CN114338400B - SDN network dynamic control method and device - Google Patents
SDN network dynamic control method and device Download PDFInfo
- Publication number
- CN114338400B CN114338400B CN202111665306.5A CN202111665306A CN114338400B CN 114338400 B CN114338400 B CN 114338400B CN 202111665306 A CN202111665306 A CN 202111665306A CN 114338400 B CN114338400 B CN 114338400B
- Authority
- CN
- China
- Prior art keywords
- information
- node
- network
- retransmission
- target tracking
- 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 63
- 230000005540 biological transmission Effects 0.000 claims description 27
- 238000007637 random forest analysis Methods 0.000 claims description 23
- 238000012549 training Methods 0.000 claims description 18
- 238000005070 sampling Methods 0.000 claims description 6
- 238000012216 screening Methods 0.000 claims description 5
- 238000001514 detection method Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000004590 computer program Methods 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 2
- 238000004891 communication Methods 0.000 abstract description 2
- 239000000523 sample Substances 0.000 description 10
- 238000005457 optimization Methods 0.000 description 8
- 238000003066 decision tree Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 230000003287 optical effect Effects 0.000 description 3
- 238000012217 deletion Methods 0.000 description 2
- 230000037430 deletion Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006855 networking Effects 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 235000008694 Humulus lupulus Nutrition 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 230000001934 delay Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
Landscapes
- Mobile Radio Communication Systems (AREA)
Abstract
The disclosure provides an SDN network dynamic control method and device; relates to the technical field of wireless communication. The method comprises the following steps: acquiring network link information and network node information of an SDN network in real time; inputting the acquired information into a trained retransmission prediction model to obtain first retransmission information of each node in the SDN network, and determining a target tracking node; calculating the feature importance information of the target tracking node based on the trained retransmission prediction model; determining key characteristics of the corresponding target tracking nodes; based on the key characteristics, corresponding control information is generated, and the control information is issued to the corresponding target tracking node so as to adjust the configuration information of the target tracking node. The method and the device can solve the problem of network delay caused by high packet loss retransmission rate of the wireless SDN in the prior art.
Description
Technical Field
The disclosure relates to the technical field of wireless networks, and in particular relates to an SDN network dynamic control method, an SDN network dynamic control device, a computer-readable storage medium and electronic equipment.
Background
The software defined network SDN (Software Defined Networking) is used as a software-based network architecture technology, supports centralized network state control, and realizes transparency of the underlying network facilities to the upper layer applications. Meanwhile, the system has flexible software programming capability, so that the automatic management and control capability of the network is improved, and the problems of limited resource scale expansion, poor networking flexibility, difficulty in rapidly meeting service requirements and the like of the current network system can be solved.
For a wireless SDN network, due to the sharing characteristic of a medium, a phenomenon that a plurality of network nodes transmit on the same channel at the same time to cause collision occurs, and further, channel transmission data packet frame loss retransmission is caused. When the packet loss rate is high, a large number of retransmissions of the link are caused, thereby causing network delay.
It should be noted that the information disclosed in the above background section is only for enhancing understanding of the background of the present disclosure and thus may include information that does not constitute prior art known to those of ordinary skill in the art.
Disclosure of Invention
An object of an embodiment of the present disclosure is to provide an SDN network dynamic control method, an SDN network dynamic control device, a computer readable storage medium and an electronic device, so as to solve a problem of network delay caused when a packet loss retransmission rate of a wireless SDN network is high in a related technology to a certain extent.
According to a first aspect of the present disclosure, there is provided an SDN network dynamic control method, the method comprising:
Acquiring network link information and network node information of an SDN network in real time;
Inputting the network link information and the network node information into a trained retransmission prediction model to obtain first retransmission information of each node in an SDN network; and determining a target tracking node based on the first retransmission information; the retransmission prediction model comprises a random forest model;
Calculating the feature importance information of the target tracking node based on the trained retransmission prediction model; determining key features of the corresponding target tracking nodes based on the feature importance information;
based on the key characteristics, corresponding control information is generated, and the control information is issued to the corresponding target tracking node so as to adjust the configuration information of the target tracking node.
In an exemplary embodiment of the present disclosure, based on the foregoing solution, before the collecting, in real time, network link information and network node information of the SDN network, the method further includes:
and sending the detection data packet to each node in the SDN at preset time intervals.
In an exemplary embodiment of the disclosure, based on the foregoing aspect, after the determining the target tracking node based on the first retransmission information, the method further includes:
And dynamically adjusting the information acquisition frequency of the target tracking node.
In an exemplary embodiment of the present disclosure, based on the foregoing scheme, the method further includes:
Acquiring historical network link information, historical network node information and corresponding historical first retransmission information of an SDN network; the historical network link information comprises historical channel state information, historical signal state information and historical link packet loss information of each node, and the historical network node information comprises historical node position information and historical node energy information;
and taking the historical channel state information, the historical signal state information, the historical link packet loss information, the historical node position information and the historical node energy information of each node as a group of training samples, taking corresponding historical first retransmission information as a sample label, and generating a random forest model in a random sampling and splitting mode until reaching a training termination condition to obtain a trained random forest model.
In an exemplary embodiment of the present disclosure, based on the foregoing solution, the calculating, based on the trained retransmission prediction model, feature importance information of the target tracking node includes:
And determining the feature importance of each target tracking node by calculating the base index or the out-of-bag data error rate of each target tracking node based on the trained retransmission prediction model.
In an exemplary embodiment of the present disclosure, based on the foregoing solution, the determining, based on the feature importance information, a key feature of a corresponding target tracking node includes:
based on the size of the feature importance information, sequencing the features of each target tracking node, and screening out features positioned in the first k bits as candidate features; k is a positive number;
deleting any candidate feature from the trained retransmission prediction model, and determining second retransmission information of the model after deleting the feature;
and selecting the candidate feature corresponding to the minimum second retransmission information as the key feature of the current target tracking node.
In an exemplary embodiment of the present disclosure, based on the foregoing scheme, the network link information includes channel state information, signal state information, link packet loss information of each node, and the network node information includes node position information, node energy information; the generating corresponding control information based on the key features to adjust configuration information of the target tracking node includes:
when the key feature is channel state information, generating channel adjustment control information to adjust channel allocation of the target tracking node;
when the key feature is signal state information, generating signal state adjustment control information to adjust the signal state of the target tracking node;
When the key feature is node position information, generating node position adjustment control information to adjust the position of the target tracking node or the neighbor node;
when the key characteristic is node energy information, generating node energy adjustment control information to adjust the transmission energy of the target tracking node or the neighbor node;
And when the key characteristic is link packet loss information, generating path planning control information to re-plan the transmission path of the network node.
According to a second aspect of the present disclosure, there is provided an SDN network dynamic control device, including:
The acquisition module is used for acquiring network link information and network node information of the SDN in real time;
The tracking node determining module is used for inputting the network link information and the network node information into a trained retransmission prediction model so as to obtain first retransmission information of each node in the SDN network; and determining a target tracking node based on the first retransmission information; the retransmission prediction model comprises a random forest model;
The calculation module is used for calculating the feature importance information of the target tracking node based on the trained retransmission prediction model; determining key features of the corresponding target tracking nodes based on the feature importance information;
and the control module is used for generating corresponding control information based on the key characteristics and transmitting the control information to a corresponding target tracking node so as to adjust the configuration information of the target tracking node.
According to a third aspect of the present disclosure, there is provided a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the method of any one of the above.
According to a fourth aspect of the present disclosure, there is provided an electronic device comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to perform the method of any of the above via execution of the executable instructions.
Exemplary embodiments of the present disclosure may have some or all of the following advantages:
In the dynamic control method for the SDN provided by the exemplary embodiment of the disclosure, on one hand, network link information and network node information of the SDN are acquired in real time, and a target tracking node, namely a high-frequency retransmission node and a topology region thereof, is dynamically determined through a trained retransmission prediction model, so that the node with high retransmission rate can be dynamically tracked, the node with too high retransmission rate can be quickly and timely found, and the occurrence of the too high packet error rate is avoided. On the other hand, calculating the feature importance information of the target tracking node based on the trained retransmission prediction model; determining key features of the corresponding target tracking nodes based on the feature importance information; based on the key characteristics, corresponding control information is generated and is issued to the corresponding target tracking node, so that the target tracking node adjusts the configuration information of the target tracking node based on the control information, key influence factors of the node with high retransmission rate can be analyzed, and further the corresponding node is controlled and adjusted, and network delay is avoided. In addition, based on the determined network retransmission rate, the network condition can be prevented from being erroneously estimated, and the network estimation accuracy is improved.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the disclosure and together with the description, serve to explain the principles of the disclosure. It will be apparent to those of ordinary skill in the art that the drawings in the following description are merely examples of the disclosure and that other drawings may be derived from them without undue effort.
Fig. 1 shows a schematic diagram of an exemplary scenario architecture of an existing SDN network dynamic control method and apparatus.
Fig. 2 schematically illustrates a flow chart of an SDN network dynamic control method in accordance with an embodiment of the disclosure.
Fig. 3 schematically illustrates a flow chart of determining key features in one embodiment according to the present disclosure.
Fig. 4 schematically illustrates a flowchart of a specific implementation procedure of an SDN network dynamic control method according to an embodiment of the disclosure.
Fig. 5 schematically illustrates a block diagram of an SDN network dynamic control device in another embodiment according to the disclosure.
Fig. 6 illustrates a block diagram of an electronic device suitable for use in implementing embodiments of the present disclosure.
Detailed Description
Example embodiments will now be described more fully with reference to the accompanying drawings. However, the exemplary embodiments may be embodied in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of the example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of embodiments of the present disclosure. One skilled in the relevant art will recognize, however, that the aspects of the disclosure may be practiced without one or more of the specific details, or with other methods, components, devices, steps, etc. In other instances, well-known technical solutions have not been shown or described in detail to avoid obscuring aspects of the present disclosure.
Furthermore, the drawings are merely schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus a repetitive description thereof will be omitted. Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software or in one or more hardware modules or integrated circuits or in different networks and/or processor devices and/or microcontroller devices.
Fig. 1 shows a schematic diagram of a system architecture 100 of an exemplary application environment to which an SDN network dynamic control method and apparatus of an embodiment of the present disclosure may be applied. As shown in fig. 1, a system architecture 100 may include an SDN controller 101, a network device 102.SDN controller 101 may be a single server, a group of servers, or a cluster of servers, and network device 102 may contain wireless sensors, wireless routers, handsets, or vehicles with wireless transmission modules, etc. The SDN controller 101 may perform overall control on the network device 102 through a southbound interface, where the network device 102 may be multiple, each network device corresponds to a radio network node, and the radio network nodes form an overall routing transport network. The present example does not limit the number of servers and network devices 102 corresponding to SDN controller 101.
The SDN network dynamic control method provided by the embodiment of the present disclosure may be executed in the SDN controller 101, and accordingly, the SDN network dynamic control device is generally disposed in the SDN controller 101.
For a wireless SDN network, due to the sharing characteristic of a medium, a phenomenon that a plurality of network nodes transmit on the same channel at the same time to generate conflict can occur, and the frame loss retransmission of a channel transmission data packet is easy to occur. Especially links with path or packet error rates up to 19 hops or more approaching or exceeding 40% can lead to a large number of retransmissions, thus causing network delays.
In current network architectures there is no mechanism to determine the number of retransmissions required for each data packet, and in related methods, data retransmissions are typically handled locally at the network node, and related information is maintained at the local MAC layer. The method aims at solving the problem that no mechanism can track the retransmission times of the data packet on the basis of node granularity (namely hop by hop). The disclosure provides a dynamic control method for a wireless SDN network.
The following describes the technical scheme of the embodiments of the present disclosure in detail:
Referring to fig. 2, an SDN network dynamic control method according to an exemplary embodiment provided by the present disclosure may include the following steps:
step S210, collecting network link information and network node information of the SDN network in real time.
In this example embodiment, an SDN network may include a plurality of wireless nodes, each node collecting its network link information and network node information in real-time. Link information and node information of network nodes can be collected through a topology discovery module of the SDN controller. A new information acquisition module can also be added in the topology discovery module of the SDN controller to acquire link information and node information. The collected network link information may be stored in a link database, and the collected network node information may be stored in a node database.
In this example embodiment, the network link information may include channel state information, signal state information, and link packet loss information. The channel state information may include channel quality information, the number of shared channels between the node and the neighbor node, etc., the signal state information may include signal strength information RSSI, signal to interference plus noise ratio SINR, the link packet loss information may include packet loss rate of all links of the node and the number of link packet loss, the network node information may include node position information and node energy information, the node position information may include GPS coordinate information of the node, distance information between the node and the neighbor node, etc., and the node energy information may include node transmission energy, transmission energy of the neighbor node, etc.
Step S220, inputting the network link information and the network node information into a trained retransmission prediction model to obtain first retransmission information of each node in the SDN network; and determining a target tracking node based on the first retransmission information; the retransmission prediction model comprises a random forest model.
In this example embodiment, the network link information and the network node information of each node may be used as input of a pre-trained retransmission prediction model, or may be used as input of a model after preprocessing (such as data cleaning or normalization) the network link information and the network node information of each node. And the model output is the retransmission information of the node. The retransmission prediction model may be a random forest model, and the training data set is used for gradually training to generate a trained random forest model. The first retransmission information may be a packet loss retransmission rate or a packet loss retransmission number, or other retransmission information, or composite information of a plurality of retransmission information, which is not particularly limited in this example.
In this exemplary embodiment, a plurality of nodes with the highest first retransmission information may be used as target tracking nodes, and a topology area formed by the target tracking nodes may be a target topology area. Therefore, the node with higher retransmission rate can be locked for key monitoring and more accurate control adjustment.
Step S230, calculating the feature importance information of the target tracking node based on the trained retransmission prediction model; and determining key features of the corresponding target tracking nodes based on the feature importance information.
In this example embodiment, the feature importance may be determined by selecting a feature importance metric to indicate the importance information of a feature, such as by a base index or an out-of-bag data error rate. Features of higher importance may be used as key features, which may be plural or one, and this example is not limited thereto.
Step S240, based on the key features, generates corresponding control information, and issues the control information to a corresponding target tracking node to adjust configuration information of the target tracking node.
In this example embodiment, the key feature may be a feature of the network link information or the network node information subordinate. The corresponding OpenFlow control information may be generated based on a key feature, for example, the key feature is a distance between the current node and the neighboring node, and the corresponding control information may be to adjust a position of the current node or the neighboring node. The OpenFlow control information of each target tracking node can be issued to the wireless node through a node configuration module of the SDN controller. The configuration information may include information on channels, transmission energy, location, routing, etc. of the nodes.
In the dynamic control method for the SDN provided by the exemplary embodiment of the disclosure, on one hand, network link information and network node information of the SDN are acquired in real time, and a target tracking node, namely a high-frequency retransmission node and a topology region thereof, is dynamically determined through a trained retransmission prediction model, so that the node with high retransmission rate can be dynamically tracked, the node with too high retransmission rate can be quickly and timely found, and the occurrence of the too high packet error rate is avoided. On the other hand, calculating the feature importance information of the target tracking node based on the trained retransmission prediction model; determining key features of the corresponding target tracking nodes based on the feature importance information; based on the key characteristics, corresponding control information is generated and is issued to the corresponding target tracking node, so that the target tracking node adjusts the configuration information of the target tracking node based on the control information, key influence factors of the node with high retransmission rate can be analyzed, and further the corresponding node is controlled and adjusted, and network delay is avoided. In addition, based on the determined network retransmission rate, the network condition can be prevented from being erroneously estimated, and the network estimation accuracy is improved.
In another embodiment, the above steps are described in more detail below.
In some embodiments, before the collecting network link information and network node information of the SDN network in real time, the method further includes:
and sending the detection data packet to each node in the SDN at preset time intervals.
In this example embodiment, a probe Packet "Packet Out" may be sent to a corresponding node at a preset fixed time interval for each node, and the sending Packet time intervals of different nodes may be the same or different. The time interval may also be adapted as the probing process proceeds. For example, the initial time interval of each node may be set to be the same, and then the variability of each node is adjusted according to the node tracking condition. For example, the initial time interval may be set to 1 second, i.e. the probe packets are sent every 1 second. The network topology information (link information and node information) is collected in the form of the data packet, so that the network topology information can be comprehensively and timely mastered, and excessive network interference cannot be caused, and normal network transmission of a data plane is not affected.
In some embodiments, after the determining a target tracking node based on the first retransmission information, the method further comprises:
And dynamically adjusting the information acquisition frequency of the target tracking node.
In this example embodiment, the target tracking node is a node with a higher retransmission rate, that is, a node that needs to be focused and adjusted, so that the information acquisition frequency of the nodes can be increased to obtain more accurate node related information, so that the frame loss retransmission information of the nodes can be predicted more accurately, and the frame loss retransmission root analysis is performed on the nodes.
In some embodiments, the information acquisition frequency of the node with the lower frame loss retransmission rate may be adjusted, for example, the information acquisition frequency of the node is reduced.
In some embodiments, the method further comprises:
Firstly, acquiring historical network link information, historical network node information and corresponding historical first retransmission information of an SDN network; the historical network link information comprises historical channel state information, historical signal state information and historical link packet loss information of each node, and the historical network node information comprises historical node position information and historical node energy information.
In the present exemplary embodiment, network link information, network node information, and corresponding retransmission information, which are a period of time (e.g., several days, ten days, one month, or several months) forward of the current time, are acquired as historical network link information, historical network node information, and corresponding historical first retransmission information for each node.
In this example embodiment, the historical channel state information may be channel state information of a period of time before the current time. The historical signal state information, the historical link packet loss information, the historical node position information and the historical node energy information are defined similarly to the historical channel state information. The historical first retransmission information may be retransmission rate information or retransmission number information of a period of time before the current time, or may be average retransmission rate or average retransmission number of the node in the period of time, or may be other operation results of the retransmission information in the period of time, which is not limited in particular in this example.
In this example embodiment, the channel state information may include channel quality information, the number of shared channels of the node and the neighbor node, etc., the signal state information may include signal strength information RSSI, signal to interference plus noise ratio SINR, the link packet loss information may include packet loss rate of all links of the node and the number of link packet loss, the network node information includes node position information and node energy information, the node position information may include GPS coordinate information of the node, distance information of the node and the neighbor node, etc., and the node energy information may include node transmission energy, transmission energy of the neighbor node, etc.
Then, taking the historical channel state information, the historical signal state information, the historical link packet loss information, the historical node position information and the historical node energy information of each node as a group of training samples, taking corresponding historical first retransmission information as a sample label, and generating a random forest model in a random sampling and splitting mode until reaching a training termination condition to obtain a trained random forest model.
In this example embodiment, a set of information (historical channel state information, historical signal state information, historical link packet loss information, historical node position information, historical node energy information) of each node may be used as a training sample, and the historical first retransmission information corresponding to the sample is used as a sample tag for training to generate a random forest model.
For example, a training generation process of a random forest model is as follows:
(1) There are N training samples, then there are N randomly selected samples to put back. Training a decision tree by using the selected N samples, and taking the N samples as samples at the root node of the decision tree.
(2) When each sample has M attribute characteristics, when each tree node of the decision tree needs to be split, randomly selecting M attribute characteristics from the M attribute characteristics, wherein M < < M. And then adopting a preset strategy (such as information gain) from the m attribute features to select 1 attribute feature as the splitting attribute of the node.
(3) In the decision tree forming process, each tree node is split according to the step (2) until the training termination condition is reached. The training termination condition in this example may be the maximum number of decision trees or that all tree nodes cannot be split again.
A plurality of decision trees may be generated according to steps (1) - (3) to construct a random forest model.
In some embodiments, the calculating the feature importance information of the target tracking node based on the trained retransmission prediction model includes:
and determining the feature importance information of each target tracking node by calculating the base index or the out-of-bag data error rate of each target tracking node based on the trained retransmission prediction model.
For example, the importance of feature X is evaluated using the out-of-bag data error rate, which is calculated as: for each decision tree, selecting corresponding out-of-bag data OOB to calculate out-of-bag data errors errOOB; randomly adding noise interference to the characteristic X of all samples of the out-of-bag data OOB, and calculating out-of-bag data error errOOB again; let N trees in the forest, the importance of feature X = Σ (errOOB 2-errOOB 1)/N.
The importance of each feature is calculated by the same method.
In some embodiments, referring to fig. 3, the determining key features of the corresponding target tracking node based on the feature importance information includes:
Step S310, sorting the features of each target tracking node based on the size of the feature importance information, and screening out the features positioned in the previous k bits as candidate features; k is a positive number.
In this exemplary embodiment, the features may be ranked from large to small in importance based on the feature importance information, and features with importance in the top k bits may be screened as candidate features.
And step S320, deleting any candidate feature from the trained retransmission prediction model, and determining second retransmission information of the model after deleting the feature.
In this exemplary embodiment, one candidate feature is deleted from the trained retransmission prediction model, and the model output after the feature deletion, that is, the second retransmission information, is determined. And executing the operation on each candidate feature to obtain the second retransmission information corresponding to the deleted candidate feature. The second retransmission information may be a predicted retransmission rate or a number of retransmissions after deleting the candidate feature.
Step S330, selecting the candidate feature corresponding to the minimum second retransmission information as the key feature of the current target tracking node.
In this exemplary embodiment, the smallest second retransmission information indicates that the corresponding deleted feature (i.e., candidate feature) is a key factor affecting the current node. That is, if a feature is removed to predict, the frame loss retransmission rate of the model is significantly reduced, and the feature is considered as a source of the excessive frame loss retransmission rate.
In some embodiments, the network link information includes channel state information, signal state information, and link packet loss information of each node, and the network node information includes node position information and node energy information; the generating corresponding control information based on the key features to adjust configuration information of the target tracking node includes:
When the key feature is channel state information, generating channel adjustment control information to adjust channel allocation of the target tracking node; when the key feature is signal state information, generating signal state adjustment control information to adjust the signal state of the target tracking node; when the key feature is node position information, generating node position adjustment control information to adjust the position of the target tracking node or the neighbor node; when the key characteristic is node energy information, generating node energy adjustment control information to adjust the transmission energy of the target tracking node or the neighbor node; and when the key characteristic is link packet loss information, generating path planning control information to re-plan the transmission path of the network node.
For example, if the analysis based on the key features is due to network interference, then a reassignment of channels is performed. And if the analysis reason based on the key characteristics is that the transmission energy of the node is too low or the transmission energy of the neighbor node is too high, adjusting the transmission energy of the corresponding node. If the analysis based on the key features is because the distance between the node and the neighbor node is too high and the distance can be adjusted by means of physical movement, the corresponding node is moved and the position is managed.
For example, referring to fig. 4, the present disclosure method is employed to implement an SDN network dynamic control process.
In step S401, the SDN controller sends a probe Packet "Packet Out" to each network device (wireless node) at preset time intervals. In this example, the preset time interval may be set to 1 second.
In step S402, the topology service module of the SDN controller collects network topology information, that is, link information and node information of each node.
In this example, the link information may include channel state information, signal state information, link packet loss information, and the like, and the node information may include location information of a current node, transmission energy information of a current node, location information of a neighbor node, transmission energy information of a neighbor node, and the like, and the neighbor node may be a 1-hop neighbor node of the current node.
In step S403, the application layer acquires the link information and the node information of each node acquired by the SDN controller through the northbound interface.
In step S404, the retransmission optimization application of the application layer extracts the link information of each node and the feature information in the node information.
In this example, the extracted feature information may include: the distance between the current node and the 1-hop neighbor, the number of channels used by the current node and the 1-hop neighbor, the transmission energy of the 1-hop neighbor of the current node, the retransmission number of the 1-hop neighbor of the current node, the RSSI of each link of the current node, the SINR of each link of the current node, the packet loss rate of each link of the current node, and the historical average retransmission number of the current node (i.e. the average retransmission number in a period of time before the current moment).
Step S405, the retransmission optimization application of the application layer inputs the extracted characteristic information into a trained retransmission prediction model to predict the retransmission rate of each node in the current network topology. In this example, the trained retransmission prediction model may be a random forest model.
In step S406, the retransmission optimization application of the application layer determines the target tracking node at the current moment based on the retransmission rate of each node. In this example, a node whose retransmission rate is greater than a preset threshold may be used as the target tracking node, that is, a node where network retransmission occurs frequently.
Step S407, the retransmission optimization application of the application layer calculates the feature importance information of the target tracking node. In this example, the feature importance information for each target tracking node may be determined by calculating the base index or out-of-bag data error rate for each target tracking node.
In step S408, the retransmission optimization application of the application layer determines key features of the corresponding target tracking node based on the feature importance information.
In this example, one or more features with the highest importance information may be used as key features of the node.
In step S409, the retransmission optimization application of the application layer sends the key feature to the SDN controller through a northbound interface.
In step S410, the SDN controller generates corresponding control information based on the key feature.
In step S411, the node configuration module of the SDN controller issues the control information to the corresponding target tracking node through the southbound interface to adjust the configuration information thereof.
In this example, when the key feature is channel state information, channel adjustment control information is generated to adjust channel allocation of the target tracking node; when the key feature is signal state information, generating signal state adjustment control information to adjust the signal state of the target tracking node; when the key feature is node position information, generating node position adjustment control information to adjust the position of the target tracking node or the neighbor node; when the key characteristic is node energy information, generating node energy adjustment control information to adjust the transmission energy of the target tracking node or the neighbor node; and when the key characteristic is link packet loss information, generating path planning control information to re-plan the transmission path of the network node. For example, if the transmission energy of the 1-hop neighbor of the random forest of the current node is the most important feature, it may be determined that the physical location of the node and its neighbor needs to be adjusted.
In this example, the retransmission optimization application of the application layer may run on the same server as the SDN controller, or may run on a different server.
The method comprises the steps that SDN controllers are used for collecting SDN wireless network basic topology information; analyzing high-frequency retransmission nodes and topology areas by using a random forest model; for high-frequency retransmission nodes and topology areas, the information acquisition frequency is dynamically improved, so that node frame loss retransmission information is more accurately measured, and root cause analysis with high retransmission rate is performed. Further, according to the analyzed reasons, corresponding OpenFlow control information is generated, the wireless network node is achieved, network delay and blocking are avoided, and service quality of the wireless SDN is improved.
In the related art, the retransmission of the wireless multi-node network uses an automatic retransmission mechanism of a link layer, and a scheme for reducing the retransmission at a system level is not used. And the method is suitable for the limited field of view situation of the distributed system. The SDN network is a global view, so that retransmission of the multi-node network can be globally optimized. Since the needed information for optimization needs to be collected by using a southbound interface and bandwidth needs to be consumed, the present disclosure adopts selective tracking of nodes with high retransmission rates. Meanwhile, the SDN may directly perform node configuration information adjustment to reduce retransmissions, such as channel reassignment, transmission capability adjustment, route recalculation, and the like.
The method and the system analyze the collected topology information by using a random forest model, determine nodes and area topologies frequently generating network retransmission in the current network topology structure, selectively track predicted high-frequency retransmission nodes and topology areas, dynamically improve information collection frequency and improve node regulation and control accuracy.
Further, in the present exemplary embodiment, there is also provided an SDN network dynamic control device 500, where the SDN network dynamic control device 500 may be used for a server. Referring to fig. 6, the SDN network dynamic control device 500 may include:
The collection module 510 may be configured to collect network link information and network node information of the SDN network in real time.
The tracking node determining module 520 may be configured to input the network link information and the network node information into a trained retransmission prediction model, so as to obtain first retransmission information of each node in the SDN network; and determining a target tracking node based on the first retransmission information; the retransmission prediction model comprises a random forest model.
A calculation module 530, configured to calculate feature importance information of the target tracking node based on the trained retransmission prediction model; and determining key features of the corresponding target tracking nodes based on the feature importance information.
The control module 540 may be configured to generate corresponding control information based on the key features, and send the control information to a corresponding target tracking node to adjust configuration information of the target tracking node.
In one exemplary embodiment of the present disclosure, the apparatus 500 further includes:
and the data packet sending module can be used for sending the detection data packet to each node in the SDN network at preset time intervals.
In one exemplary embodiment of the present disclosure, the apparatus 500 further includes:
and the acquisition frequency adjustment module can be used for dynamically adjusting the information acquisition frequency of the target tracking node.
In one exemplary embodiment of the present disclosure, the apparatus 500 further includes:
The historical information acquisition module can be used for acquiring historical network link information, historical network node information and corresponding historical first retransmission information of the SDN network; the historical network link information comprises historical channel state information, historical signal state information and historical link packet loss information of each node, and the historical network node information comprises historical node position information and historical node energy information.
The training module can be used for taking the historical channel state information, the historical signal state information, the historical link packet loss information, the historical node position information and the historical node energy information of each node as a group of training samples, taking the corresponding historical first retransmission information as a sample label, and generating a random forest model in a random sampling and splitting mode until reaching a training termination condition to obtain a trained random forest model.
In one exemplary embodiment of the present disclosure, the computing module 530 is further configured to:
and determining the feature importance information of each target tracking node by calculating the base index or the out-of-bag data error rate of each target tracking node based on the trained retransmission prediction model.
In one exemplary embodiment of the present disclosure, the computing module 530 includes:
The screening module can be used for sorting the characteristics of each target tracking node based on the size of the characteristic importance information and screening out the characteristics positioned at the first k bits as candidate characteristics; k is a positive number.
And the deletion determining module can be used for deleting any candidate feature from the trained retransmission prediction model and determining second retransmission information of the model after deleting the feature.
The selecting module can be used for selecting the candidate feature corresponding to the minimum second retransmission information as the key feature of the current target tracking node.
In one exemplary embodiment of the present disclosure, the control module 540 may also be configured to:
When the key feature is channel state information, generating channel adjustment control information to adjust channel allocation of the target tracking node; when the key feature is signal state information, generating signal state adjustment control information to adjust the signal state of the target tracking node; when the key feature is node position information, generating node position adjustment control information to adjust the position of the target tracking node or the neighbor node; when the key characteristic is node energy information, generating node energy adjustment control information to adjust the transmission energy of the target tracking node or the neighbor node; and when the key characteristic is link packet loss information, generating path planning control information to re-plan the transmission path of the network node.
The specific details of each module or unit in the above SDN network dynamic control device are described in detail in the corresponding SDN network dynamic control method, so that details are not repeated here.
As another aspect, the present application also provides a computer-readable medium that may be contained in the electronic device described in the above embodiment; or may exist alone without being incorporated into the electronic device. The computer-readable medium carries one or more programs which, when executed by one of the electronic devices, cause the electronic device to implement the methods described in the embodiments below. For example, the electronic device may implement the steps shown in fig. 2 to 4, and so on.
It should be noted that the computer readable medium shown in the present disclosure may be a computer readable signal medium or a computer readable storage medium, or any combination of the two. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this disclosure, a computer-readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In the present disclosure, however, the computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, with the computer-readable program code embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.
In addition, in an exemplary embodiment of the present disclosure, an electronic device capable of implementing the above method is also provided. Those skilled in the art will appreciate that the various aspects of the present disclosure may be implemented as a system, method, or program product. Accordingly, various aspects of the disclosure may be embodied in the following forms, namely: an entirely hardware embodiment, an entirely software embodiment (including firmware, micro-code, etc.) or an embodiment combining hardware and software aspects may be referred to herein as a "circuit," module "or" system.
An electronic device 600 according to such an embodiment of the present disclosure is described below with reference to fig. 6. The electronic device 600 shown in fig. 6 is merely an example and should not be construed to limit the functionality and scope of use of embodiments of the present disclosure in any way.
As shown in fig. 6, the electronic device 600 is in the form of a general purpose computing device. Components of electronic device 600 may include, but are not limited to: the at least one processing unit 610, the at least one memory unit 620, a bus 630 connecting the different system components (including the memory unit 620 and the processing unit 610), a display unit 640.
Wherein the storage unit stores program code that is executable by the processing unit 610 such that the processing unit 610 performs steps according to various exemplary embodiments of the present disclosure described in the above-described "exemplary methods" section of the present specification.
The storage unit 620 may include readable media in the form of volatile storage units, such as Random Access Memory (RAM) 6201 and/or cache memory unit 6202, and may further include Read Only Memory (ROM) 6203.
The storage unit 620 may also include a program/utility 6204 having a set (at least one) of program modules 6205, such program modules 6205 including, but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of which may include an implementation of a network environment.
Bus 630 may be a local bus representing one or more of several types of bus structures including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or using any of a variety of bus architectures.
The electronic device 600 may also communicate with one or more external devices 700 (e.g., keyboard, pointing device, bluetooth device, etc.), one or more devices that enable a user to interact with the electronic device 600, and/or any device (e.g., router, modem, etc.) that enables the electronic device 600 to communicate with one or more other computing devices. Such communication may occur through an input/output (I/O) interface 650. Also, electronic device 600 may communicate with one or more networks such as a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network, such as the Internet, through network adapter 660. As shown, network adapter 660 communicates with other modules of electronic device 600 over bus 630. It should be appreciated that although not shown, other hardware and/or software modules may be used in connection with electronic device 600, including, but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RA identification systems, tape drives, data backup storage systems, and the like.
From the above description of embodiments, those skilled in the art will readily appreciate that the example embodiments described herein may be implemented in software, or in combination with the necessary hardware. Thus, the technical solution according to the embodiments of the present disclosure may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (may be a CD-ROM, a U-disk, a mobile hard disk, etc.) or on a network, and includes several instructions to cause a computing device (may be a personal computer, a server, a terminal device, or a network device, etc.) to perform the method according to the embodiments of the present disclosure.
Furthermore, the above-described figures are only schematic illustrations of processes included in the method according to the exemplary embodiments of the present disclosure, and are not intended to be limiting. It will be readily appreciated that the processes shown in the above figures do not indicate or limit the temporal order of these processes. In addition, it is also readily understood that these processes may be performed synchronously or asynchronously, for example, among a plurality of modules.
It should be noted that although the steps of the methods of the present disclosure are illustrated in a particular order in the figures, this does not require or imply that the steps must be performed in that particular order or that all of the illustrated steps must be performed in order to achieve desirable results. Additionally or alternatively, certain steps may be omitted, multiple steps combined into one step to perform, and/or one step decomposed into multiple steps to perform, etc., all are considered part of the present disclosure.
It should be understood that the present disclosure disclosed and defined herein extends to all alternative combinations of two or more of the individual features mentioned or evident from the text and/or drawings. All of these different combinations constitute various alternative aspects of the present disclosure. Embodiments of the present disclosure describe the best mode known for carrying out the disclosure and will enable one skilled in the art to utilize the disclosure.
Claims (10)
1. The SDN network dynamic control method is characterized by comprising the following steps:
Acquiring network link information and network node information of an SDN network in real time;
Inputting the network link information and the network node information into a trained retransmission prediction model to obtain first retransmission information of each node in an SDN network; determining a target tracking node based on the first retransmission information, wherein the target tracking node is a high-frequency retransmission node; the retransmission prediction model is a random forest model generated by adopting a random sampling and splitting mode based on historical network link information, historical network node information and corresponding historical first retransmission information of an SDN network;
Calculating the feature importance information of the target tracking node based on the trained retransmission prediction model; determining key features of the corresponding target tracking nodes based on the feature importance information;
based on the key characteristics, corresponding control information is generated, and the control information is issued to the corresponding target tracking node so as to adjust the configuration information of the target tracking node.
2. The SDN network dynamic control method of claim 1, wherein prior to the collecting network link information and network node information of the SDN network in real time, the method further comprises:
and sending the detection data packet to each node in the SDN at preset time intervals.
3. The SDN network dynamic control method of claim 1, wherein after the determining a target tracking node based on the first retransmission information, the method further comprises:
And dynamically adjusting the information acquisition frequency of the target tracking node.
4. The SDN network dynamic control method of claim 1, the method further comprising:
Acquiring historical network link information, historical network node information and corresponding historical first retransmission information of an SDN network; the historical network link information comprises historical channel state information, historical signal state information and historical link packet loss information of each node, and the historical network node information comprises historical node position information and historical node energy information;
and taking the historical channel state information, the historical signal state information, the historical link packet loss information, the historical node position information and the historical node energy information of each node as a group of training samples, taking corresponding historical first retransmission information as a sample label, and generating a random forest model in a random sampling and splitting mode until reaching a training termination condition to obtain a trained random forest model.
5. The SDN network dynamic control method of claim 1, wherein calculating feature importance information of the target tracking node based on the trained retransmission prediction model includes:
and determining the feature importance information of each target tracking node by calculating the base index or the out-of-bag data error rate of each target tracking node based on the trained retransmission prediction model.
6. The SDN network dynamic control method of any one of claims 1-5, wherein determining key features of a corresponding target tracking node based on the feature importance information includes:
based on the size of the feature importance information, sequencing the features of each target tracking node, and screening out features positioned in the first k bits as candidate features; k is a positive number;
deleting any candidate feature from the trained retransmission prediction model, and determining second retransmission information of the model after deleting the feature;
and selecting the candidate feature corresponding to the minimum second retransmission information as the key feature of the current target tracking node.
7. The SDN network dynamic control method of claim 1, wherein the network link information includes channel state information, signal state information, link packet loss information for each node, the network node information includes node location information, node energy information; the generating corresponding control information based on the key features to adjust configuration information of the target tracking node includes:
when the key feature is channel state information, generating channel adjustment control information to adjust channel allocation of the target tracking node;
when the key feature is signal state information, generating signal state adjustment control information to adjust the signal state of the target tracking node;
When the key feature is node position information, generating node position adjustment control information to adjust the position of the target tracking node or the neighbor node;
when the key characteristic is node energy information, generating node energy adjustment control information to adjust the transmission energy of the target tracking node or the neighbor node;
And when the key characteristic is link packet loss information, generating path planning control information to re-plan the transmission path of the network node.
8. An SDN network dynamic control device, characterized by comprising:
The acquisition module is used for acquiring network link information and network node information of the SDN in real time;
The tracking node determining module is used for inputting the network link information and the network node information into a trained retransmission prediction model so as to obtain first retransmission information of each node in the SDN network; determining a target tracking node based on the first retransmission information, wherein the target tracking node is a high-frequency retransmission node; the retransmission prediction model is a random forest model generated by adopting a random sampling and splitting mode based on historical network link information, historical network node information and corresponding historical first retransmission information of an SDN network;
The calculation module is used for calculating the feature importance information of the target tracking node based on the trained retransmission prediction model; determining key features of the corresponding target tracking nodes based on the feature importance information;
and the control module is used for generating corresponding control information based on the key characteristics and transmitting the control information to a corresponding target tracking node so as to adjust the configuration information of the target tracking node.
9. A computer readable storage medium, characterized in that a computer program is stored thereon, which computer program, when being executed by a processor, implements the method according to any of claims 1-7.
10. An electronic device, comprising:
one or more processors;
Storage means for storing one or more programs which when executed by the one or more processors cause the one or more processors to implement the method of any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111665306.5A CN114338400B (en) | 2021-12-31 | 2021-12-31 | SDN network dynamic control method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111665306.5A CN114338400B (en) | 2021-12-31 | 2021-12-31 | SDN network dynamic control method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114338400A CN114338400A (en) | 2022-04-12 |
| CN114338400B true CN114338400B (en) | 2024-05-14 |
Family
ID=81021677
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111665306.5A Active CN114338400B (en) | 2021-12-31 | 2021-12-31 | SDN network dynamic control method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114338400B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105791175A (en) * | 2014-12-26 | 2016-07-20 | 电信科学技术研究院 | Method and equipment for controlling transmission resources in software defined network SDN |
| CN107181720A (en) * | 2016-03-11 | 2017-09-19 | 中兴通讯股份有限公司 | A kind of method and device of software definition networking SDN secure communications |
| CN110460502A (en) * | 2019-09-10 | 2019-11-15 | 西安电子科技大学 | Application rs traffic recognition methods under VPN based on distribution characteristics random forest |
| CN112055953A (en) * | 2018-05-01 | 2020-12-08 | 思科技术公司 | Managing multicast service chains in a cloud environment |
| KR20210075501A (en) * | 2019-12-13 | 2021-06-23 | 주식회사 엘지유플러스 | Network management apparatus and control method thereof |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9985869B2 (en) * | 2015-06-09 | 2018-05-29 | International Business Machines Corporation | Support for high availability of service appliances in a software-defined network (SDN) service chaining infrastructure |
| US10700982B2 (en) * | 2017-07-31 | 2020-06-30 | Cisco Technology, Inc. | Path selection for an application based on a performance score in a software-defined network |
-
2021
- 2021-12-31 CN CN202111665306.5A patent/CN114338400B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105791175A (en) * | 2014-12-26 | 2016-07-20 | 电信科学技术研究院 | Method and equipment for controlling transmission resources in software defined network SDN |
| CN107181720A (en) * | 2016-03-11 | 2017-09-19 | 中兴通讯股份有限公司 | A kind of method and device of software definition networking SDN secure communications |
| CN112055953A (en) * | 2018-05-01 | 2020-12-08 | 思科技术公司 | Managing multicast service chains in a cloud environment |
| CN110460502A (en) * | 2019-09-10 | 2019-11-15 | 西安电子科技大学 | Application rs traffic recognition methods under VPN based on distribution characteristics random forest |
| KR20210075501A (en) * | 2019-12-13 | 2021-06-23 | 주식회사 엘지유플러스 | Network management apparatus and control method thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114338400A (en) | 2022-04-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Subramanya et al. | Machine learning-driven service function chain placement and scaling in MEC-enabled 5G networks | |
| Zhou et al. | Performance analysis of age of information in ultra-dense Internet of Things (IoT) systems with noisy channels | |
| US8667126B2 (en) | Dynamic rate heartbeating for inter-node status updating | |
| JP6556356B2 (en) | Wireless network monitoring method and network node implementing this method | |
| Chen et al. | Minimizing age-of-information for fog computing-supported vehicular networks with deep Q-learning | |
| CN111491312B (en) | Method and equipment for predicting allocation, acquisition and training of wireless resources and neural network | |
| Li et al. | Particle filtering with applications in networked systems: a survey | |
| CN114760669B (en) | Flow prediction-based route decision method and system | |
| Pacheco et al. | Calibration-aided edge inference offloading via adaptive model partitioning of deep neural networks | |
| CN112202644B (en) | Collaborative network measurement method and system oriented to hybrid programmable network environment | |
| US20200092176A1 (en) | Machine learning algorithms for quality of service assurance in network traffic | |
| Hashmi et al. | Towards real-time user qoe assessment via machine learning on lte network data | |
| CN116782185B (en) | Resource scheduling method and device for vehicle-mounted wireless short-distance communication | |
| Montoya et al. | Energy-Efficient and Delay Sensitive Routing Paths Using Mobility Prediction in Mobile WSN: Mathematical Optimization, Markov Chains, and Deep Learning Approaches | |
| CN114338400B (en) | SDN network dynamic control method and device | |
| Soto-Vergel et al. | LoRa Performance in Industrial Environments: Analysis of Different ADR Algorithms | |
| CN116723136B (en) | Network data detection method applying FCM clustering algorithm | |
| CN112910699A (en) | Intelligent fault detection method and device for power internet of things | |
| Strelkovskaya et al. | Spline-approximation and spline-extrapolation methods in telecommunication problems | |
| Ma et al. | A variation-aware approach for task allocation in wireless distributed computing systems | |
| Ernest et al. | Age of information outage probability analysis for computation offloading in IIoT networks | |
| KR20190114126A (en) | Software-defined network controller for controlling message transmission to switch based on processing completion delay time of message and method thereof | |
| Ren et al. | End-to-end network SLA quality assurance for C-RAN: a closed-loop management method based on digital twin network | |
| CN112637904A (en) | Load balancing method and device and computing equipment | |
| CN117979388B (en) | Ad hoc network video communication method and system |
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 |