CN115225139A - Planning method for multiple control domains of SDN (software defined network) - Google Patents
Planning method for multiple control domains of SDN (software defined network) Download PDFInfo
- Publication number
- CN115225139A CN115225139A CN202210832203.1A CN202210832203A CN115225139A CN 115225139 A CN115225139 A CN 115225139A CN 202210832203 A CN202210832203 A CN 202210832203A CN 115225139 A CN115225139 A CN 115225139A
- Authority
- CN
- China
- Prior art keywords
- planning
- satellite
- sdn
- satellite network
- individual
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B7/00—Radio transmission systems, i.e. using radiation field
- H04B7/14—Relay systems
- H04B7/15—Active relay systems
- H04B7/185—Space-based or airborne stations; Stations for satellite systems
- H04B7/1851—Systems using a satellite or space-based relay
- H04B7/18519—Operations control, administration or maintenance
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Astronomy & Astrophysics (AREA)
- Aviation & Aerospace Engineering (AREA)
- General Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Radio Relay Systems (AREA)
Abstract
The invention relates to a method for planning multiple control domains of a satellite network SDN, which comprises the following steps: s1: coding a satellite controller node and a switch node in an LEO satellite network, and randomly initializing a multi-control-domain planning scheme under a constraint relation; s2: calculating an individual fitness value; s3: determining the binding energy and the binding state of each hierarchy; s4: introducing self-adaptive weight and a reverse learning mechanism into an updating strategy of an emission part and an absorption part of the photon to update the electronic state; s5: the steps S3 to S4 are circulated to carry out iterative optimization until the maximum iteration times is reached, and the optimal combination energy and combination state are output; s6: and decoding to form an optimal planning strategy of the multiple control domains of the satellite network SDN. The invention uses the atomic orbit search algorithm to carry out optimization and improvement, introduces the self-adaptive weight and the reverse learning mechanism to update the electronic state so as to be suitable for the planning problem of the satellite network multi-control domain, thereby improving the network delay performance and realizing the network balanced load.
Description
Technical Field
The invention belongs to a satellite network technology, and particularly relates to a planning method of a satellite network SDN multi-control domain based on an improved atomic orbit search algorithm. According to the method, a satellite network delay and network load balancing model is constructed in a satellite network architecture based on an SDN, the planning effect of multiple control domains is optimized through an improved atomic orbit search algorithm, the network delay is further reduced, and the load balancing effect of a controller is improved.
Background
Software Defined Networking (SDN) decouples the control plane and the data plane, can allocate Network resources in a global view, and makes an effective resource allocation strategy, and has the advantage of flexible centralized control. SDN abstracts different, distinguishable layers of the network, making the network agile and flexible. The SDN is composed of an application plane, a data plane, and a control plane. The application plane is composed of various network applications; the data plane consists of physical switches in the network; the control plane serves as the core of the SDN and manages policies and traffic of the entire network. The three planes of the SDN mainly implement communication through API interfaces, the northbound interface establishes communication between the application plane and the control plane, and communication and interaction between the control plane and the data plane are mainly implemented by the southbound interface. Due to the adoption of the SDN, the idea of traditional network distributed management is changed, bottom hardware equipment can be managed through a controller of a control plane, network services are arranged, the operation and maintenance cost is reduced, and the operation and maintenance efficiency is improved.
With the rapid development of satellite internet technology and the rapid growth of satellite network users, the disadvantages of inflexible network configuration, coexistence of multiple protocols, different network heterogeneity and the like existing in the traditional satellite network gradually emerge. By taking the mature experience of the SDN in the ground network as a reference, the SDN is introduced into the satellite network, so that the satellite of the data plane only needs to complete simple data forwarding and hardware configuration, and the satellite of the control plane completes functions of flow table issuing, resource configuration and the like, thereby simplifying the satellite network configuration and reducing the network cost.
An Atomic orbital search Algorithm (AOS) is a novel intelligent optimization algorithm proposed in 2021, and is mainly proposed based on a quantum atom theory, and the algorithm performs iterative optimization by using the basic principles of electron density configuration, absorption and emission of atoms to energy, and has the advantages of strong optimization capability and high convergence rate. One electron in the atomic orbit search algorithm represents an individual population, the energy state of the electron corresponds to the objective function value of the individual population, the probability of the photon acting on the electron is represented by the photon rate, if the random probability is greater than the photon rate, the photon has a certain effect on the electron, otherwise, the photon is considered to have no effect on the electron, and the effect can be a particle effect or a magnetic field effect and the like. The effect of photons on electrons can be judged based on the binding energy of each layer, and when the energy state of the electrons is larger than the average value of all the energy states of the electrons, the emission effect of the photons is embodied, otherwise, the absorption effect of the photons is embodied.
Even if an atomic orbit search algorithm is adopted, the planning problem of the existing satellite network SDN multi-control domain is still embodied in that the satellite network time is prolonged, and the load is not balanced enough.
The present invention has been made in view of the above circumstances.
Disclosure of Invention
The invention aims to solve the problem of planning of multiple control domains of a satellite network under the joint influence of satellite network delay and load balancing factors, and provides a method for planning multiple control domains of a satellite network SDN based on an improved atomic orbit search algorithm. The invention uses an atomic orbit search algorithm to carry out optimization and improvement, introduces a self-adaptive weight and a reverse learning mechanism to update the electronic state so as to be suitable for the planning problem of the satellite network multi-control domain, and matches the relationship between a controller and an exchanger according to the planning method of the multi-control domain so as to improve the network delay performance and realize the network balanced load.
The invention provides a planning method for multiple control domains of a satellite network SDN, which adopts a satellite network framework based on the SDN, solves a satellite network delay and load balancing model based on an improved atomic orbit algorithm, and decodes the satellite network delay and load balancing model into the planning method for the multiple control domains of the satellite network according to the output electrons with the lowest energy state, and comprises the following steps:
s1: coding a satellite controller node and a switch node in an LEO satellite network, and randomly initializing a multi-control-domain planning scheme under a constraint relation;
s2: calculating an individual fitness value;
s3: determining the binding energy and binding state of each hierarchy;
s4: introducing self-adaptive weight and a reverse learning mechanism into an updating strategy of the emission and absorption part of the photon to update the electronic state;
s5: the steps S3 to S4 are circulated to carry out iterative optimization until the maximum iteration times is reached, and the optimal combination energy and combination state are output;
s6: and decoding to form an optimal planning strategy of the SDN multi-control domain.
Furthermore, a control plane of the SDN adopts a distributed deployment mode, a main controller is deployed at a ground station, a zone controller is deployed at a GEO satellite, a part of LEO satellites are selected to be deployed at a slave controller, an LEO satellite iridium satellite network topological structure is adopted, and all LEO satellites are regarded as switch nodes.
Further, the satellite network delay includes inter-satellite propagation delay, queuing delay in a control domain, and task processing delay of a controller node.
Further, the load balancing condition is represented by a load balancing parameter BL, and the load balancing parameter is used for evaluating the degree of difference of the loads of the LEO satellite controller nodes, and is represented by formula (10):
in the formula (f) i Presentation controller c i It controls the traffic request situation in the domain, f p Presentation controller c p Which controls traffic request conditions within the domain.
Further, in S3, the average value of the electronic coding sequences in each layer is recorded as the binding state, and the binding energy is the average value of each individual fitness value.
Further, in S4, the adaptive weight is represented by equation (12)
Wherein mgen is the maximum number of iterations, and gen is the current number of iterations.
Further, in S4, the emission of photons is represented by formula (13)
The absorption of photons is represented by formula (14)
Wherein, the first and the second end of the pipe are connected with each other,
and
current and updated individuals, alpha, respectively, for the ith electron of the k-th layer i 、γ i Is a random number vector between (0, 1), LE is the individual of the lowest energy state in the atom, BS is the binding state of the atom; LE k Being the individual of the lowest energy state in the k-th layer, BS k Is the binding state in the k-th layer, and w is the adaptive weight.
Further, the reverse learning mechanism in S4 includes:
1) For the overall optimal individual, reverse learning is performed on LE, as shown in formula (15)
Wherein the content of the first and second substances,
represents an overall optimal individual LE (the individual with the lowest energy state in the atom) after reverse learning, lb is the number of the minimum control domain, wherein 1, m is the number of the control domains, r is a random number of (0, 1), and LE is the individual with the lowest energy state in the atom;
2) For each layer of optimal individuals, to LE k Reverse learning is performed as shown in equation (16)
Wherein the content of the first and second substances,
for optimizing individual LE for k layer k (the lowest energy state in the k-th layer) of the individuals subjected to reverse learning, lb is the number of the smallest control field, where 1, m is the number of control fields, r is a random number of (0, 1), LE k Is the individual of the lowest energy state in the k-th layer;
3) By comparison
And with
The individual with the smaller fitness value is reserved as
(
An individual representing the (i + 1) th electron of the k layer which is changed after the ith electron of the k layer is updated; performing reverse learning on the overall optimal individual and the optimal individual of each layer, and comparing to obtain the (i + 1) th electronic individual
But do not
And when the situation occurs, the constraint condition of the individual is damaged, and the constraint condition is repaired in a constraint way for the individual with the damaged constraint condition after being updated (the constraint repair refers to randomly generating a group of new multi-control-domain planning methods meeting the constraint condition), so that the method is suitable for planning the multi-control-domain of the satellite network.
The invention has the advantages that: the satellite network SDN control domain planning method mainly integrates the advantages and the characteristics that LEO satellites have low time delay and excellent signal quality, and mainly researches a planning method of an LEO satellite control domain, adopts a classic LEO satellite iridium satellite network topological structure, takes all LEO satellites as switch nodes, constructs a model with the goal of optimizing satellite network time delay and network load balance, uses an improved atomic orbit search algorithm to obtain an optimal control domain planning method, and matches the relationship between a controller and a switch according to the planning method to improve network time delay performance and realize network balance load.
Drawings
Fig. 1 is a flowchart of a method for planning multiple control domains of a satellite network SDN according to the present invention.
Figure 2 is a schematic diagram of an SDN-based satellite network architecture employed by the present invention.
Fig. 3 is a schematic diagram of the series of electronic codes used in the present invention.
Fig. 4 is a schematic diagram of the variation of the adaptive weight w employed in the present invention.
Fig. 5 is a comparison of the results of the three algorithms.
Fig. 6 is an analysis diagram of objective function values for different numbers of control domains.
Fig. 7 is a graph of network load balancing parameter analysis for different numbers of control domains.
Fig. 8 is a network latency analysis graph for different numbers of control domains.
Detailed Description
The method for planning multiple control domains in a satellite network SDN according to the present invention is further described with reference to fig. 1 to 8. It is clear that the described embodiments are only a part of the embodiments of the invention, and are not exhaustive of all embodiments. It should be noted that, in the present embodiment, features of the embodiment and the embodiment may be combined with each other without conflict.
As shown in fig. 1, a schematic flow chart of a method for planning multiple control domains of a satellite network SDN according to the present invention is provided, the method for planning multiple control domains of a satellite network includes steps of using a satellite network architecture based on SDN, solving a satellite network delay and load balancing model based on an improved atomic orbit algorithm, and decoding the model into the planning multiple control domains of the satellite network according to the output electrons with the lowest energy state, and the method includes:
s1: coding a satellite controller node and a switch node in an LEO satellite network, and randomly initializing a multi-control-domain planning scheme under a constraint relation;
s2: calculating an individual fitness value;
s3: determining the binding energy and binding state of each hierarchy;
s4: the updating strategy of the emission and absorption part of the photons is improved, and the electronic state is updated by introducing a self-adaptive weight and a reverse learning mechanism;
s5: step S3 to step S4 are circulated to carry out iterative optimization until the maximum iterative times are reached, and the optimal binding energy and binding state are output;
s6: and decoding to form an optimal planning strategy of the multiple control domains of the satellite network SDN.
Detailed description of the preferred embodiment 1
1. Satellite network architecture based on SDN
Fig. 2 is a schematic diagram of an SDN-based satellite network architecture employed in the present invention. A control plane of the SDN-based satellite network adopts a distributed deployment mode, a master controller is deployed at a ground station, a zone controller is deployed at a GEO satellite, and a part of LEO satellites are selected to deploy slave controllers. The method for planning the control domain of the LEO satellite is taken as an example, a classic LEO satellite iridium satellite network topological structure is adopted, and all LEO satellites are taken as switch nodes.
2. SDN control domain planning method model analysis
The embodiment analyzes the planning method of the LEO satellite control domain, constructs a problem model aiming at the satellite network delay and network load balancing conditions, and provides a basis for the planning of the LEO satellite control domain.
(1) Description of constraints
In order to form an effective satellite control domain planning scheme, the current situation of an LEO satellite network and an SDN principle are combined, and the constraint relationship existing in the control domain planning problem is comprehensively analyzed and described as follows
(1) Controlling domain management constraints. The control domains are managed by the controllers and do not overlap.
Wherein the controller set C = [ C ] 1 ,c 2 ...c i ...c m ]Control Domain set D = [ D = [ D ] 1 ,D 2 ...D i ...D m ]，D a Presentation controller c a Control field of D b Presentation controller c a Each control domain corresponds to a corresponding controller, and each control domain can have a plurality of switches.
(2) Governing constraints of the control domain and the switch. X ij Indicating the association of the control Domain with the switch, X ij Can take the value of "0" or "1", if the switch s j At c i Within the control field is denoted by "1", s j Out of c i Use of "0" in control fields "And (4) showing. Any switch can only be under the control of a single control domain. S = [ S ] 1 ,s 2 ,...s j ...s n ]Is a collection of switch nodes.
Further, each controller may command multiple switches.
(3) Controller processing power constraints. The total load in any control domain cannot exceed the processing power of the current controller. Mu.s j For a switch s j Data stream request rate of lambda i Is a controller c i The processing power of (1).
(2) Model construction
1) The satellite network delay mainly comprises inter-satellite propagation delay, queuing delay in a control domain and task processing delay of a controller node.
(1) Inter-satellite propagation delay. Because the satellite network has large scale and complex topological structure, the propagation delay generated by the link distance between the satellites cannot be ignored, and is mainly controlled by a control domain D i Inner LEO satellite exchanger node and controller c i The propagation delay Tci and the propagation delay Tcc between LEO satellite controller nodes in the satellite network are as follows:
in the formula (d) ij Represents the shortest link distance between the node i and the node j, and the same applies to d ir Representing the shortest link distance of node i from node r.
(2) And controlling the queuing time delay in the domain. Queuing time delay Q in LEO satellite control domain can be obtained through Little principle i ：
(3) And (5) task processing delay. Satellite controller node c i Is the time, task processing delay W, generated by the controller processing the data stream in its control domain i Is shown as
To sum up, the average total delay of the satellite network is denoted as T:
2) The load balancing condition of the current network can be represented by a load balancing parameter BL, and the load balancing parameter can evaluate the difference degree of the loads of the LEO satellite controller nodes:
in the formula (f) i Presentation controller c i It controls the traffic request situation in the domain, f p Presentation controller c p Which controls the traffic request situation within the domain.
The LEO satellite network multi-controller deployment problem model established in this embodiment to optimize satellite network delay and network load balancing is as follows:
min F＝α*BL+β*T (11)
wherein α + β =1, 0. Ltoreq. α, β. Ltoreq.1.
3. Implementation steps of satellite network SDN control domain planning method based on improved atomic orbit search algorithm
The overall flow of the method for planning multiple control domains of the satellite network SDN based on the atomic orbit search algorithm is shown in fig. 1.
Solving is carried out on the satellite network time delay and load balancing model based on an improved atomic orbit algorithm, and decoding is carried out to obtain the satellite network multi-control-domain planning method according to the electrons with the lowest energy state.
1) And coding a satellite controller node and a switch node in the LEO satellite network, and randomly initializing a multi-control-domain planning scheme under a constraint relation.
According to the iridium network topology, 66 LEO satellites in the iridium constellation are in total, so that the 66 LEO satellites are all regarded as satellite switch nodes, and a set of satellite switch nodes can be represented as S = [ S ] 1 ,s 2 ...s j ...s 66 ]Set of satellite controller nodes C = [ C = [ C ] 1 ,c 2 ...c i ...c m ]Set of satellite controller domains D = [ D ] 1 ,D 2 ...D i ...D m ]And the number of the satellite controller nodes is m. In order to allocate 66 switches to m controller nodes, that is, to plan 66 switches to m control domains, the method is implemented by setting an array with a length of 66 and an array element size of 1-m, and based on the constraint relationship of the embodiment, a constraint restriction is performed to form an effective array, that is, an electronic coding sequence, where one electron represents a feasible solution, as shown in fig. 3, an electronic coding sequence X = [ m,3, 2,4,1, \ 82305, 4, m]. Npop electronic coding sequences are then generated according to the population number npop.
In FIG. 3, "X [1 ]]= m "illustrates the first switch being dropped to control domain D m In (A), "X2]=3 "illustrates the second switch being dropped to control domain D 3 In the middle, control domain planning is performed in sequence.
2) Individual fitness value calculation
Because each electron has an energy state corresponding to the fitness value of an individual, the individual fitness value of the population is calculated.
3) Determining binding energy and binding state of each hierarchy
The average of the electronic coding sequences in each layer was recorded as the binding state, and the binding energy was the average of the individual fitness values.
4) Updating electronic states
The action of photons, particles or magnetic fields on electrons changes the electron population and its energy state. Because the original algorithm judges whether the photon acts on the electron based on the photon rate, if the generated random number is larger than the photon rate, the influence of the photon action is considered. Photons can absorb and emit electrons, and the two functions are judged based on the binding energy of the hierarchy. When the binding energy is higher than the level, the emission is performed, otherwise, the absorption is performed. In order to adapt to the multi-control-domain planning method, the updating strategy of the emission and absorption parts of the photons is improved.
(1) Photon action
Formula (12) shows that w is the adaptive weight, mgen is the maximum iteration number, gen is the current iteration number, w decreases from 1 to 0 with the increase of the iteration number, the change amplitude of w is small in the early stage of iteration, the change amplitude is large in the later stage of iteration, and the change is shown in fig. 4.
In the formulae (13) and (14),
and
current and updated individuals, alpha, respectively, for the ith electron of the k-th layer i 、γ i Is a random number vector between (0, 1), and w is an adaptive weight. Formula (13) is expressed as the emission of photons, where LE is the lowest energy state of the atoms and BS is the binding state of the atoms. Formula (14) is shown as absorption of photons, where LE k Being the individual of the lowest energy state in the k-th layer, BS k Is in a bonded state in the k-th layer. Considering that the atomic orbit algorithm is easy to generate the premature phenomenon in the later local search, and the updating of the electronic individuals mainly carries out random exploration on the original individuals around the overall optimal individuals or the optimal individuals of each layer, the self-adaptive weight is added on the individuals with the lowest energy state, so that the convergence speed and the accuracy of the algorithm are improved. In the embodiment, the adaptive weight is mainly applied to the action stage of photons on electrons as shown in formulas (13) and (14), the w change amplitude in the early stage of iteration is small, the electronic individuals mainly change around the optimal individuals or the optimal individuals of each layer, and the w change amplitude in the later stage of iteration is increased, so that the situation that the electronic individuals fall into a local extreme value is avoided.
(2) Action of particles or magnetic fields
When the generated random number is smaller than the photon velocity, it means that the effect of the photon on the electron is not possible, and the movement of the electron between the different layers around the nucleus is considered based on the effect of the particle or the magnetic field, and the absorption or emission of energy is generated. Because electrons are randomly changed under the action of particles or magnetic fields and the like in the traditional atomic orbit search algorithm, the invention changes the traditional updating formula, introduces a reverse learning mechanism, and simultaneously performs reverse learning on the overall optimal individual and the optimal individual of each layer, thereby further expanding the search range and increasing the global search capability.
And aiming at the overall optimal individual, performing reverse learning on the LE, as shown in the formula
In the formula (15), lb is the number of the minimum control field, here, 1, m is the number of control fields, and r is a random number of (0, 1).
For each layer of optimal individuals, for LE k Reverse learning is performed as shown in the formula
In the formula (16), lb is the number of the minimum control field, where 1,m is the number of control fields, and r is a random number of (0,1).
And performing reverse learning on the integral optimal individual and the optimal individual of each layer, and then performing constraint restoration, so that the method is suitable for planning of multiple control domains of the satellite network. By comparison
And with
The individual with smaller fitness value is reserved as
5) Outputting the optimal binding energy and binding state
And (5) circulating the steps 3) to 4) to carry out iterative optimization until the maximum iteration number mgen is reached, and outputting the optimal binding energy and binding state.
6) Decoding
And decoding according to the finally output optimal combination state to form an optimal planning strategy of the satellite network multi-control-domain.
Specific example 2
Simulation and performance evaluation
The invention discloses a method for planning a satellite network SDN control domain based on an atomic orbit search algorithm, which solves the reasonable planning of the satellite network SDN control domain by improving the traditional atomic orbit search algorithm. The improved atomic orbital search algorithm (IAOS), the Particle Swarm Optimization (PSO) and the atomic orbital search Algorithm (AOS) provided by the invention are subjected to experimental simulation, the iteration number mgen is set to be 200, and the population number npop is set to be 200. When the number m of satellite control domains is 12, the operation results of the different algorithms are shown in fig. 5. Compared with PSO and AOS, the IAOS has the best optimizing result, and compared with AOS, the IAOS is remarkably improved in both global exploration capacity and convergence rate.
Different control domain numbers are set and a primary simulation result is randomly selected. As shown in fig. 6 to 8, as the number of control domains increases, both the objective function value and the network delay are in a downward trend, and the load balancing parameter is in an upward trend, because the connectable paths between the inter-satellite controller nodes and the switch nodes increase with the increase of the number of control domains, the selectivity of the controller for processing the switch request increases, the switch request is processed by adopting the principle of proximity, the load balancing parameter between the controllers becomes larger, and the optimization advantage of IAOS is not affected by the number of control domains in view of the overall performance, and a better satellite network SDN control domain planning method can still be obtained.
It should be understood that the above-described embodiments of the present invention are examples for clearly illustrating the invention, and are not to be construed as limiting the embodiments of the present invention, and it will be obvious to those skilled in the art that various changes and modifications can be made on the basis of the above description, and it is not intended to exhaust all embodiments, and obvious changes and modifications can be made on the basis of the technical solutions of the present invention.
Claims (8)
1. A planning method for multiple control domains of a satellite network SDN is characterized by comprising the following steps:
s1: coding a satellite controller node and a switch node in an LEO satellite network, and randomly initializing a multi-control-domain planning scheme under a constraint relation;
s2: calculating an individual fitness value;
s3: determining the binding energy and binding state of each hierarchy;
s4: introducing self-adaptive weight and a reverse learning mechanism into an updating strategy of an emission part and an absorption part of the photon to update the electronic state;
s5: the steps S3 to S4 are circulated to carry out iterative optimization until the maximum iteration times is reached, and the optimal combination energy and combination state are output;
s6: and decoding to form an optimal planning strategy of the SDN multi-control domain.
2. The method for planning multiple control domains in a satellite network SDN according to claim 1, wherein the control plane in the satellite network SDN is deployed in a distributed manner, the master controller is deployed at a ground station, the area controller is deployed at a GEO satellite, a part of the LEO satellites are deployed at the slave controllers, a LEO satellite iridium network topology is adopted, and all LEO satellites are considered as switch nodes.
3. The method for planning multiple control domains in a satellite network SDN according to claim 1, wherein the time delay of the satellite network comprises inter-satellite propagation time delay, queuing time delay in a control domain, and task processing time delay of a controller node.
4. The method for planning multiple control domains in an SDN of a satellite network according to claim 1, wherein the load balancing of the satellite network is represented by a load balancing parameter BL, and the load balancing parameter BL is used for evaluating the degree of difference of the loads of the LEO satellite controller nodes, and is represented by formula (10):
in the formula, f i Presentation controller c i It controls the traffic request situation in the domain, f p Presentation controller c p Which controls the traffic request situation within the domain.
5. The method for planning multiple control domains in an SDN as defined in claim 1, wherein in S3, an average value of the electronic coding sequences in each layer is recorded as a binding state, and the binding energy is an average value of each individual fitness value.
6. The method for planning multiple control domains in a satellite network SDN according to claim 1, wherein in S4, the adaptive weight is represented by formula (12)
Wherein mgen is the maximum iteration number, and gen is the current iteration number.
7. The method for planning multiple control domains in SDN of claim 1, wherein in S4, the emission of photons is represented by formula (13)
The absorption of photons is represented by formula (14)
Wherein, the first and the second end of the pipe are connected with each other,
and
current and updated individuals, respectively, of the ith electron of the k-th layer, alpha i 、γ i Is a random number vector between (0, 1), LE is the individual of the lowest energy state in the atom, BS is the binding state of the atom; LE k Being the individual of the lowest energy state in the k-th layer, BS k Is in a bonded state in the k-th layer.
8. The method for planning multiple control domains in an SDN network according to claim 1, wherein the reverse learning mechanism in S4 includes:
1) For the overall optimal individual LE, reverse learning is performed on the LE, as shown in formula (15)
Wherein, the first and the second end of the pipe are connected with each other,
representing the whole optimal individual LE after reverse learning, lb is the number of the minimum control domain, wherein 1, m is the number of the control domains, r is a random number of (0, 1), and LE is the individual of the lowest energy state in the atom;
2) Optimal individual LE for each layer k To LE, to LE k Reverse learning is performed as shown in equation (16)
Wherein the content of the first and second substances,
for optimizing the individual LE for the k-th layer k For the individuals subjected to reverse learning, lb is the number of the smallest control field, where 1,m is the number of control fields, r is a random number of (0, 1), LE k Is the individual of the lowest energy state in the k-th layer;
3) By comparison
And with
The individual with smaller fitness value is reserved as
Performing reverse learning on the overall optimal individual and the optimal individual of each layer, and comparing to obtain the (i + 1) th electronic individual
But do not
Constraint limits of individuals may be damaged, and when the situation occurs, constraint repair is carried out on the individuals with the constraint conditions damaged after updating, so that the individuals are suitable for planning of multiple control domains of the satellite network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210832203.1A CN115225139B (en) | 2022-07-15 | 2022-07-15 | Planning method for multiple control domains of SDN (software defined network) |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210832203.1A CN115225139B (en) | 2022-07-15 | 2022-07-15 | Planning method for multiple control domains of SDN (software defined network) |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115225139A true CN115225139A (en) | 2022-10-21 |
| CN115225139B CN115225139B (en) | 2023-03-10 |
Family
ID=83610956
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210832203.1A Active CN115225139B (en) | 2022-07-15 | 2022-07-15 | Planning method for multiple control domains of SDN (software defined network) |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115225139B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090210899A1 (en) * | 2008-02-19 | 2009-08-20 | Marc Lawrence-Apfelbaum | Methods and apparatus for enhanced advertising and promotional delivery in a network |
| CN107046501A (en) * | 2017-05-16 | 2017-08-15 | 北京邮电大学 | For SDN determining method of path, device, computer equipment and storage medium |
| CN108880663A (en) * | 2018-07-20 | 2018-11-23 | 大连大学 | Incorporate network resource allocation method based on improved adaptive GA-IAGA |
| CN109981169A (en) * | 2019-04-11 | 2019-07-05 | 重庆邮电大学 | A kind of channel allocation method of the VLC-RF network based on AF panel |
| CN110688219A (en) * | 2019-09-05 | 2020-01-14 | 浙江理工大学 | Adaptive weight load balancing algorithm based on reverse chaotic cuckoo search |
| CN113852562A (en) * | 2021-09-23 | 2021-12-28 | 华北电力大学 | Communication and load resource joint optimization device based on improved particle swarm optimization |
| CN114422366A (en) * | 2022-03-28 | 2022-04-29 | 中国人民解放军战略支援部队航天工程大学 | Satellite network SDN multi-controller deployment method and system |
-
2022
- 2022-07-15 CN CN202210832203.1A patent/CN115225139B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090210899A1 (en) * | 2008-02-19 | 2009-08-20 | Marc Lawrence-Apfelbaum | Methods and apparatus for enhanced advertising and promotional delivery in a network |
| CN107046501A (en) * | 2017-05-16 | 2017-08-15 | 北京邮电大学 | For SDN determining method of path, device, computer equipment and storage medium |
| CN108880663A (en) * | 2018-07-20 | 2018-11-23 | 大连大学 | Incorporate network resource allocation method based on improved adaptive GA-IAGA |
| CN109981169A (en) * | 2019-04-11 | 2019-07-05 | 重庆邮电大学 | A kind of channel allocation method of the VLC-RF network based on AF panel |
| CN110688219A (en) * | 2019-09-05 | 2020-01-14 | 浙江理工大学 | Adaptive weight load balancing algorithm based on reverse chaotic cuckoo search |
| CN113852562A (en) * | 2021-09-23 | 2021-12-28 | 华北电力大学 | Communication and load resource joint optimization device based on improved particle swarm optimization |
| CN114422366A (en) * | 2022-03-28 | 2022-04-29 | 中国人民解放军战略支援部队航天工程大学 | Satellite network SDN multi-controller deployment method and system |
Non-Patent Citations (1)
| Title |
|---|
| 孙晨华等: "5G地面移动通信技术在低轨星座的适应性分析", 《无线电工程》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115225139B (en) | 2023-03-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113328938B (en) | Network autonomous intelligent management and control method based on deep reinforcement learning | |
| CN105430706A (en) | WSN (Wireless Sensor Networks) routing optimization method based on improved PSO (particle swarm optimization) | |
| CN115242295B (en) | Satellite network SDN multi-controller deployment method and system | |
| CN109889255B (en) | Satellite network reconstruction method based on improved bee colony algorithm | |
| WO2021036414A1 (en) | Co-channel interference prediction method for satellite-to-ground downlink under low earth orbit satellite constellation | |
| CN105323166A (en) | Cloud computing-oriented routing method based on network efficiency priority | |
| CN114629543B (en) | Satellite network self-adaptive flow scheduling method based on deep supervised learning | |
| CN116319355A (en) | Satellite network topology generation method based on deep reinforcement learning | |
| CN114567365A (en) | Routing method and system for low-earth-orbit satellite network load balancing | |
| CN115225139B (en) | Planning method for multiple control domains of SDN (software defined network) | |
| Girgis et al. | Routing and capacity assignment problem in computer networks using genetic algorithm | |
| CN116886176A (en) | Predictable inter-satellite routing method based on link utility function | |
| CN116634504A (en) | Unmanned aerial vehicle networking topology relation and bandwidth allocation optimization strategy based on improved NSGA-II algorithm | |
| CN115134928B (en) | Wireless Mesh network congestion control method with optimized frequency band route | |
| Ibraheem et al. | A multi QoS genetic-based adaptive routing in wireless mesh networks with Pareto solutions | |
| Qiao et al. | A service function chain deployment scheme of the software defined satellite network | |
| CN111542069B (en) | Method for realizing wireless AP deployment optimization based on rapid non-dominant genetic algorithm | |
| Meraihi et al. | A hybrid quantum evolutionary algorithm with cuckoo search algorithm for QoS multicast routing problem | |
| CN117014355A (en) | TSSDN dynamic route decision method based on DDPG deep reinforcement learning algorithm | |
| CN112364992A (en) | Scene-constrained model pruning method for intelligent network search | |
| CN108769944B (en) | MP-MR-MC wireless sensor network data collection method for bridge structure health monitoring | |
| Wan et al. | An improved space information networks topology algorithm | |
| CN117320027B (en) | Controller deployment method of satellite network | |
| CN116506352B (en) | Network data continuing forwarding selection method based on centralized reinforcement learning | |
| CN113037425B (en) | Multi-target controller placement method based on evolution perception in network |
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 |