CN114071549A - 5G base station cluster KPI prediction method and system based on multi-reservoir fuzzy cognitive map - Google Patents
5G base station cluster KPI prediction method and system based on multi-reservoir fuzzy cognitive map Download PDFInfo
- Publication number
- CN114071549A CN114071549A CN202111353692.4A CN202111353692A CN114071549A CN 114071549 A CN114071549 A CN 114071549A CN 202111353692 A CN202111353692 A CN 202111353692A CN 114071549 A CN114071549 A CN 114071549A
- Authority
- CN
- China
- Prior art keywords
- kpi
- base station
- fuzzy
- cognitive map
- reservoir
- 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.)
- Pending
Links
- 230000001149 cognitive effect Effects 0.000 title claims abstract description 64
- 238000000034 method Methods 0.000 title claims abstract description 40
- 230000001364 causal effect Effects 0.000 claims abstract description 12
- 238000007781 pre-processing Methods 0.000 claims abstract description 10
- 230000006870 function Effects 0.000 claims description 19
- 239000011159 matrix material Substances 0.000 claims description 17
- 238000004590 computer program Methods 0.000 claims description 14
- 230000008569 process Effects 0.000 claims description 12
- 210000002569 neuron Anatomy 0.000 claims description 9
- 238000003860 storage Methods 0.000 claims description 9
- 239000000126 substance Substances 0.000 claims description 9
- 230000004913 activation Effects 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 3
- 210000005036 nerve Anatomy 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000012417 linear regression Methods 0.000 claims description 2
- 238000013507 mapping Methods 0.000 claims description 2
- 150000001875 compounds Chemical class 0.000 claims 2
- 230000000630 rising effect Effects 0.000 claims 2
- 238000004891 communication Methods 0.000 abstract description 2
- 208000018910 keratinopathic ichthyosis Diseases 0.000 description 50
- 238000010586 diagram Methods 0.000 description 9
- 238000012545 processing Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 229910052918 calcium silicate Inorganic materials 0.000 description 1
- 235000012241 calcium silicate Nutrition 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 210000002364 input neuron Anatomy 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 210000004205 output neuron Anatomy 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/08—Testing, supervising or monitoring using real traffic
-
- 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
- G06N3/043—Architecture, e.g. interconnection topology based on fuzzy logic, fuzzy membership or fuzzy inference, e.g. adaptive neuro-fuzzy inference systems [ANFIS]
Abstract
The invention belongs to the technical field of communication, and provides a 5G base station cluster KPI prediction method and a system based on a multi-reservoir fuzzy cognitive map, wherein the method comprises the following steps: acquiring KPI original sequence data of all base stations in a 5G base station cluster; preprocessing KPI original sequence data to obtain a KPI dynamic time sequence of a base station network; based on the fuzzy cognitive map model, corresponding the dynamic time sequence to concept nodes in the fuzzy cognitive map; on the basis of an original reasoning structure with fuzzy feedback, adding fuzzy feedback, replacing each concept node of the concept nodes in the fuzzy cognitive map with a back-rising state network to obtain a multi-reservoir fuzzy cognitive map model, and obtaining the state characteristics of the KPI of each base station by applying multi-library learning; and based on the state characteristics of the KPI, obtaining the KPI at the next moment by adopting iterative reasoning of fuzzy causal relationship dynamics.
Description
Technical Field
The invention belongs to the technical field of communication, and particularly relates to a 5G base station cluster KPI prediction method and system based on a multi-reservoir fuzzy cognitive map.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
Compared with 2G, 3G and 4G networks, the frequency band of the 5G network is higher, the loss is large in the transmission process, more intensive base stations need to be built, the KPIs of the base station network have important significance for assistant decision-making by mining rules behind the KPIs, the change trend of a certain area on key indexes can be known in advance to a certain extent through prediction of the KPIs of the base station network of the area, and then operators can be assisted to establish reasonable emergency strategies.
The problems existing in the prior art are as follows:
the time series data prediction models commonly used at present are DNN, LSTM, GRU and their variants and the like, the models have more parameters and longer training time, and the interpretability of the models is covered by the black box characteristics. Therefore, how to implement a model with high performance prediction while maintaining interpretability is a problem for predicting KPIs of 5G base station clusters.
Disclosure of Invention
In order to solve at least one technical problem existing in the background technology, the invention provides a load-bearing network flow prediction method and a load-bearing network flow prediction system based on a Gramami angular field.
In order to achieve the purpose, the invention adopts the following technical scheme:
the invention provides a 5G base station cluster KPI prediction method based on a multi-reservoir fuzzy cognitive map, which comprises the following steps:
acquiring KPI original sequence data of all base stations in a 5G base station cluster;
preprocessing KPI original sequence data to obtain a KPI dynamic time sequence of a base station network;
based on the fuzzy cognitive map model, corresponding the dynamic time sequence to concept nodes in the fuzzy cognitive map;
on the basis of an original reasoning structure with fuzzy feedback, adding fuzzy feedback, replacing each concept node of the concept nodes in the fuzzy cognitive map with a back-rising state network to obtain a multi-reservoir fuzzy cognitive map model, and obtaining the state characteristics of the KPI of each base station by applying multi-library learning;
and based on the state characteristics of the KPI, obtaining the KPI at the next moment by adopting iterative reasoning of fuzzy causal relationship dynamics.
The second aspect of the invention provides a 5G base station cluster KPI prediction system based on a multi-reservoir fuzzy cognitive map, which comprises:
a data acquisition module configured to: acquiring KPI original sequence data of all base stations in a 5G base station cluster;
a data pre-processing module configured to: preprocessing KPI original sequence data to obtain a KPI dynamic time sequence of a base station network;
a state feature extraction module configured to: based on the fuzzy cognitive map model, corresponding the dynamic time sequence to concept nodes in the fuzzy cognitive map;
on the basis of an original reasoning structure with fuzzy feedback, adding fuzzy feedback, replacing each concept node of the concept nodes in the fuzzy cognitive map with a back-rising state network to obtain a multi-reservoir fuzzy cognitive map model, and obtaining the state characteristics of the KPI of each base station by applying multi-library learning;
a KPI prediction module of a base station configured to: and based on the state characteristics of the KPI, obtaining the KPI of the base station at the next moment by adopting iterative reasoning of fuzzy causal relationship dynamics.
A third aspect of the invention provides a computer-readable storage medium.
A computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps in the multi-reservoir fuzzy cognitive map based 5G base station cluster KPI prediction method as described above.
A fourth aspect of the invention provides a computer apparatus.
A computer apparatus comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor when executing the program implementing the steps in the multi-reservoir fuzzy cognitive map based 5G base station cluster KPI prediction method as described above.
Compared with the prior art, the invention has the beneficial effects that:
on the basis of an original reasoning structure with fuzzy feedback, each concept node in the graph is replaced by an ESN module, multi-library calculation is applied, the state characteristics of a certain KPI in the 5G network are learned, the capability of the ESN for processing a large-scale non-stationary time sequence and the advantages of the FCM in the reasoning and interpretation aspect are fully utilized, the prediction of the KPI in the 5G base station network is realized, and the added ESN enables the provided model to have higher prediction accuracy. The 5G base station cluster in a certain area can be regarded as a graph structure, which is very similar to the structure of the FCM, and by the technology, the high-precision interpretable prediction of the KPI of the 5G base station cluster can be realized.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the invention and together with the description serve to explain the invention and not to limit the invention.
FIG. 1 is an overall flow diagram of a prediction method based on an MR-FCM model;
fig. 2 is an echo state network structure.
Detailed Description
The invention is further described with reference to the following figures and examples.
It is to be understood that the following detailed description is exemplary and is intended to provide further explanation of the invention as claimed. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the invention. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
Example one
Interpretation of terms:
KPI: (Key Performance Indicator), a base station network cluster generates a large amount of data at every moment, such as user experience rate, connection number density, end-to-end delay, mobility, traffic density, and the like, and these data are referred to as KPI of the base station network.
In the FCM, a fuzzy cognitive graph is a weighted directed graph which is composed of concept nodes, state values and relations and has n concept nodes, fuzzy logic and a neural network are combined to form a strong model for system state prediction and interpretable knowledge representation, in the fuzzy cognitive graph with n nodes, each node represents a concept in a system, the concept can be events, targets, trends and the like of the system, each concept describes the attribute of the concept through the state values, causal influence relations among the concepts are represented by directed arcs, and the size of the causal influence degree is reflected by a weight value.
The ESN is an echo state network, is one of library computing systems, is simple to train, and has good prediction capability on a nonlinear or non-Gaussian dynamic system. ESNs form a hidden layer of the network by randomly placing large scale sparsely connected neurons, commonly referred to as "pools".
The reserve pool has the following characteristics: 1) contains a relatively large number of neurons compared to classical neural networks; 2) randomly generating connection relations among the neurons; 3) the connections between neurons are sparse.
As shown in fig. 1, the embodiment provides a 5G base station cluster KPI prediction method based on a multi-reservoir fuzzy cognitive map, which includes the following steps:
s1, acquiring KPI original sequence data of all base station networks in the 5G base station cluster;
s2, preprocessing the KPI original sequence data to obtain KPI dynamic time sequence of the base station network
S3, based on the fuzzy cognitive map model, the dynamic time sequence is corresponding to the concept node in the fuzzy cognitive map;
and for a certain KPI, modeling the base station network into an FCM, wherein the KPI of each base station in the base station network corresponds to a concept node.
S4, adding fuzzy feedback on the basis of an original reasoning structure with the fuzzy feedback, replacing each concept node of the concept nodes in the fuzzy cognitive map with a back-rising state network to obtain a multi-reservoir fuzzy cognitive map model, and obtaining the state characteristics of the KPI of each base station network by applying multi-library learning;
and S5, obtaining the KPI at the next moment by adopting iterative reasoning of fuzzy causal relationship dynamics based on the state characteristics of the KPI.
In S2, the process of obtaining the KPI dynamic time series includes:
the KPI indicator of each base station network obtained from S1 is normalized by using the maximum and minimum normalization method, taking the indicator of the user experience rate as an example, the obtained original user experience rate is:
after normalization, the method comprises the following steps:
wherein x isk(T) is the user experience rate of the kth base station in the base station cluster for time length T, K being 1,2, …, K; the maximum and minimum normalized formula is
xmin，xmaxAre each xkMinimum and maximum values among (T).
The 5G base station cluster in a certain area can be regarded as a graph structure, which is similar to the structure of the fuzzy cognitive map model, so that the 5G base station cluster is analogized to the fuzzy cognitive map model.
In S4, replacing each concept node of the concept nodes in the fuzzy cognitive map with a back-rise state network to obtain a multi-reservoir fuzzy cognitive map model includes:
the construction formula of the fuzzy cognitive map model is as follows:
where f (-) is a transformation function that acts to control the state value of the node to [0,1 ].
Wherein i and j denote the ith base station network and the jth base station network, and the concept node of the ith base station network is represented by ciIndicating that the weight on the connection side of the ith base station network node and the jth base station network node is wijIs shown in the specification, wherein wij∈[-1,1]。
Concept node c of ith base station networkiConcept node c with jth base station networkjThere are three relationships between:
when w isij>At 0, ciAnd cjIs in positive correlation; when w isij<At 0, ciAnd cjIs in negative correlation;
when w isijWhen equal to 0, ciAnd cjNo cause-effect relationship exists; each concept node ciHaving a state value a at different timesi∈[0,1]。
Concept node c of ith base station networkiIs a dynamic time sequence
Wherein
Concept node c representing ith base station networkiThe state value at time t.
Concept node c of ith base station networkiThe state value at time t +1 depends on node ciAt time tState value added with node ciAnd multiplying the state value of the concept node of the other connected base station network at the time t by the sum of the corresponding weights.
The multi-reservoir fuzzy cognitive map model comprises the following steps:
taking the user experience rate as an example, after the replacement, the formula (1) is changed into the following formula (2)
Wherein the content of the first and second substances,
is a membership function which functions to control the output value to 0,1]ESN ofj(. h) is the calculation of the jth ESN module, wjkIs the weight on the edge connected to the kth ESN module.
For the input of the jth ESN module at time t,
the state updating formula of the jth ESN module and the reserve pool is as follows:
wherein the content of the first and second substances,
respectively an input weight matrix and a reserve pool weight matrix of the jth ESN module,
is the state of the reserve pool at time t, and f is the activation function of the neurons in the reserve pool; the output of the jth ESN module at time (t +1) is:
wherein the content of the first and second substances,
is the output weight matrix of the jth ESN module,
is the output value at time t, foutIs an activation function of the output nerve. Thus, the state characteristic of the index of the user experience rate of each base station is learned through each ESN module.
Further, in the present invention, the number M of input neurons of each ESN network may be set to 1, the number N of reserve pool neurons may be set to 200, and the number P of output neurons may be set to 1.
As shown in fig. 2, wherein the state network comprises an input layer, a reserve pool, and an output layer;
mapping the input signal u (t) from a low-dimensional input space to a high-dimensional state space, and adopting a linear regression method to output the weight W of the network in the high-dimensional state spaceoutTraining is performed while inputting the weight WinReserve pool weight WresFeedback connection weight WbackAnd the training time of the network is kept unchanged.
The network of the fly-back state comprises M input units, the input at time t is
N internal units in the state of
P output units, the output at time t being
Wherein [. ]]TRepresenting a transpose operation of the matrix.
The update of the pool internal state vector is shown in equation (3):
x(t+1)＝f[Winu(t+1)+Wresx(t)] (3)
wherein the content of the first and second substances,
representing the value of the ESNs internal state vector at time t,
respectively representing an input connection weight matrix, a reserve pool connection weight matrix and an output connection weight matrix. f ═ f1,f2,…,fN]Representing the activation function of neurons within the reservoir, in general fi(i ═ 1,2, …, N) takes the hyperbolic tangent function, as shown in equation (4):
the output of the entire network is shown in equation (5):
y(t+1)＝fout[Wout(x(t+1),y(t))], (5)
wherein f isout＝(f1 out,…,fP out) Is a linear function, (x (t +1), y (t)) is the concatenation of the input, reserve, previous output weight matrices.
Therefore, the state characteristic of the index of the user experience rate of each base station is learned through each ESN module, the process learns new characteristics through an output layer of the ESN module, the state updating of the reserve pool is only the influence of input values on the reserve pool, the state of the reserve pool is updated, but what is needed is that the value of the output layer is used as an intermediate state, namely the KPI state characteristic is learned. The purpose of replacing the original node by the ESN module is to make up for the defect that the FCM cannot model the long-term dependency relationship.
In S5, obtaining the KPI at the next time by using iterative inference of fuzzy causal relationship dynamics based on the state characteristics of the KPI includes:
wherein the ESNjThe calculation procedure of (-) is as follows:
1) the input of the jth ESN at time t is shown in equation (6):
2) the status update of the reserve pool of the jth ESN at time (t +1) is shown in equation (7):
wherein the content of the first and second substances,
respectively an input weight matrix and a reserve pool weight matrix of the jth ESN module,
is the state of the reserve pool at time t, and f is the activation function of the neurons in the reserve pool;
3) the output of the jth ESN at time (t +1) is shown in equation (8):
wherein the content of the first and second substances,
is the output weight matrix of the jth ESN module,
is the output value at time t, foutIs an activation function of the output nerve.
Equation (2) can be expressed in the form of a vector as follows:
wherein the content of the first and second substances,
is the state value of all node modules of the multi-reservoir fuzzy cognitive map at the time t,
is the state value, W, of the kth node of the multi-reservoir fuzzy cognitive map at the moment (t +1)k＝{w1k,…,wkk,…,wKkIs the kth column of the weight matrix W. After the inverse transformation of equation (9), it can be expressed as:
the learning problem of the multi-reservoir fuzzy cognitive map model weight can be converted into the solution of least squares, and in order to solve WkThe following objective function is constructed:
the optimization of the objective function adopts the Adam algorithm. After the model is trained, a weight matrix W is obtained, and the user experience rate of the base station of the 5G base station network at the next moment can be solved through the known updating rules of W and MR-FCM. By predicting the experience rate of a base station network user in a certain area, the change trend of the experience rate of the user in the area can be obtained in advance to a certain extent, and operators can be assisted to establish a reasonable emergency strategy.
The optimization process of the objective function adopts an Adam algorithm, and the specific optimization process comprises the following steps:
in this embodiment, taking the prediction of the user experience rate as an example, the overall idea is as follows:
first, the user experience for each base station in the cluster of base stationsRate indicator
Is pretreated to obtain
Secondly, the preprocessed data pass through the ESN (-) calculation process of the ESN module to obtain the state characteristics of the user experience rate. Finally, an iterative formulation by fuzzy causal dynamics
And obtaining the user experience rate of each base station at the next moment.
Considering the difference between the use of holiday traffic and the use of user traffic during non-holiday, two measurement indexes, namely growth entropy and standard deviation, are adopted to measure the difference, then the Granami angular field sequence imaging technology is utilized to carry out dimension-increasing processing on a one-dimensional network traffic sequence, and a processed picture is input to a ConvLSTM-F layer, so that the characteristics of the original sequence in time and space can be extracted, and the prediction of the intelligent traffic of the bearer network is realized. In the prediction model, due to the use of the ConvLSTM-F module, not only can the time-aspect information of the sequence be extracted, but also the spatial information can be further extracted, so that the prediction precision of the model is improved while the input information is enriched.
Example two
The embodiment provides a 5G base station cluster KPI prediction system based on a multi-reservoir fuzzy cognitive map, which includes:
a data acquisition module configured to: acquiring KPI original sequence data of all base stations in a 5G base station cluster;
a data pre-processing module configured to: preprocessing KPI original sequence data to obtain a KPI dynamic time sequence of a base station network;
a state feature extraction module configured to: based on the fuzzy cognitive map model, corresponding the dynamic time sequence to concept nodes in the fuzzy cognitive map;
on the basis of an original reasoning structure with fuzzy feedback, adding fuzzy feedback, replacing each concept node of the concept nodes in the fuzzy cognitive map with a back-rising state network to obtain a multi-reservoir fuzzy cognitive map model, and obtaining the state characteristics of the KPI of each base station by applying multi-library learning;
a KPI prediction module of a base station configured to: and based on the state characteristics of the KPI, obtaining the KPI of the base station at the next moment by adopting iterative reasoning of fuzzy causal relationship dynamics.
EXAMPLE III
The present embodiment provides a computer-readable storage medium, on which a computer program is stored, which when executed by a processor implements the steps in the method for predicting the network traffic of a bearer network based on the grassplot angle field as described above.
Example four
The embodiment provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor executes the program to implement the steps in the method for predicting the network traffic of the carrier network based on the grammite angular field.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. The 5G base station cluster KPI prediction method based on the multi-reservoir fuzzy cognitive map is characterized by comprising the following steps:
acquiring KPI original sequence data of all base station networks in a 5G base station cluster;
preprocessing KPI original sequence data to obtain a KPI dynamic time sequence of a base station network;
based on the fuzzy cognitive map model, corresponding the dynamic time sequence to concept nodes in the fuzzy cognitive map;
on the basis of an original reasoning structure with fuzzy feedback, adding fuzzy feedback, replacing each concept node of the concept nodes in the fuzzy cognitive map with a back-rising state network to obtain a multi-reservoir fuzzy cognitive map model, and obtaining the state characteristics of the KPI of each base station by applying multi-library learning;
and based on the state characteristics of the KPI, obtaining the KPI at the next moment by adopting iterative reasoning of fuzzy causal relationship dynamics.
2. The multi-reservoir fuzzy cognitive map based 5G base station cluster KPI prediction method according to claim 1, wherein said KPI dynamic time series obtaining process comprises: and normalizing the acquired KPI of each base station by using a maximum and minimum normalization method.
3. The multi-reservoir fuzzy cognitive map-based 5G base station cluster KPI prediction method of claim 1, wherein the multi-reservoir fuzzy cognitive map model is:
in the formula (I), the compound is shown in the specification,
is a function of degree of membership, ESNj(. h) is the calculation of the jth network module in the fallback state, wjkIs the weight on the edge connected to the kth state-rising network module,
the j is the input of the network module in the rising state at the time t, and the time t is the current time.
4. The multi-reservoir fuzzy cognitive map based 5G base station cluster KPI prediction method according to claim 1, wherein said obtaining KPI at next time by iterative inference using fuzzy causal relation dynamics is solved by converting learning problem of multi-reservoir fuzzy cognitive map model weight into least squares.
5. The multi-reservoir fuzzy cognitive map based 5G base station cluster KPI prediction method of claim 1, wherein the learning problem of the multi-reservoir fuzzy cognitive map model weight is expressed as:
wherein D is(t)Is the state value of all node modules of the multi-reservoir fuzzy cognitive map at the time t,
is the state value, W, of the kth node of the multi-reservoir fuzzy cognitive map at the moment t +1kIs the kth column of the weight matrix W.
6. The multi-reservoir fuzzy cognitive map-based 5G base station cluster KPI prediction method according to claim 1, wherein the calculation process of said back-rise state network module is as follows:
wherein the content of the first and second substances,
is the output of the jth network module in the back-up stateThe weight matrix is output, and the weight matrix is output,
is the output value at time t, foutIs an activation function of the output nerve;
wherein the content of the first and second substances,
in the formula (I), the compound is shown in the specification,
respectively an input weight matrix and a reserve pool weight matrix of the jth network module in the state of rising back,
is the state of the reservoir at time t, and f is the activation function of the neurons in the reservoir.
7. The multi-reservoir fuzzy cognitive map based 5G base station cluster KPI prediction method of claim 1, wherein said back-rise state network comprises an input layer, a reserve pool and an output layer; the training process of the network with the state of the rollback comprises the following steps: and mapping the input signal from a low-dimensional input space to a high-dimensional state space, training the output weight of the network in the high-dimensional state space by adopting a linear regression method, and keeping the input weight, the reserve pool weight and the feedback connection weight unchanged in the network training process.
8. 5G base station cluster KPI prediction system based on many reservoirs fuzzy cognitive map, characterized by, including:
a data acquisition module configured to: acquiring KPI original sequence data of all base station networks in a 5G base station cluster;
a data pre-processing module configured to: preprocessing KPI original sequence data to obtain a KPI dynamic time sequence of a base station network;
a state feature extraction module configured to: based on the fuzzy cognitive map model, corresponding the dynamic time sequence to concept nodes in the fuzzy cognitive map;
on the basis of an original reasoning structure with fuzzy feedback, adding fuzzy feedback, replacing each concept node of the concept nodes in the fuzzy cognitive map with a back-rising state network to obtain a multi-reservoir fuzzy cognitive map model, and obtaining the state characteristics of the KPI of each base station by applying multi-library learning;
a KPI prediction module of a base station configured to: and based on the state characteristics of the KPI, obtaining the KPI of the base station at the next moment by adopting iterative reasoning of fuzzy causal relationship dynamics.
9. A computer readable storage medium, having stored thereon a computer program, which when executed by a processor performs the steps in the multi-reservoir fuzzy cognitive map based 5G base station cluster KPI prediction method according to any of claims 1-7.
10. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program implements the steps in the multi-reservoir fuzzy cognitive map based 5G base station cluster KPI prediction method according to any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111353692.4A CN114071549A (en) | 2021-11-16 | 2021-11-16 | 5G base station cluster KPI prediction method and system based on multi-reservoir fuzzy cognitive map |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111353692.4A CN114071549A (en) | 2021-11-16 | 2021-11-16 | 5G base station cluster KPI prediction method and system based on multi-reservoir fuzzy cognitive map |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114071549A true CN114071549A (en) | 2022-02-18 |
Family
ID=80273439
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111353692.4A Pending CN114071549A (en) | 2021-11-16 | 2021-11-16 | 5G base station cluster KPI prediction method and system based on multi-reservoir fuzzy cognitive map |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114071549A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103906135A (en) * | 2012-12-28 | 2014-07-02 | 中国科学院声学研究所 | P2P node selection method and system used in cellular network |
| CN109492760A (en) * | 2018-12-21 | 2019-03-19 | 北京航空航天大学 | Interpretable prediction technique based on Depth Blur cognition graph model |
| CN110187636A (en) * | 2019-05-30 | 2019-08-30 | 济南大学 | Method for building up, device and the application of model suitable for cement pre-grinding process |
| CN113365298A (en) * | 2021-04-22 | 2021-09-07 | 山东师范大学 | MR network signal intensity prediction method and system based on high-order fuzzy cognitive map |
-
2021
- 2021-11-16 CN CN202111353692.4A patent/CN114071549A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103906135A (en) * | 2012-12-28 | 2014-07-02 | 中国科学院声学研究所 | P2P node selection method and system used in cellular network |
| CN109492760A (en) * | 2018-12-21 | 2019-03-19 | 北京航空航天大学 | Interpretable prediction technique based on Depth Blur cognition graph model |
| CN110187636A (en) * | 2019-05-30 | 2019-08-30 | 济南大学 | Method for building up, device and the application of model suitable for cement pre-grinding process |
| CN113365298A (en) * | 2021-04-22 | 2021-09-07 | 山东师范大学 | MR network signal intensity prediction method and system based on high-order fuzzy cognitive map |
Non-Patent Citations (1)
| Title |
|---|
| 刘灿娜: "多维时间序列的图表示学习及预测研究", 《中国优秀硕士学位论文全文数据库》, 15 February 2024 (2024-02-15) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109102126B (en) | Theoretical line loss rate prediction model based on deep migration learning | |
| Basirat et al. | The quest for the golden activation function | |
| CN111581343A (en) | Reinforced learning knowledge graph reasoning method and device based on graph convolution neural network | |
| CN106529818B (en) | Water quality assessment Forecasting Methodology based on Fuzzy Wavelet Network | |
| CN104732274A (en) | Intelligent computer | |
| Ahmadi et al. | Learning fuzzy cognitive maps using imperialist competitive algorithm | |
| CN109543043A (en) | A kind of automatic monitoring method of the power communication big data mistake of knowledge based map reasoning | |
| CN107451230A (en) | A kind of answering method and question answering system | |
| CN109934422A (en) | Neural network wind speed prediction method based on time series data analysis | |
| CN112381320A (en) | Community personnel flow prediction method, system, storage medium and equipment | |
| CN112766603A (en) | Traffic flow prediction method, system, computer device and storage medium | |
| CN112434813A (en) | Multiple multidimensional language reasoning method based on attribute-oriented language concept lattice | |
| CN111722139A (en) | Lithium battery health monitoring model self-learning method based on micro-continuous mapping | |
| US6317730B1 (en) | Method for optimizing a set of fuzzy rules using a computer | |
| Handoyo et al. | The fuzzy inference system with rule bases generated by using the fuzzy C-means to predict regional minimum wage in Indonesia | |
| Yao et al. | A stability criterion for discrete-time fractional-order echo state network and its application | |
| Zhang et al. | Hybrid intelligence utilization for construction site layout | |
| CN112785051A (en) | Cloud resource prediction method based on combination of EMD and TCN | |
| CN117116048A (en) | Knowledge-driven traffic prediction method based on knowledge representation model and graph neural network | |
| CN114071549A (en) | 5G base station cluster KPI prediction method and system based on multi-reservoir fuzzy cognitive map | |
| Kang et al. | Generalized fuzzy cognitive maps: a new extension of fuzzy cognitive maps | |
| CN114881308A (en) | Internet vehicle speed prediction method based on meta-learning | |
| Wan et al. | Deep learning-based management for wastewater treatment plants under blockchain environment | |
| Chang et al. | Developing a prototype satellite-based cyber-physical system for smart wastewater treatment | |
| Muhamediyeva et al. | Methods of Obtaining Z-solutions of a Multi-criterial Problem |
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 |