved particle swarm algorithm for multi-controller deployment problem in software defined network Download PDFInfo
- Publication number
- CN110149226B CN110149226B CN201910402207.4A CN201910402207A CN110149226B CN 110149226 B CN110149226 B CN 110149226B CN 201910402207 A CN201910402207 A CN 201910402207A CN 110149226 B CN110149226 B CN 110149226B
- Authority
- CN
- China
- Prior art keywords
- particle
- controller
- particles
- worst
- network
- 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
- 239000002245 particle Substances 0.000 title claims abstract description 118
- 239000013598 vector Substances 0.000 claims abstract description 47
- 238000004364 calculation method Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 6
- 238000013507 mapping Methods 0.000 claims description 5
- 230000003321 amplification Effects 0.000 claims description 3
- 238000003199 nucleic acid amplification method Methods 0.000 claims description 3
- 238000005457 optimization Methods 0.000 description 3
- 238000000034 method Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- 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
- H04L41/12—Discovery or management of network topologies
-
- 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
- H04L41/142—Network analysis or design using statistical or mathematical methods
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Feedback Control In General (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses an improved particle swarm algorithm for a multi-controller deployment problem in a software defined network, wherein each particle is represented as a solution of the multi-controller deployment problem by generating random particles and velocity vectors; in each iteration, the positions of the particles are updated according to the velocity vectors, and meanwhile, the velocity vectors of the particles are updated according to the global optimal value and the worst value of the particle swarm and the historical global optimal value and the worst value of the particles, so that the situation that the particles fall into a global optimal solution is avoided, and the reliability of an output result can be effectively improved; and finally, outputting the optimal particles in the particle swarm after the iteration condition is met, wherein the corresponding deployment scheme can effectively improve the corresponding performance, effectively improve the overall performance of the network and optimize the user experience.
Description
Technical Field
The invention relates to the technical field of Software Defined Networking (SDN), in particular to an improved particle swarm algorithm for a multi-controller deployment problem in the SDN.
Background
From 2016 to 2021, the global IP network data traffic will increase nearly three times, and the network traffic of the whole internet will rise from 1000EB per year to 2.3ZB per year, as represented by cisco-dominated network traffic forecasting projects.
Since future network traffic is predicted to increase greatly, the conventional IP network architecture has not been adapted to the demands of future networks. The software defined network, as a newly proposed network architecture, has many advantages over the traditional network, and is more adaptable to the challenges of future network scenarios.
Research on the development of SDN has shown that in large wide area networks, a single controller cannot meet the requirements of the network, and therefore in 2012, Heller et al proposed a multi-controller deployment problem in wide area networks. The controller deployment problem generally refers to how to determine the number of controllers, allocate the locations of the controllers and the mapping relationship between the controllers and the switches in the software defined network so as to optimize the network performance. In consideration of performance indexes such as reliability of a network, time delay of network data flow in the network, communication time delay between controllers, load balance between controllers and the like, in a multi-controller environment, different deployment positions of the controllers affect the performance of the performance indexes. And selecting a proper controller position, and determining a proper mapping relation between the controller and the switch, so that the overall performance of the network can be effectively improved, and the user experience is optimized.
The problem of multi-controller deployment is an NP-Hard problem, the research and solution schemes of the problem comprise an exhaustion algorithm, a greedy algorithm, a clustering algorithm, an integer programming algorithm and the like, but the obtained solution schemes are not ideal.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides an improved particle swarm algorithm for solving the problem of multi-controller deployment in a software defined network.
The technical scheme adopted by the invention for solving the technical problems is as follows:
an improved particle swarm algorithm for a multi-controller deployment problem in a software defined network comprises the following steps:
step S1, acquiring a network topology G, the number k of controllers required by the network topology G and a performance index function f to be optimized, and randomly generating a plurality of particles P to form a particle swarm P, wherein each particle P represents a solution of the multi-controller deployment problem (i.e. each particle represents the position of a group of controllers); specifically, the generating step of the particle group P includes:
step A1, integer numbering is carried out on the network topology G, each number represents a node, the shortest path distance between each node is calculated, and the shortest path distance is stored in a distance matrix D.
Step a2, randomly selects k number of numbers as a particle p, also denoted as a set of controller locations, at which to deploy both controllers and switches.
Step a3, according to the shortest path principle, allocates controllers to other switches in the network, ensuring that each switch is connected only to the controller nearest to it.
And step A4, calculating the total distance between the controller and the switch according to the controller position, the switch position and the mapping relation between the switch and the controller determined in the step A3 and the shortest path distance between each node in the network stored in the distance matrix D, and taking the average value of the total distance as the average delay of the network to be used as an index for evaluating the particle performance.
Step A5, circulating the steps A2-A4 to obtain a plurality of particles P which form a particle group P.
Step S2 is to randomly set a velocity vector V for each particle p, and to mark the set of velocity vectors V as V, specifically, the velocity vector V is a vector of 1 × k (k is regarded as a one-dimensional vector).
Step S3, updating the position of each particle p according to the current position and velocity vector of the particle p, where the calculation formula is:
pnew = pold + v;
wherein p isnewThe updated position for the particle p.
poldIs the current position of the particle p.
v is the velocity vector of particle p; further, when the velocity vector v is not an integer, rounding processing is performed on the velocity vector v.
Step S4, searching the historical optimal value and the historical worst value of each particle p, and marking as pbestAnd pworstSimultaneously searching the historical optimal value and the historical worst value of all the particles P in the whole particle swarm P, and recording as gbestAnd gworst。
Step S5, updating the velocity vector of the particle p, the update formula is:
vnew= w*vold
+c1*r1*(pbest-pold)+c2*r2(gbest-pold)
+c3* (1-r1)*(pold-pworst)+c4*(1-r2)*(pold-gworst);
wherein v isnewThe updated velocity vector for particle p.
w is the velocity inertia vector when the particle p updates the velocity vector, and the calculation formula is as follows:
w = 0.5 + (fgbest-fgworst)/fgbest;
wherein f isgbestFor the historical optimum g of all the particles P in the entire particle swarm PbestThe corresponding performance indicator function.
fgworstFor the historical worst value g of all the particles P in the whole particle group PworstThe corresponding performance indicator function.
c1、c2、c3And c4Is 1.
r1And r2Is in the interval of [0, 1 ]]The random number of (2).
S6, setting an algorithm termination condition, circularly iterating S3-S6, judging whether the algorithm termination condition is reached, if so, terminating the circular iteration, outputting the particles p after the last circulation, namely, the optimal solution of the multi-controller deployment problem, and if not, outputting vnewAs the velocity vector v in step S3, continuing the loop iteration of steps S3-S6; in this embodiment, when the number of iteration cycles of the algorithm exceeds 50, the algorithm is terminated; or g obtained in step S5bestAnd gworstIs compared to g of the last iteration of the loopbestAnd gworstIf the amplification does not exceed 10%, the algorithm is terminated.
The invention has the beneficial effects that: the method of the invention has the following advantages:
1. the particle swarm optimization method improves the particle algorithm, can dynamically change the evolution speed of the particles according to the population state, takes the worst value and the optimal value in the particle swarm into consideration at the same time, takes the worst value and the optimal value as one of the parameters of the particle evolution speed, enlarges the search space, effectively avoids the particle swarm optimization from falling into the local optimal solution, and is more favorable for finding the global optimal solution;
2. the solution obtained by the algorithm is close to the optimal value obtained by an exhaustive algorithm, and the delay of the network can be effectively reduced;
3. the algorithm has low operation complexity, can quickly obtain a solution, and effectively reduces the time required by solving compared with an exhaustive algorithm;
4. the position of the controller is used for representing a particle, the performance of the particle is calculated according to the position, and when other performance indexes such as the reliability of a network and the like are replaced, the algorithm is still effective and is suitable for various optimization scenes.
Drawings
The invention is further illustrated with reference to the following figures and examples.
FIG. 1 is a functional block diagram of the present invention;
fig. 2 is a flow chart of the present invention.
Detailed Description
Referring to fig. 1 and 2, an improved particle swarm algorithm for a multi-controller deployment problem in a software defined network includes the following steps:
step S1, obtaining a network topology G, the number k of controllers required for the network topology G, and a performance index function f (in this embodiment, the performance index function is selected as an average time delay of the network), and randomly generating a plurality of particles P to form a particle group P, where each particle P represents a solution to the multiple controller deployment problem (i.e., each particle represents a position of a group of controllers); specifically, the generating step of the particle group P includes:
step A1, integer numbering is carried out on the network topology G, each number represents a node, the shortest path distance between each node is calculated, and the shortest path distance is stored in a distance matrix D.
Step a2, randomly selects k number of numbers as a particle p, also denoted as a set of controller locations, at which to deploy both controllers and switches.
Step a3, according to the shortest path principle, allocates controllers to other switches in the network, ensuring that each switch is connected only to the controller nearest to it.
Step A4, controller position, switch position and mapping between switch and controller determined according to step A3
And (4) calculating the total distance between the controller and the switch by combining the shortest path distance between each node in the network stored in the distance matrix D, and taking the average value of the total distance as the average delay of the network to be used as an index for evaluating the performance of the particle (namely the fitness of the particle).
Step A5, circulating the steps A2-A4 to obtain a plurality of particles P which form a particle group P.
Step S2 is to randomly set a velocity vector V for each particle p, and to mark the set of velocity vectors V as V, specifically, the velocity vector V is a vector of 1 × k (k is regarded as a one-dimensional vector).
Step S3, updating the position of each particle p according to the current position and velocity vector of the particle p, where the calculation formula is:
pnew = pold + v;
wherein p isnewThe updated position for the particle p.
poldIs the current position of the particle p.
v is the velocity vector of particle p; further, since in step a1, an integer is used as the node number in the network, in this embodiment, when the velocity vector v is not an integer, rounding is performed on the velocity vector v to ensure that the updated particle position can still be mapped to the corresponding network node.
Step S4, searching the historical optimal value and the historical worst value of each particle p, and marking as pbestAnd pworstSimultaneously searching the historical optimal value and the historical worst value of all the particles P in the whole particle swarm P, and recording as gbestAnd gworst。
Step S5, updating the velocity vector of the particle p, the update formula is:
vnew= w*vold
+c1*r1*(pbest-pold)+c2*r2(gbest-pold)
+c3* (1-r1)*(pold-pworst)+c4*(1-r2)*(pold-gworst);
wherein v isnewThe updated velocity vector for particle p.
w is the velocity inertia vector when the particle p updates the velocity vector, and the calculation formula is as follows:
w = 0.5 + (fgbest-fgworst)/fgbest;
wherein f isgbestFor the historical optimum g of all the particles P in the entire particle swarm PbestThe corresponding performance indicator function.
fgworstFor the historical worst value g of all the particles P in the whole particle group PworstThe corresponding performance indicator function.
c1、c2、c3And c4Is 1.
r1And r2Is in the interval of [0, 1 ]]The random number of (2).
Similarly, when v is obtainednewWhen the value of (b) is not an integer, for vnewRounding up is performed to ensure that subsequently updated particle positions can still be mapped to corresponding network nodes.
S6, setting an algorithm termination condition, circularly iterating S3-S6, judging whether the algorithm termination condition is reached, if so, terminating the circular iteration, outputting the particles p after the last circulation, namely, the optimal solution of the multi-controller deployment problem, and if not, outputting vnewAs the velocity vector v in step S3, continuing the loop iteration of steps S3-S6; in this embodiment, the algorithm termination condition is: when the iteration number of the algorithm loop exceeds 50, the algorithm is terminated; or g obtained in step S5bestAnd gworstIs compared to g of the last iteration of the loopbestAnd gworstIf the amplification does not exceed 10%, the algorithm is terminated.
The invention generates random particles and velocity vectors, and each particle is represented as a solution of the multi-controller deployment problem; in each iteration, the positions of the particles are updated according to the velocity vectors, and meanwhile, the velocity vectors of the particles are updated according to the global optimal value and the worst value of the particle swarm and the historical global optimal value and the worst value of the particles, so that the situation that the particles fall into a global optimal solution is avoided, and the reliability of an output result can be effectively improved; and finally, outputting the optimal particles in the particle swarm after the iteration condition is met, wherein the corresponding deployment scheme can effectively improve the corresponding performance, effectively improve the overall performance of the network and optimize the user experience.
The above embodiments do not limit the scope of the present invention, and those skilled in the art can make equivalent modifications and variations without departing from the overall concept of the present invention.
Claims (4)
1. An improved particle swarm algorithm for a multi-controller deployment problem in a software defined network is characterized by comprising the following steps of:
step S1, acquiring a network topology G, the number k of controllers required by the network topology G and a performance index function f required to be optimized, and randomly generating a plurality of particles P to form a particle swarm P, wherein each particle P represents a solution of the multi-controller deployment problem, namely each particle represents the position of a group of controllers;
step S2, randomly setting a velocity vector of each particle p as V, and recording a set formed by the velocity vectors V as V;
step S3, updating the position of each particle p according to the current position and velocity vector of the particle p, where the calculation formula is:
pnew = pold + v;
wherein p isnewUpdated positions for the particle p;
poldis the current position of the particle p;
v is the velocity vector of particle p;
step S4, searching the historical optimal value and the historical history of each particle pWorst value, denoted as pbestAnd pworstSimultaneously searching the historical optimal value and the historical worst value of all the particles P in the whole particle swarm P, and recording as gbestAnd gworst;
Step S5, updating the velocity vector of the particle p, the update formula is:
vnew= w*vold
+c1*r1*(pbest-pold)+c2*r2(gbest-pold)
+c3* (1-r1)*(pold-pworst)+c4*(1-r2)*(pold-gworst);
wherein v isnewAn updated velocity vector for particle p;
w is the velocity inertia vector when the particle p updates the velocity vector, and the calculation formula is as follows:
w = 0.5 + (fgbest-fgworst)/fgbest;
wherein f isgbestFor the historical optimum g of all the particles P in the entire particle swarm PbestA corresponding performance indicator function;
fgworstfor the historical worst value g of all the particles P in the whole particle group PworstA corresponding performance indicator function;
c1、c2、c3and c4Is 1;
r1and r2Is in the interval of [0, 1 ]]The random number of (2);
s6, setting an algorithm termination condition, circularly iterating S3-S6, judging whether the algorithm termination condition is reached, if so, terminating the circular iteration, outputting the particles p after the last circulation, namely, the optimal solution of the multi-controller deployment problem, and if not, outputting vnewAs the velocity vector v in step S3, continuing the loop iteration of steps S3-S6;
in the step S1, the generating step of the particle group P includes:
step A1, carrying out integer numbering on the network topology G, wherein each number represents a node, calculating the shortest path distance between each node, and storing the shortest path distance into a distance matrix D;
step A2, randomly selecting k numbers as a particle p, also expressed as a group of controller positions, and deploying controllers and switches at the positions simultaneously;
step A3, distributing controllers for other switches in the network according to the principle of shortest path, and ensuring that each switch is only connected with the controller nearest to the switch;
step A4, according to the controller position, the switch position and the mapping relation between the switch and the controller determined in the step A3, the shortest path distance between each node in the network stored by the distance matrix D is combined to calculate the total distance between the controller and the switch, and the average value of the total distance is taken as the average delay of the network and is used as an index for evaluating the particle performance;
step A5, circulating the steps A2-A4 to obtain a plurality of particles P which form a particle group P.
2. The improved particle swarm algorithm for the multi-controller deployment problem in the software defined network as claimed in claim 1, wherein in step S2, the velocity vector v is a vector of 1 × k.
3. The improved particle swarm algorithm for the multi-controller deployment problem in the software defined network as claimed in claim 1, wherein in step S3, when the velocity vector v is not an integer, the velocity vector v is rounded.
4. The improved particle swarm algorithm for the multi-controller deployment problem in the software defined network as claimed in claim 1, wherein in the step S6, the algorithm is terminated when the iteration number of the algorithm loop exceeds 50; or g obtained in step S5bestAnd gworstIs compared to g of the last iteration of the loopbestAnd gworstIf the amplification does not exceed 10%, the algorithm is terminated.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910402207.4A CN110149226B (en) | 2019-05-14 | 2019-05-14 | Improved particle swarm algorithm for multi-controller deployment problem in software defined network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910402207.4A CN110149226B (en) | 2019-05-14 | 2019-05-14 | Improved particle swarm algorithm for multi-controller deployment problem in software defined network |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110149226A CN110149226A (en) | 2019-08-20 |
CN110149226B true CN110149226B (en) | 2022-01-28 |
Family
ID=67595327
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910402207.4A Active CN110149226B (en) | 2019-05-14 | 2019-05-14 | Improved particle swarm algorithm for multi-controller deployment problem in software defined network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110149226B (en) |
Families Citing this family (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113452540B (en) * | 2020-03-27 | 2022-08-30 | 中国电信股份有限公司 | SDN configuration method and device and computer readable storage medium |
CN111650878B (en) * | 2020-06-15 | 2021-05-04 | 北京理工大学 | Method for optimizing programmability of flow when multiple controllers in software defined network fail |
CN112217666B (en) * | 2020-09-29 | 2022-02-25 | 中国银行股份有限公司 | Software defined network optimization method and device |
CN113904933B (en) * | 2021-09-30 | 2023-08-25 | 中国船舶重工集团公司第七一九研究所 | Synchronous optimization method for performance and reliability of digital network control system |
CN114244720A (en) * | 2021-12-17 | 2022-03-25 | 湘潭大学 | Multi-controller deployment method based on improved particle swarm algorithm in SDN environment |
CN115190021B (en) * | 2022-04-24 | 2024-04-12 | 北京中电飞华通信有限公司 | Deterministic time delay service oriented slice deployment method and related equipment |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107743147A (en) * | 2017-11-06 | 2018-02-27 | 柏域信息科技(上海)有限公司 | Dynamic cloud storage SDN controllers collocation method and device based on global optimization |
CN108900330A (en) * | 2018-06-25 | 2018-11-27 | 电子科技大学 | A kind of multi-controller dispositions method defining network suitable for large scope software |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140257767A1 (en) * | 2013-03-09 | 2014-09-11 | Bigwood Technology, Inc. | PSO-Guided Trust-Tech Methods for Global Unconstrained Optimization |
-
2019
- 2019-05-14 CN CN201910402207.4A patent/CN110149226B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107743147A (en) * | 2017-11-06 | 2018-02-27 | 柏域信息科技(上海)有限公司 | Dynamic cloud storage SDN controllers collocation method and device based on global optimization |
CN108900330A (en) * | 2018-06-25 | 2018-11-27 | 电子科技大学 | A kind of multi-controller dispositions method defining network suitable for large scope software |
Non-Patent Citations (2)
Title |
---|
Xibo Yao ; Hua Wang ; .Maximizing Network Utilization for SDN Based on Particle Swarm Optimization.《IEEE》.2017, * |
樊自甫 ; 姚杰 ; 杨先辉.基于时延优化的软件定义网络控制层部署策略.《计算机应用》.2018, * |
Also Published As
Publication number | Publication date |
---|---|
CN110149226A (en) | 2019-08-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110149226B (en) | Improved particle swarm algorithm for multi-controller deployment problem in software defined network | |
CN110601973B (en) | Route planning method, system, server and storage medium | |
CN107317697B (en) | Route configuration method of OSPF (open shortest Path first) and SDN (software defined network) hybrid network | |
CN108076158B (en) | Minimum load route selection method and system based on naive Bayes classifier | |
CN113194034A (en) | Route optimization method and system based on graph neural network and deep reinforcement learning | |
CN111147370B (en) | Method and control device for determining forwarding path | |
EP2883334B1 (en) | Techniques for flooding optimization for link state protocols in a network topology | |
CN108075975B (en) | Method and system for determining route transmission path in Internet of things environment | |
CN111245722B (en) | SDN data center network flow forwarding method based on genetic algorithm | |
WO2015175311A1 (en) | Probing technique for predictive routing in computer networks | |
CN110995858A (en) | Edge network request scheduling decision method based on deep Q network | |
CN110679120B (en) | Communication network node | |
CN110311863B (en) | Routing path determination method and device | |
CN113612692B (en) | Centralized optical on-chip network self-adaptive route planning method based on DQN algorithm | |
Plakunov et al. | Data center resource mapping algorithm based on the ant colony optimization | |
CN111405634B (en) | Method and device for self-adaptive clustering of wireless sensor network | |
CN116962438A (en) | Gradient data synchronization method, system, electronic equipment and readable storage medium | |
CN113098771B (en) | Distributed self-adaptive QoS routing method based on Q learning | |
CN113094857A (en) | Energy-saving controller layout method for software defined vehicle network | |
CN110233763B (en) | Virtual network embedding algorithm based on time sequence difference learning | |
CN116389266A (en) | Method and device for slicing digital twin network based on reinforcement learning | |
CN110390395A (en) | Improved genetic algorithm suitable for self-adaptive mutation crossing of SDN multi-controller deployment problem | |
Wei et al. | G-Routing: Graph Neural Networks-Based Flexible Online Routing | |
US11595292B2 (en) | Dynamic node cluster discovery in an unknown topology graph | |
CN110768906A (en) | SDN-oriented energy-saving routing method based on Q learning |
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 |