CN113411841A - 5G slice cutting and joining method and device and computing equipment - Google Patents
5G slice cutting and joining method and device and computing equipment Download PDFInfo
- Publication number
- CN113411841A CN113411841A CN202010186555.5A CN202010186555A CN113411841A CN 113411841 A CN113411841 A CN 113411841A CN 202010186555 A CN202010186555 A CN 202010186555A CN 113411841 A CN113411841 A CN 113411841A
- Authority
- CN
- China
- Prior art keywords
- slice
- network
- information
- matrix
- cutover
- 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
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/16—Central resource management; Negotiation of resources or communication parameters, e.g. negotiating bandwidth or QoS [Quality of Service]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0893—Assignment of logical groups to network elements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/12—Discovery or management of network topologies
Abstract
The invention discloses a 5G slice cutting and joining method, a device and computing equipment, wherein the method comprises the following steps: acquiring the information of a to-be-cut topological graph of the slicing network; decomposing the cutover topological graph information to obtain a slice adjacency matrix and a slice characteristic matrix; the method comprises the following steps that a slice adjacency matrix represents the connection relation between nodes in a cutover topological graph, and a slice characteristic matrix represents the influence attribute of a plurality of cutoffs to be performed on each slice node; inputting the slice adjacency matrix and the slice characteristic matrix into a trained slice network cut-joint combiner to predict to obtain cut-joint combination information; and performing the cutting processing according to the cutting and jointing combined information. By the method, after the slice adjacency matrix and the slice characteristic matrix are separated from the cutting topological graph information, the segmentation adjacency matrix and the slice characteristic matrix are input into the cutting network cutting combiner, the combination mode of the cutting to be performed can be predicted, and the cutting combination can be automatically completed.
Description
Technical Field
The invention relates to the technical field of communication, in particular to a 5G slice cutting and joining method, a device and computing equipment.
Background
A Network Slice (Network Slice) is an end-to-end logical function and a physical or virtual resource set required by the end-to-end logical function, including an access Network, a transmission Network, a core Network, and the like, and the Network Slice can be regarded as a virtualized "private Network" in a 5G Network; the network slice is constructed based on the uniform infrastructure of the NFV, and low-cost and efficient operation is achieved. Network slice techniques may enable logical isolation of a communication network, allowing network elements and functionality to be configured and reused in each network slice to meet specific industry application needs. And the cutover refers to high-risk operations such as capacity expansion, upgrading, transformation, replacement and configuration change of the existing network equipment by an operator, so the cutover is necessarily accompanied by risks.
However, the slicing network has many nodes and complex relationships, and compared with the conventional network, due to different individual requirements of users, the slicing networks of each user are different, the slicing network cutover complexity and difficulty are also improved in multiples, the slicing network cutover quality and efficiency directly affect the SLA (Service-Level Agreement) of the slicing user, and if the slicing network cutover is not reasonably combined, the slicing application is frequently affected. At present, slicing network cutting and splicing are mainly judged by manual experience, so that the efficiency is low, mistakes are easy to make, and the requirement on the skills of personnel is high.
Disclosure of Invention
In view of the above problems, embodiments of the present invention are proposed to provide a 5G slice segmentation and joining method, apparatus and computing device that overcome or at least partially solve the above problems.
According to an aspect of an embodiment of the present invention, there is provided a 5G slice segmentation and joining method, including:
acquiring the information of a to-be-cut topological graph of the slicing network;
decomposing the cutover topological graph information to obtain a slice adjacency matrix and a slice characteristic matrix; the method comprises the following steps that a slice adjacency matrix represents the connection relation between nodes in a cutover topological graph, and a slice characteristic matrix represents the influence attribute of a plurality of cutoffs to be performed on each slice node;
inputting the slice adjacency matrix and the slice characteristic matrix into a trained slice network cut-joint combiner to predict to obtain cut-joint combination information;
and performing the cutting processing according to the cutting and jointing combined information.
According to another aspect of the embodiments of the present invention, there is provided a 5G slice secant joining apparatus, including:
the acquisition module is suitable for acquiring the information of a topological graph to be cut of the slice network;
the decomposition module is suitable for decomposing the cutover topological graph information to obtain a slice adjacency matrix and a slice characteristic matrix; the method comprises the following steps that a slice adjacency matrix represents the connection relation between nodes in a cutover topological graph, and a slice characteristic matrix represents the influence attribute of a plurality of cutoffs to be performed on each slice node;
the merging module is suitable for inputting the slice adjacency matrix and the slice characteristic matrix into a trained slice network cut-join merger to predict and obtain cut-join merging information;
and the cutting module is suitable for performing cutting processing according to the cutting and jointing combined information.
According to still another aspect of an embodiment of the present invention, there is provided a computing device including: the system comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete mutual communication through the communication bus;
the memory is used for storing at least one executable instruction, and the executable instruction enables the processor to execute the operations corresponding to the 5G slice cutting and joining method.
According to a further aspect of the embodiments of the present invention, there is provided a computer storage medium having at least one executable instruction stored therein, the executable instruction causing a processor to perform operations corresponding to the above-mentioned 5G slice segmentation-joining method.
According to the 5G slice cutting and joining combination method, the device and the computing equipment, the slice adjacency matrix and the slice characteristic matrix are separated from the to-be-cut topological graph information of the slice network and are input into the trained slice network cutting and joining combiner, the cutting and joining combination information can be obtained in a predictable manner, and further, the original slice network cutting and joining can be combined properly and then the cutting and joining processing is carried out. Therefore, according to the scheme of the embodiment, the cutting and jointing information can be automatically predicted and output, the cutting and jointing which has a large influence on the slicing user is combined, and the influence on the slicing user is reduced; meanwhile, compared with a manual judgment and combination mode, the automatic prediction output can improve the cutting efficiency and accuracy of the slicing network.
The foregoing description is only an overview of the technical solutions of the embodiments of the present invention, and the embodiments of the present invention can be implemented according to the content of the description in order to make the technical means of the embodiments of the present invention more clearly understood, and the detailed description of the embodiments of the present invention is provided below in order to make the foregoing and other objects, features, and advantages of the embodiments of the present invention more clearly understandable.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the embodiments of the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
fig. 1 shows a flowchart of a 5G slice segmentation-joining merging method provided by an embodiment of the present invention;
FIG. 2 is a flow chart of a method for 5G slice segmentation and joining according to another embodiment of the present invention;
FIG. 3 illustrates a block diagram of a slice network cutover combiner in one particular example of the invention;
FIG. 4 illustrates a complete flow of a split-join merge method in one particular example of the invention;
FIG. 5 is a schematic structural diagram of a 5G slice secant-join apparatus provided in an embodiment of the present invention;
fig. 6 shows a schematic structural diagram of a computing device provided by an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.
Before describing embodiments of the present invention, several terms referred to herein are introduced:
1. graph Convolutional neural Networks (GCNs for short):
the graph convolutional neural network is proposed by Thomas Kpif in 2017 in the paper Semi-collaborative classification with graph conditional networks, and provides a new idea for processing graph (graph) structure data, and the convolutional neural network commonly used for images in deep learning is applied to the graph data.
Each graph neural network layer can be written as such a non-linear function:
Hl+1＝f(Hl，A)
wherein H(0)X is the input data, H(L)Z is the output data and L is the number of layers of the neural network, and different models are determined by selecting different f () and parameters.
Wherein W(l)Is a parameter matrix of the ith neural network layer, is a nonlinear activation function such as ReLU, A is a adjacency matrix, and D is a node-degree diagonal matrix of A.
2. The slice management framework mainly comprises CSMF, NSMF and NSSMF.
Among them, CSMF (Communication Service Management Function): the method and the system complete the requirement ordering and processing of the user business communication service, are responsible for converting the communication service requirement of the operator/third-party client into the requirement on the network slice, send the requirement on the network slice (such as creating, terminating, modifying the instance request of the network slice) to the NSMF through an interface between the NSMF and the NSMF, and acquire the management data (such as performance, fault data and the like) of the network slice from the NSMF.
Wherein, NSMF (Network Slice Management Function): the network slicing management system is responsible for receiving network slicing requirements sent by the CSMF, managing life cycle, performance, faults and the like of the network slicing examples, arranging the composition of the network slicing examples, decomposing the requirements of the network slicing examples into the requirements of each network slicing sub-network example or network function, and sending network slicing sub-network example management requests to each NSSMF.
Wherein, NSSMF (Network Slice Subnet Management Function): receiving a network slice subnet deployment request issued by NSMF, managing a network slice subnet instance, arranging the composition of the network slice subnet instance, mapping the SLA requirement of the network slice subnet into the QoS requirement of the network service, and issuing a deployment request of the network service to the NFVO system of the ETSI NFV domain.
Fig. 1 shows a flowchart of a 5G slice segmentation-joining method according to an embodiment of the present invention. The method may be used to guide the making of a cutover and joining of network slices, and may be performed by any computing device having data processing capabilities.
As shown in fig. 1, the method comprises the steps of:
step S110: and acquiring the information of the topological graph to be cut of the slicing network.
Wherein the cutover topology canDenoted G ═ V, E, V is the set of sliced network nodes V ═ { V1, V2, V3, …, VN }, E is the set of edges, E is the connection between network node Vi and network node Vj, then E is the set of edgesij1, otherwise eij＝0。
In this embodiment, the cutover topological graph information to be performed by the slice network plan within the nearest preset time period may be automatically obtained, where the cutover topological graph information further includes information reflecting the cutover influence.
Step S120: decomposing the cutover topological graph information to obtain a slice adjacency matrix and a slice characteristic matrix; the slice adjacency matrix represents the connection relation between the nodes in the cutover topological graph, and the slice characteristic matrix represents the influence attribute of a plurality of cutoffs to be performed on each slice node.
Specifically, the cutover topological graph information is decomposed to obtain a slice adjacency matrix and a slice feature matrix, wherein each element in the slice adjacency matrix represents a connection relation between corresponding nodes, the slice feature matrix comprises M attribute values, and influence of M slice network cutoffs to be performed on each slice node is correspondingly planned.
Step S130: and inputting the slice adjacency matrix and the slice characteristic matrix into a trained slice network cut-joint combiner to predict to obtain cut-joint combined information.
The slicing network cut-joint combiner is obtained by training a slicing adjacency matrix sample, a slicing characteristic matrix sample and a correspondingly marked optimal combination scheme information sample which correspond to a cut-joint topological graph of a historical slicing network.
Specifically, after the slice adjacency matrix and the slice feature matrix are input to the slice network cut-join combiner, the slice network cut-join combiner can finally output a cut-join scheme with the minimum influence node times and range, and cut-join combination information is obtained, wherein the cut-join combination information refers to information for combining M slice networks to be performed.
Step S140: and performing the cutting processing according to the cutting and jointing combined information.
After the cutting and jointing combined information is obtained through prediction, the M cutting and jointing network cutting and jointing to be performed can be combined according to the cutting and jointing combined information, and cutting and jointing processing is performed according to the combined result, so that the influence on a cutting user is reduced.
According to the 5G slice segmentation-joining merging method provided by the embodiment, the slice adjacency matrix and the slice feature matrix are decomposed from the to-be-performed segmentation topological graph information of the slice network and are input into the trained slice network segmentation-joining merger, so that the segmentation-joining merging information can be obtained in a predictable manner, and further, the original slice network segmentation-joining can be appropriately merged, and then, the segmentation-joining processing is performed. Therefore, according to the scheme of the embodiment, the cutting and jointing information can be automatically predicted and output, the cutting and jointing which has a large influence on the slicing user is combined, and the influence on the slicing user is reduced; meanwhile, compared with a manual judgment and combination mode, the automatic prediction output can improve the cutting efficiency and accuracy of the slicing network.
Fig. 2 shows a flowchart of a 5G slice segmentation-joining method according to another embodiment of the present invention. In this embodiment, the problem of split-join merging of 5G slices is regarded as a node classification (node classification) problem, and finally, the merged split to which each split-joined affected node belongs is output. The slicing network cut-join merger is of a GCNs structure, in the cutting-join scheme, a graph refers to a cut-join topological graph of a slicing network, each node represents a slicing network node, and each edge represents a relation between the slicing network nodes. Using GCNs to extract spatial features of the cutover topology, the goal is to learn a mapping of signals or features on the cutover topology G ═ V, E, with inputs including the slice adjacency matrix a and the slice feature matrix X, and the model will produce a node-level output or a graph-level output.
As shown in fig. 2, the method comprises the steps of:
step S210: training a slice network cut-join combiner, wherein the slice network cut-join combiner is of a graph convolution neural network structure.
In the embodiment of the invention, a slicing network cut-join merger is constructed by a graph convolutional neural network (GCNs) structure, the problem of cut-join merger is converted into the problem of merging nodes in a cut-join topological graph by using the learning advantages of the GCNs on graph structure data, and the slicing network cut-join to be performed is automatically merged.
Specifically, the process of training the slice network cut-join combiner is as follows: collecting cutover topological graph information of a historical slice network as a total data set, marking each piece of data in the data set to obtain a cutover combined information sample, and decomposing cutover topological graph information corresponding to each piece of data to obtain a slice adjacency matrix sample and a slice characteristic matrix sample; taking the slice adjacency matrix sample and the slice characteristic matrix sample as training input data, and taking the corresponding cut-joint and information sample as training output data; and training the graph convolution neural network model by using the training input data and the training output data, and constructing the slice network cut-join combiner according to the model weight when the graph convolution neural network model converges.
Wherein, the marking refers to that M to-be-performed slice network cutting and jointing are combined into K cutting and jointing (K)<M), i.e. classify the cut-affected node into class K. The information of the cutover topological graph is converted into the input of the model, namely a slice adjacency matrix sample A 'and a slice feature matrix sample X', the slice adjacency matrix sample A 'represents the connection relation of each node in the slice network topology, the slice feature matrix sample X' comprises M attribute values, the influence of M slice network cutoffs to be performed correspondingly to the plan on each slice node is achieved, for example, the attribute value of 0 represents that the node is not influenced by the next cutover, and the attribute value of 1 represents that the node is influenced by the next cutover. And, the cutover topology information of the network slice may be represented as G ═ V, E, V is a set of network nodes of the network slice V ═ V1, V2, V3, …, VN }, E is a set of edges, E is a connection between the network node Vi and the network node Vj, and E is a set of edgesij1, otherwise eij＝0。
And, composing the slice adjacency matrix sample A 'and the slice feature matrix sample X' as follows:
the sliced adjacency matrix sample a' is the connection relationship of each node in the sliced network topology,eijrepresenting the connection relation between networking nodes Vi and Vj, wherein the connection between the nodes is 1, the disconnection between the nodes is 0, and the shape of the slice adjacency matrix sample A' is N x N (N is the number of the nodes);
and the slice feature matrix sample X' contains M attribute values, the influence of M slice network cutovers to be performed correspondingly to the plan on each slice node is realized, and the jth cutover of each node i is represented as Xij，xij0 means that the j-th cut does not affect the node i, xij1 represents that the jth cut will affect node i, and the sample X' of the slice feature matrix is a feature matrix with the shape of N × M (N is the number of nodes, and M is the maximum number of cuts to be made in the data set);
the label of each piece of data can be written into a label matrix Y, namely the label matrix Y is the optimal slice cut-joint merging scheme of the manual label and is merged into K cut-joints (K < M), namely, the nodes are classified into K types, the maximum merged cut-joint number in the data set is K, and the shape is N x 1.
Furthermore, the total data set may be divided into training data and test data, and 80% of the entire data set may be taken as training data, and the remaining 20% as test data. And training by using a training set, so that the closer the reconstructed data is to the original data, the better the reconstructed data is, and evaluating the verification model by using a test set.
Fig. 3 shows a block diagram of a slice network cutover combiner in one specific example of the present invention. As shown in fig. 3, a graph network model is built by using a deep learning framework, the model is composed of a plurality of graph convolution layers (GCNs) and a full connection layer (Dense), the graph convolution neural network is used for extracting the node connection relation of the slice network cutover topological graph and the node range influenced by the slice cutover to be performed, and the nodes and edges of the slice topology are projected into a low-dimensional vector space. The full-connection layer finds the relation between the network slicing and cutting and joining scheme of the slice to be performed and the optimal slicing and cutting and joining scheme by learning the slicing and cutting topological vector representation after the spatial features are extracted, and finally outputs the cutting and joining scheme with the minimum influence on the node times and range.
Wherein the first layer is an input layer: inputting a slice adjacency matrix sample A 'and a slice feature matrix sample X';
the second layer is a Graph Conv layer (Graph Conv): the number of convolution kernels is 256, the activation function is set to be 'relu', and the convolution layer is utilized to extract the topological features of the slice network;
the third layer is the Graph convolutional layer (Graph Conv): the number of convolution kernels is 128, and the activation function is set to 'relu';
the fourth layer is a Graph convolutional layer (Graph Conv): the number of convolution kernels is 64, the activation function is set to "lambda", and a potential space vector representation Z of the slice topology nodes and edges is output, wherein Z is GCN (X ', a');
the fifth layer is a fully connected layer (sense): the number of fully connected neurons is 64, the activation function is set to be 'relu', and potential space vector representation Z of the slice network topology is input;
the sixth layer is a fully connected layer (sense): the number of fully connected neurons is 128, the activation function is set to be 'relu', and potential space vector representation Z of the slice network topology is input;
the seventh layer is an output layer and is composed of a full connection layer (Dense): the number of the neurons is set to be K, the K is the maximum cutting number after the slicing cutting and the jointing, and the activation function is set to be sigmoid. And outputting a predicted optimal slice cut-join merging scheme, merging the M to-be-performed slice network cut-joins into K cut-joins (K < M), namely classifying the cut-joins affected nodes into K classes.
An error between the predicted optimal slice-sectioning merge-scheme and the true optimal slice-sectioning merge-scheme is then calculated, the training objective being to minimize the error. The objective function selects a 'binary _ cross' class two logarithmic loss function. The training round number is set to 1000(epochs 1000), and the gradient descent optimization algorithm selects an adam optimizer for improving the learning speed of the conventional gradient descent (optimizer adam'). The neural network can find the optimal weight value which enables the target function to be minimum through gradient descent, and the neural network can learn the weight value automatically through training. Training is performed with a training set so that the smaller the objective function, the better, and the validation model is evaluated with a test set after each round of training. And deriving the weight of the model after the model converges.
In the above step S210, the process of model training is explained in detail, and the online prediction process will be explained below.
Step S220: and acquiring the information of the cutting topological graph to be performed by the slicing network from a network slice management functional module in the slice management architecture.
And automatically acquiring the information of the cutting topological graph to be performed by the slicing network plan within the current nearest preset time period from the NSMF module.
Step S230: decomposing the cutover topological graph information to obtain a slice adjacency matrix and a slice characteristic matrix; the slice adjacency matrix represents the connection relation between the nodes in the cutover topological graph, and the slice characteristic matrix represents the influence attribute of a plurality of cutoffs to be performed on each slice node.
Wherein the slice adjacency matrix A is a matrix of N x N; if the network node V in the topological graph information is cut overiAnd a networking node VjAre connected, then the element e in the adjacent matrix of the sliceijIs a first connection value; and/or if the networking node V in the topological graph information is cut overiAnd a networking node VjAre not connected, then slice the element e in the adjacent matrixijIs the second connection value. For example, the first connection value is 1 and the second connection value is 0.
And the slice feature matrix X is a matrix of N X M; if the network node V in the topological graph information is cut overiAffected by the j-th cut, the element x in the slice feature matrixijIs a first attribute value; and/or if the networking node Vi in the cutover topological graph information is not influenced by the jth cutover, the value of the element xij in the slice feature matrix is the second attribute value. For example, the first attribute value is 1, and the second attribute value is 0.
It should be noted that, the above-mentioned process of obtaining the slice adjacency matrix a and the slice feature matrix X through decomposition, and the composition of the slice adjacency matrix a and the slice feature matrix X, can be referred to the above description of the slice adjacency matrix a 'and the slice feature matrix X' in the training process.
Step S240: and inputting the slice adjacency matrix and the slice characteristic matrix into a trained slice network cut-joint combiner to predict to obtain cut-joint combined information.
Wherein, the cutting number in the cutting topological graph information is M; the number of cutovers in the cutover merging information is K; wherein K is less than M, and K and M are both natural numbers more than 0. In other words, the slice network cut-join combiner can combine M cut-joins to be performed into K cut-joins, and an optimal slice cut-join combining scheme is obtained.
In practical implementation, before inputting the slice adjacency matrix and the slice feature matrix into the slice network cut-join combiner for prediction, the slice adjacency matrix a and the slice feature matrix X may be first transmitted to the data preprocessing module for preprocessing, and processed into a format conforming to the input of the slice network cut-join combiner, for example, the slice network cut-join combiner requires that the input slice feature matrix X is a 50 × 50 matrix, whereas since the number of cuts to be performed is only 48, the actually decomposed slice feature matrix X is a 50 × 48 matrix, the preprocessing may be to complement the insufficient rows and columns in the matrix, in this case, to complement 0 of 50 × 2.
Step S250: and sending the cutting and jointing combined information to a network slice management function module in a slice management architecture for the network slice management function module to perform cutting and jointing processing.
The optimal slice-cutover-merging scheme is fed back to the NSMF, instructing it to operate according to the merged slice-cutover scheme.
Fig. 4 shows the complete flow of the split-join merge method in one specific example of the invention. As shown in fig. 4, the flow is as follows: 1. automatically acquiring the information of a cutover topological graph to be performed by a slice network plan in a recent period of time; 2. converting the topological graph corresponding to the cut topological graph information into a slice adjacency matrix A and a slice characteristic matrix X; 3. inputting the slice adjacency matrix A and the slice characteristic matrix X into a data preprocessing module for matrix preprocessing; 4. inputting the preprocessed matrix into a slicing network cutting and joining combiner; 5. the output of the slicing network cut-joint combiner obtains the optimal slicing cut-joint combination scheme; 6. the optimal slice cutover merge recipe is sent to the NSMF to instruct the merged slice cutover operation.
According to the 5G slice segmentation-joining merging method provided by the embodiment, a historical slice network segmentation topological graph is collected as a total data set, an optimal slice segmentation-joining merging scheme is manually marked on the data set, then the slice segmentation-joining topological graph is converted into model inputs, namely a slice adjacent matrix sample and a slice characteristic matrix sample, the slice adjacent matrix sample is a connection relation of each node in the slice network segmentation topology, the slice characteristic matrix sample comprises M attribute values, and the influence of M slice network segmentations to be performed on each slice node is correspondingly planned; a graph network model is built by using a deep learning framework, the model consists of a plurality of graph convolution layers (GCN) and a full connection layer (Dense), the node connection relation of a slice network cutover topological graph and the node range characteristics influenced by the to-be-sliced cutover are extracted by using a graph convolution neural network, and the nodes and edges of the slice topology are projected into a low-dimensional vector space; the full-connection layer finds the relation between the network slicing and splicing to be performed and the optimal slicing and splicing merging scheme by learning the slicing and splicing topological vector representation after the spatial features are extracted, and finally trains to obtain the slicing and network slicing and splicing merger; when on-line prediction is carried out, the slicing adjacency matrix and the slicing feature matrix which are decomposed in real time are input into the slicing network cut-join combiner, so that a cut-join scheme with the minimum affected node frequency and range can be output, M to-be-carried slicing network cut-joins are combined into K cut-joins (K < M), namely the cut-joined affected nodes are classified into K types, so that the cut-joins of the repeatedly affected nodes are combined, the influence on a slicing user is reduced to the maximum degree, and the cutting-join efficiency of the slicing network is improved.
Fig. 5 shows a schematic structural diagram of a 5G slice cutting and joining device provided by an embodiment of the present invention.
As shown in fig. 5, the apparatus includes:
an obtaining module 510, adapted to obtain information of a topology map to be cut over the slice network;
a decomposition module 520, adapted to decompose the cutover topological graph information to obtain a slice adjacency matrix and a slice feature matrix; the method comprises the following steps that a slice adjacency matrix represents the connection relation between nodes in a cutover topological graph, and a slice characteristic matrix represents the influence attribute of a plurality of cutoffs to be performed on each slice node;
a merging module 530, adapted to input the slice adjacency matrix and the slice feature matrix into a trained slice network cutover merger to predict cutover merging information;
and the cutting and jointing module 540 is suitable for performing cutting and jointing processing according to the cutting and jointing combined information.
In an alternative, the obtaining module is further adapted to: acquiring the information of a cutover topological graph to be performed by a slice network from a network slice management functional module in a slice management architecture;
the cutover module is further adapted to: and sending the cutting and jointing combined information to a network slice management function module in a slice management architecture for the network slice management function module to perform cutting and jointing processing.
In an alternative form, the slice adjacency matrix is a N x N matrix; the decomposition module is further adapted to:
if the network node V in the topological graph information is cut overiAnd a networking node VjAre connected, then the element e in the adjacent matrix of the sliceijIs a first connection value; and/or the presence of a gas in the gas,
if the network node V in the topological graph information is cut overiAnd a networking node VjAre not connected, then slice the element e in the adjacent matrixijIs the second connection value.
In an alternative, the slice feature matrix is a matrix of N × M; the decomposition module is further adapted to:
if the network node V in the topological graph information is cut overiAffected by the j-th cut, the element x in the slice feature matrixijIs a first attribute value; and/or the presence of a gas in the gas,
if the network node V in the topological graph information is cut overiIs not influenced by the j' th cut, the element x in the slice feature matrixijIs the second attribute value.
In an optional manner, the number of cutoffs in the cutover topological graph information is M; the number of cutovers in the cutover merging information is K; wherein K is less than M, and K and M are both natural numbers more than 0.
In an alternative mode, the slice network cut-join merger is a graph convolution neural network structure; the device further comprises: a training module adapted to:
collecting cutover topological graph information of a historical slice network as a total data set, marking each piece of data in the data set to obtain a cutover combined information sample, and decomposing cutover topological graph information corresponding to each piece of data to obtain a slice adjacency matrix sample and a slice characteristic matrix sample;
taking the slice adjacency matrix sample and the slice characteristic matrix sample as training input data, and taking the corresponding cut-joint and information sample as training output data;
and training a graph convolution neural network model by using the training input data and the training output data, and constructing a slice network cut-join combiner according to the model weight when the graph convolution neural network model converges.
An embodiment of the present invention provides a non-volatile computer storage medium, where the computer storage medium stores at least one executable instruction, and the computer executable instruction may execute the method for 5G slice segmentation and joining in any of the above method embodiments.
Fig. 6 is a schematic structural diagram of a computing device according to an embodiment of the present invention, and a specific embodiment of the present invention does not limit a specific implementation of the computing device.
As shown in fig. 6, the computing device may include: a processor (processor)602, a communication Interface 604, a memory 606, and a communication bus 608.
Wherein: the processor 602, communication interface 604, and memory 606 communicate with one another via a communication bus 608. A communication interface 604 for communicating with network elements of other devices, such as clients or other servers. The processor 602 is configured to execute the program 610, and may specifically execute the relevant steps in the above-described embodiment of the method for computing a 5G slice segmentation and joining of a computing device.
In particular, program 610 may include program code comprising computer operating instructions.
The processor 602 may be a central processing unit CPU or an application Specific Integrated circuit asic or one or more Integrated circuits configured to implement embodiments of the present invention. The computing device includes one or more processors, which may be the same type of processor, such as one or more CPUs; or may be different types of processors such as one or more CPUs and one or more ASICs.
And a memory 606 for storing a program 610. Memory 606 may comprise high-speed RAM memory, and may also include non-volatile memory (non-volatile memory), such as at least one disk memory.
The program 610 may specifically be configured to cause the processor 602 to perform the following operations:
acquiring the information of a to-be-cut topological graph of the slicing network;
decomposing the cutover topological graph information to obtain a slice adjacency matrix and a slice characteristic matrix; the method comprises the following steps that a slice adjacency matrix represents the connection relation between nodes in a cutover topological graph, and a slice characteristic matrix represents the influence attribute of a plurality of cutoffs to be performed on each slice node;
inputting the slice adjacency matrix and the slice characteristic matrix into a trained slice network cut-joint combiner to predict to obtain cut-joint combination information;
and performing the cutting processing according to the cutting and jointing combined information.
In an alternative, the program 610 causes the processor 602 to:
acquiring the information of a cutover topological graph to be performed by a slice network from a network slice management functional module in a slice management architecture;
and sending the cutting and jointing combined information to a network slice management function module in a slice management architecture for the network slice management function module to perform cutting and jointing processing.
In an alternative form, the slice adjacency matrix is a N x N matrix;
the program 610 causes the processor 602 to perform the following operations:
if the network node V in the topological graph information is cut overiAnd a networking node VjAre connected, then the element e in the adjacent matrix of the sliceijIs a first connection value; and/or the presence of a gas in the gas,
if the network node V in the topological graph information is cut overiAnd a networking node VjAre not connected, then slice the element e in the adjacent matrixijIs the second connection value.
In an alternative, the slice feature matrix is a matrix of N × M;
the program 610 causes the processor 602 to perform the following operations:
if the network node V in the topological graph information is cut overiAffected by the j-th cut, the element x in the slice feature matrixijIs a first attribute value; and/or the presence of a gas in the gas,
if the network node V in the topological graph information is cut overiIs not influenced by the j' th cut, the element x in the slice feature matrixijIs the second attribute value.
In an optional manner, the number of cutoffs in the cutover topological graph information is M; the number of cutovers in the cutover merging information is K; wherein K is less than M, and K and M are both natural numbers more than 0.
In an alternative mode, the slice network cut-join merger is a graph convolution neural network structure;
the program 610 causes the processor 602 to perform the following operations:
collecting cutover topological graph information of a historical slice network as a total data set, marking each piece of data in the data set to obtain a cutover combined information sample, and decomposing cutover topological graph information corresponding to each piece of data to obtain a slice adjacency matrix sample and a slice characteristic matrix sample;
taking the slice adjacency matrix sample and the slice characteristic matrix sample as training input data, and taking the corresponding cut-joint and information sample as training output data;
and training a graph convolution neural network model by using the training input data and the training output data, and constructing a slice network cut-join combiner according to the model weight when the graph convolution neural network model converges.
The algorithms or displays presented herein are not inherently related to any particular computer, virtual system, or other apparatus. Various general purpose systems may also be used with the teachings herein. The required structure for constructing such a system will be apparent from the description above. In addition, embodiments of the present invention are not directed to any particular programming language. It is appreciated that a variety of programming languages may be used to implement the teachings of embodiments of the present invention as described herein, and any descriptions of specific languages are provided above to disclose the best modes of embodiments of the invention.
In the description provided herein, numerous specific details are set forth. It is understood, however, that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.
Similarly, it should be appreciated that in the foregoing description of exemplary embodiments of the invention, various features of the embodiments of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects. However, the disclosed method should not be interpreted as reflecting an intention that: that is, the claimed embodiments of the invention require more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this invention.
Those skilled in the art will appreciate that the modules in the device in an embodiment may be adaptively changed and disposed in one or more devices different from the embodiment. The modules or units or components of the embodiments may be combined into one module or unit or component, and furthermore they may be divided into a plurality of sub-modules or sub-units or sub-components. All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and all of the processes or elements of any method or apparatus so disclosed, may be combined in any combination, except combinations where at least some of such features and/or processes or elements are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.
Furthermore, those skilled in the art will appreciate that while some embodiments herein include some features included in other embodiments, rather than other features, combinations of features of different embodiments are meant to be within the scope of the invention and form different embodiments. For example, in the following claims, any of the claimed embodiments may be used in any combination.
The various component embodiments of the invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art will appreciate that a microprocessor or Digital Signal Processor (DSP) may be used in practice to implement some or all of the functionality of some or all of the components according to embodiments of the present invention. Embodiments of the invention may also be implemented as apparatus or device programs (e.g., computer programs and computer program products) for performing a portion or all of the methods described herein. Such programs implementing embodiments of the present invention may be stored on computer-readable media or may be in the form of one or more signals. Such a signal may be downloaded from an internet website or provided on a carrier signal or in any other form.
It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. Embodiments of the invention may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The usage of the words first, second and third, etcetera do not indicate any ordering. These words may be interpreted as names. The steps in the above embodiments should not be construed as limiting the order of execution unless specified otherwise.
Claims (10)
1. A method of 5G slice scissoring and merging, comprising:
acquiring the information of a to-be-cut topological graph of the slicing network;
decomposing the cutover topological graph information to obtain a slice adjacency matrix and a slice characteristic matrix; the method comprises the following steps that a slice adjacency matrix represents the connection relation between nodes in a cutover topological graph, and a slice characteristic matrix represents the influence attribute of a plurality of cutoffs to be performed on each slice node;
inputting the slice adjacency matrix and the slice characteristic matrix into a trained slice network cut-joint combiner to predict to obtain cut-joint combination information;
and performing the cutting processing according to the cutting and jointing combined information.
2. The method according to claim 1, wherein the acquiring information of the cutover topology map to be performed by the slice network specifically includes: acquiring the information of a cutover topological graph to be performed by a slice network from a network slice management functional module in a slice management architecture;
the performing of the cutover processing according to the cutover merging information specifically comprises: and sending the cutting and jointing combined information to a network slice management function module in a slice management architecture for the network slice management function module to perform cutting and jointing processing.
3. The method of claim 1 or 2, wherein the slice adjacency matrix is a N x N matrix; the decomposing the cutover topological graph information to obtain a slice adjacency matrix and a slice feature matrix further comprises:
if the network node V in the topological graph information is cut overiAnd a networking node VjAre connected, then the element e in the adjacent matrix of the sliceijIs a first connection value; and/or the presence of a gas in the gas,
if the network node V in the topological graph information is cut overiAnd a networking node VjAre not connected, then slice the element e in the adjacent matrixijIs the second connection value.
4. The method according to claim 1 or 2, wherein the slice feature matrix is a matrix of N x M; the decomposing the cutover topological graph information to obtain a slice adjacency matrix and a slice feature matrix further comprises:
if the network node V in the topological graph information is cut overiAffected by the j-th cut, the element x in the slice feature matrixijIs a first attribute value; and/or the presence of a gas in the gas,
if the network node V in the topological graph information is cut overiIs not influenced by the j' th cut, the element x in the slice feature matrixijIs the second attribute value.
5. The method according to claim 1 or 2, wherein the number of cutovers in the cutover topology map information is M; the number of cutovers in the cutover merging information is K; wherein K is less than M, and K and M are both natural numbers more than 0.
6. The method of claim 1, wherein the slice network cut-join merger is a graph convolution neural network structure; the slicing network cut-join combiner is obtained by training the following steps:
collecting cutover topological graph information of a historical slice network as a total data set, marking each piece of data in the data set to obtain a cutover combined information sample, and decomposing cutover topological graph information corresponding to each piece of data to obtain a slice adjacency matrix sample and a slice characteristic matrix sample;
taking the slice adjacency matrix sample and the slice characteristic matrix sample as training input data, and taking the corresponding cut-joint and information sample as training output data;
and training a graph convolution neural network model by using the training input data and the training output data, and constructing a slice network cut-join combiner according to the model weight when the graph convolution neural network model converges.
7. A 5G slice severance joint and apparatus comprising:
the acquisition module is suitable for acquiring the information of a topological graph to be cut of the slice network;
the decomposition module is suitable for decomposing the cutover topological graph information to obtain a slice adjacency matrix and a slice characteristic matrix; the method comprises the following steps that a slice adjacency matrix represents the connection relation between nodes in a cutover topological graph, and a slice characteristic matrix represents the influence attribute of a plurality of cutoffs to be performed on each slice node;
the merging module is suitable for inputting the slice adjacency matrix and the slice characteristic matrix into a trained slice network cut-join merger to predict and obtain cut-join merging information;
and the cutting module is suitable for performing cutting processing according to the cutting and jointing combined information.
8. The apparatus of claim 7, wherein the acquisition module is further adapted to: acquiring the information of a cutover topological graph to be performed by a slice network from a network slice management functional module in a slice management architecture;
the cutover module is further adapted to: and sending the cutting and jointing combined information to a network slice management function module in a slice management architecture for the network slice management function module to perform cutting and jointing processing.
9. A computing device, comprising: the system comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete mutual communication through the communication bus;
the memory is configured to store at least one executable instruction that causes the processor to perform operations corresponding to the method of 5G slice segmentation-joining according to any one of claims 1-6.
10. A computer storage medium having stored therein at least one executable instruction for causing a processor to perform operations for 5G slice segmentation-joining and method correspondence as recited in any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010186555.5A CN113411841B (en) | 2020-03-17 | 2020-03-17 | 5G slice cutting and joining method and device and computing equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010186555.5A CN113411841B (en) | 2020-03-17 | 2020-03-17 | 5G slice cutting and joining method and device and computing equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113411841A true CN113411841A (en) | 2021-09-17 |
| CN113411841B CN113411841B (en) | 2022-08-02 |
Family
ID=77677119
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010186555.5A Active CN113411841B (en) | 2020-03-17 | 2020-03-17 | 5G slice cutting and joining method and device and computing equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113411841B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115696466A (en) * | 2022-09-23 | 2023-02-03 | 广州爱浦路网络技术有限公司 | S-NSSAI priority determination method and device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190021010A1 (en) * | 2017-07-05 | 2019-01-17 | Huawei Technologies Co., Ltd. | Methods and systems for network slicing |
| CN109743286A (en) * | 2018-11-29 | 2019-05-10 | 武汉极意网络科技有限公司 | A kind of IP type mark method and apparatus based on figure convolutional neural networks |
| CN110276406A (en) * | 2019-06-26 | 2019-09-24 | 腾讯科技（深圳）有限公司 | Expression classification method, apparatus, computer equipment and storage medium |
| US20200052991A1 (en) * | 2018-08-09 | 2020-02-13 | At&T Intellectual Property I, L.P. | Mobility network slice selection |
-
2020
- 2020-03-17 CN CN202010186555.5A patent/CN113411841B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190021010A1 (en) * | 2017-07-05 | 2019-01-17 | Huawei Technologies Co., Ltd. | Methods and systems for network slicing |
| US20200052991A1 (en) * | 2018-08-09 | 2020-02-13 | At&T Intellectual Property I, L.P. | Mobility network slice selection |
| CN109743286A (en) * | 2018-11-29 | 2019-05-10 | 武汉极意网络科技有限公司 | A kind of IP type mark method and apparatus based on figure convolutional neural networks |
| CN110276406A (en) * | 2019-06-26 | 2019-09-24 | 腾讯科技（深圳）有限公司 | Expression classification method, apparatus, computer equipment and storage medium |
Non-Patent Citations (3)
| Title |
|---|
| ZTE: ""view on Nextgen Multi-service Core"", 《3GPP TSG_SA\WG2_ARCH，S2-153139》 * |
| 匿名: "图卷积网络定义和简单示例详解", 《WWW.ELECFANS.COM》 * |
| 周恒等: "一种5G网络切片的编排算法", 《电信科学》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115696466A (en) * | 2022-09-23 | 2023-02-03 | 广州爱浦路网络技术有限公司 | S-NSSAI priority determination method and device |
| CN115696466B (en) * | 2022-09-23 | 2023-06-06 | 广州爱浦路网络技术有限公司 | Method and device for determining priority of S-NSSAI |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113411841B (en) | 2022-08-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113541980B (en) | Network slice fault root cause positioning method, device, computing equipment and storage medium | |
| CN108306804A (en) | A kind of Ethercat main station controllers and its communication means and system | |
| CN110489749B (en) | Business process optimization method of intelligent office automation system | |
| CN109743286A (en) | A kind of IP type mark method and apparatus based on figure convolutional neural networks | |
| CN114139626A (en) | Power communication fault analysis method and device, terminal equipment and medium | |
| Zhang et al. | A spatial fuzzy influence diagram for modelling spatial objects’ dependencies: A case study on tree-related electric outages | |
| CN113411841B (en) | 5G slice cutting and joining method and device and computing equipment | |
| CN113541986B (en) | Fault prediction method and device for 5G slice and computing equipment | |
| CN113765683B (en) | Cutover alarm shielding method and device for network slices | |
| CN113095515A (en) | Service fault information processing method and device | |
| CN102870090A (en) | Process entity graphs | |
| CN113747480B (en) | Processing method and device for 5G slice faults and computing equipment | |
| CN113806498A (en) | Online question-answering method and device for network slice instance running state | |
| CN113762978B (en) | Complaint delimiting method and device for 5G slicing user and computing equipment | |
| CN115660245A (en) | Service arrangement method and device, electronic equipment and storage medium | |
| CN113297289A (en) | Method and device for extracting business data from database and electronic equipment | |
| CN112990425A (en) | Automatic classification method of 5G network slices, device thereof, electronic equipment and computer storage medium | |
| Pankov et al. | Model studies of systems with diagnostics based on fault simulation | |
| CN113766532B (en) | Isolation evaluation method and device for network slice instance | |
| CN114500227B (en) | Alarm analysis method, device, equipment and computer storage medium | |
| CN113825151B (en) | Method and device for predicting relationship between slice alarms based on relationship graph convolution network | |
| CN114143163B (en) | Slice false alarm identification method and device based on graph attention network | |
| CN117609870B (en) | Structure recognition model training, model structure recognition method, device and medium | |
| CN117350549B (en) | Distribution network voltage risk identification method, device and equipment considering output correlation | |
| CN114079958B (en) | Disaster recovery pool selection method and device for 5G slice instance and computing equipment |
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 |