CN115622894A - Dynamic network topology prediction method and system based on historical data probability analysis - Google Patents
Dynamic network topology prediction method and system based on historical data probability analysis Download PDFInfo
- Publication number
- CN115622894A CN115622894A CN202211629149.7A CN202211629149A CN115622894A CN 115622894 A CN115622894 A CN 115622894A CN 202211629149 A CN202211629149 A CN 202211629149A CN 115622894 A CN115622894 A CN 115622894A
- Authority
- CN
- China
- Prior art keywords
- entity
- prediction
- probability
- network
- historical
- 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
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3346—Query execution using probabilistic model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/216—Parsing using statistical methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
- G06F40/295—Named entity recognition
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/142—Network analysis or design using statistical or mathematical methods
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/147—Network analysis or design for predicting network behaviour
Abstract
The invention discloses a dynamic network topology prediction method and system based on historical data probability analysis, and relates to the technical field of data processing. In the method, historical detection data of each entity in a network is obtained to form an index matrix of each entity in the network; training a predictive probability model of each entity based on the index matrix to determine a predictive probability model parameter matrix of each entity; and predicting the active state of each entity at the prediction time according to the externally input prediction time and the prediction probability model parameter matrix, and establishing a network topology view of the entity of the network in the active state at the prediction time based on the attribute value of the entity in the active state.
Description
Technical Field
The invention belongs to the technical field of data processing, and particularly relates to a dynamic network topology prediction method and system based on historical data probability analysis.
Background
The operation planning of the network is to make a preset development of related network behaviors or events at a certain future moment, and the basis for developing the network planning is to construct an effective network topology view. The network is complex and changeable, and the states of various network elements (network entities and their connection relations) in the network have randomness and uncertainty. The detection of the network is based on the perception of the current network state, and the perception of a relatively finished network requires a certain time period, so that the conventional network topology view is the fusion analysis result of the historical detection data of the network, and the displayed network topology is the result of the accumulation of the detection data in the historical time period.
Therefore, it is not time efficient to develop network planning based on the known network topology. The method is based on the existing historical network topology state data, performs network topology prediction at the future moment, and has important significance for improving timeliness of decisions such as network planning and management. Therefore, how to predict the network topology at a future time based on the existing historical network topology state data is a technical problem to be solved urgently in the field.
Disclosure of Invention
In order to solve the technical problem, the invention provides a dynamic network topology prediction scheme based on historical data probability analysis, and the scheme aims at a time-varying network dynamically evolved by time domain regularity; through statistical analysis of historical data of network perception, a corresponding entity state probability prediction model is established, a network topology prediction data set at a specific future moment is further established, a corresponding network topology view is generated, and support is provided for network behavior decision based on dynamic topology.
The invention discloses a dynamic network topology prediction method based on historical data probability analysis. The method comprises the following steps: s1, acquiring historical detection data of each entity in a network to form an index matrix of each entity in the network; the entity is a physical node in the network, and the historical detection data comprises data which are obtained by detection and are used for representing the active state of the entity in a historical period and attribute values of the entity; s2, training the prediction probability model of each entity based on the index matrix to determine a prediction probability model parameter matrix of each entity; each entity corresponds to a prediction probability model, and the prediction probability model parameter matrix is a matrix formed by prediction probability model parameters related to a period; and S3, according to the prediction time input from the outside, predicting the active state of each entity at the prediction time by using the prediction probability model of each entity and the corresponding prediction probability model parameter matrix, and establishing a network topology view of the entity in the active state of the network at the prediction time based on the attribute value of the entity in the active state.
According to the method of the first aspect, in said step S1: storing Q pieces of historical detection data of P entities in a network entity database D, and recording the storage positions of the Q pieces of historical detection data as D i I is more than or equal to 1 and less than or equal to Q, and the entity is marked as O j J is more than or equal to 1 and less than or equal to P, and the entity is marked as O j Having m j According to historical detection data, the entity O j Index vector A of S（Oj） Characterized in that:
(ii) a Wherein D is 1（Oj） Represents the entity O j Storage location of the 1 st historical probe data, D 2（Oj） Represents the entity O j The storage location of the 2 nd historical probe data, and so on, D mj（Oj） Represents the entity O j M of j Storing positions of the historical detection data; obtaining the index vector of each entity, arranging the index vector of each entity according to rows to obtain the index matrix A of each entity in the network S The length of the index vector of each of the entities is not exactly equal.
According to the method of the first aspect, for the entity O j According to the index vector A S（Oj） Reading m stored in the network entity database D j Determining the entity O based on the read attribute values in the historical detection data j A corresponding predictive probability model; characterizing the entity O based on the read historical probe data j Data of active state in the history period, training and the entity O j Corresponding predictive probability model X j To obtain said entity O j Predicted probability model parameter vector C of j (ii) a The predictive probability model parameter vector C j Including state probability prediction calculation parameters C j1 And state probability threshold C j2 Obtaining the prediction probability model parameter vectors of each entity, and arranging the prediction probability model parameter vectors of each entity in rows to obtain the prediction probability model parameter matrix A C 。
According to the method of the first aspect, in said step S3: extracting a prediction time point from the externally input prediction time, and forming a prediction time point parameter N t Based on said predicted time-point parameter N t And the parameter matrix A of the predictive probability model C Included predictive probability model parameter vectors C for said respective entities j State probability prediction in (1) calculating parameter C j1 Using said trained pre-prediction of each entityMeasuring probability model, calculating the predicted state probability P of each entity j Arranging the predicted state probabilities of each of the entities in rows to form a predicted state probability matrix A P (ii) a The prediction state probability matrix A P Predicted state probability P in j Predictive probability model parameter vector C corresponding to entity j State probability threshold C in j2 Comparing the predicted state probability P j Not less than the state probability threshold C j2 Then, the entity O is determined j Being active at the predicted time; reading attribute values contained in historical detection data stored in the network entity database D according to the index vector of the entity in the active state, determining an incidence relation between the attribute values and the entity in the active state, and generating a network topology link based on the incidence relation to form the network topology view.
The second aspect of the invention discloses a dynamic network topology prediction system based on historical data probability analysis; the system comprises: a first processing unit configured to: acquiring historical detection data of each entity in a network to form an index matrix of each entity in the network; the entity is a physical node in the network, and the historical detection data comprises data which are obtained by detection and are used for representing the active state of the entity in a historical period and attribute values of the entity; a second processing unit configured to: training a predictive probability model of each entity based on the index matrix to determine a predictive probability model parameter matrix for each entity; each entity corresponds to a prediction probability model, and the prediction probability model parameter matrix is a matrix formed by prediction probability model parameters related to a period; a third processing unit configured to: according to the prediction time of external input, predicting the active state of each entity at the prediction time by using the prediction probability model of each entity and the corresponding prediction probability model parameter matrix, and establishing a network topology view of the entity in the active state of the network at the prediction time based on the attribute value of the entity in the active state.
The system according to the second aspect, the first processing unit is configured to: storing Q pieces of historical detection data of P entities in a network entity database D, and recording the storage positions of the Q pieces of historical detection data as D i I is more than or equal to 1 and less than or equal to Q, and the entity is marked as O j J is more than or equal to 1 and less than or equal to P, and the entity is marked as O j Having m j According to historical detection data, the entity O j Index vector A of S（Oj） Characterized in that:
wherein D is 1（Oj） Represents the entity O j Storage location of the 1 st historical probe data, D 2（Oj） Represents the entity O j The storage location of the 2 nd historical probe data, and so on, D mj（Oj） Represents the entity O j M of j A storage location of the bar of historical probe data; obtaining the index vector of each entity, arranging the index vectors of each entity according to rows to obtain an index matrix A of each entity in the network S The length of the index vector of each of the entities is not exactly equal.
The system according to the second aspect, the second processing unit is configured to: for the entity O j According to the index vector A S（Oj） Reading m stored in the network entity database D j The historical detection data is read, and the entity O is determined based on the attribute values in the historical detection data j A corresponding predictive probability model; characterizing the entity O based on read historical probe data j Data of active state in the history period, training and the entity O j Corresponding predictive probability model X j To obtain said entity O j Predicted probability model parameter vector C of j (ii) a The predictive probability model parameter vector C j Including state probability prediction calculation parameters C j1 Sum state probability threshold C j2 Obtaining the prediction probability model parameter vector of each entity, and calculating the prediction probability model of each entityThe type parameter vectors are arranged according to rows, then the prediction probability model parameter matrix A C 。
The system according to the second aspect, the third processing unit is configured to: extracting a prediction time point from the externally input prediction time, and forming a prediction time point parameter N t Based on said predicted time-point parameter N t And the prediction probability model parameter matrix A C Included predictive probability model parameter vectors C for said respective entities j State probability prediction calculation parameter C in (1) j1 Calculating the predicted state probability P of each entity by using the trained predicted probability model of each entity j Arranging the predicted state probabilities of each of the entities in rows to form a predicted state probability matrix A P (ii) a The prediction state probability matrix A P Predicted state probability P in j Predictive probability model parameter vector C corresponding to entity j State probability threshold C in j2 Comparing the predicted state probability P j Not less than the state probability threshold C j2 Then, the entity O is determined j Being active at the predicted time; reading attribute values contained in historical detection data stored in the network entity database D according to the index vector of the entity in the active state, determining an incidence relation between the attribute values and the entity in the active state, and generating a network topology link based on the incidence relation to form the network topology view.
A third aspect of the invention discloses an electronic device. The electronic device comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the dynamic network topology prediction method based on historical data probability analysis according to any one of the first aspect of the disclosure when executing the computer program.
A fourth aspect of the invention discloses a computer-readable storage medium. The computer readable storage medium has stored thereon a computer program which, when executed by a processor, implements the steps in a dynamic network topology prediction method based on historical data probability analysis according to any one of the first aspect of the present disclosure.
In conclusion, the network planning is developed through the network prediction topology view generated by the invention, so that the timeliness and the accuracy of the network planning can be effectively improved, and the planned network topology and the network at the future moment are implemented. The technical scheme of the invention is a system for predicting the network topology at the future moment based on the historical network detection data aiming at the network with the dynamic change characteristic. The problem of insufficient timeliness for developing operations such as network planning and management on the existing known network topology view is solved.
Drawings
In order to more clearly illustrate the embodiments or prior art solutions of the present invention, the drawings used in the embodiments or prior art descriptions will be briefly described below, it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without inventive efforts.
FIG. 1 is a flowchart illustrating a dynamic network topology prediction method based on historical data probability analysis according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention discloses a dynamic network topology prediction method based on historical data probability analysis. FIG. 1 is a flowchart illustrating a dynamic network topology prediction method based on historical data probability analysis according to an embodiment of the present invention; as shown in fig. 1, the method includes: s1, acquiring historical detection data of each entity in a network to form an index matrix of each entity in the network; the entity is a physical node in the network, and the historical detection data comprises data which are obtained by detection and are used for representing the active state of the entity in a historical period and attribute values of the entity; s2, training the prediction probability model of each entity based on the index matrix to determine a prediction probability model parameter matrix of each entity; each entity corresponds to a prediction probability model, and the prediction probability model parameter matrix is a matrix formed by prediction probability model parameters related to a period; and S3, according to the prediction time of external input, predicting the active state of each entity at the prediction time by using the prediction probability model of each entity and the corresponding prediction probability model parameter matrix, and establishing a network topology view of the entity in the active state of the network at the prediction time based on the attribute value of the entity in the active state.
In some embodiments, in said step S1: storing Q pieces of historical detection data of P entities in a network entity database D, and recording the storage positions of the Q pieces of historical detection data as D i I is more than or equal to 1 and less than or equal to Q, and the entity is marked as O j J is more than or equal to 1 and less than or equal to P, and the entity is marked as O j Having m j According to historical detection data, the entity O j Index vector A of S（Oj） Characterized in that:
(ii) a Wherein D is 1（Oj） Represents the entity O j Storage location of the 1 st historical probe data, D 2（Oj） Represents the entity O j The storage location of the 2 nd historical probe data, and so on, D mj（Oj） Represents the entity O j M of j A storage location of the bar of historical probe data; obtaining the index vector of each entity, arranging the index vectors of each entity according to rows to obtain the index vectors of each entityIndex matrix A for each entity in the network S The length of the index vector of each of the entities is not exactly equal.
In some embodiments, for said entity O j According to the index vector A S（Oj） Reading m stored in the network entity database D j The historical detection data is read, and the entity O is determined based on the attribute values in the historical detection data j A corresponding predictive probability model; characterizing the entity O based on the read historical probe data j Data of active state in the history period, training and the entity O j Corresponding predictive probability model X j To obtain said entity O j Predictive probability model parameter vector C j (ii) a The predictive probability model parameter vector C j Comprising a state probability prediction calculation parameter C j1 Sum state probability threshold C j2 Obtaining the prediction probability model parameter vectors of each entity, and arranging the prediction probability model parameter vectors of each entity according to rows to obtain the prediction probability model parameter matrix A C 。
In some embodiments, in said step S3: extracting a prediction time point from the externally input prediction time, and forming a prediction time point parameter N t Based on said predicted time-point parameter N t And the prediction probability model parameter matrix A C Included predictive probability model parameter vectors C for said respective entities j State probability prediction calculation parameter C in (1) j1 Calculating the predicted state probability P of each entity by using the trained predicted probability model of each entity j Arranging the predicted state probabilities of each of the entities in rows to form a predicted state probability matrix A P (ii) a The prediction state probability matrix A P Predicted state probability P in j Predictive probability model parameter vector C corresponding to entity j State probability threshold C in j2 Comparing the predicted state probability P j Not less than the state probability threshold C j2 Then, the entity O is determined j At the predicted timeIn an active state; reading attribute values contained in historical detection data stored in the network entity database D according to the index vector of the entity in the active state, determining an incidence relation between the attribute values and the entity in the active state, and generating a network topology link based on the incidence relation to form the network topology view.
The second aspect of the invention discloses a dynamic network topology prediction system based on historical data probability analysis; the system comprises: a first processing unit configured to: acquiring historical detection data of each entity in a network to form an index matrix of each entity in the network; the entity is a physical node in the network, and the historical detection data comprises data which are obtained by detection and are used for representing the active state of the entity in a historical period and attribute values of the entity; a second processing unit configured to: training a predictive probability model of each entity based on the index matrix to determine a predictive probability model parameter matrix of each entity; each entity corresponds to a prediction probability model, and the prediction probability model parameter matrix is a matrix formed by prediction probability model parameters related to a period; a third processing unit configured to: according to the prediction time of external input, predicting the active state of each entity at the prediction time by using the prediction probability model of each entity and the corresponding prediction probability model parameter matrix, and establishing a network topology view of the entity in the active state of the network at the prediction time based on the attribute value of the entity in the active state.
In some embodiments, the first processing unit is configured to: storing Q pieces of historical detection data of P entities in a network entity database D, and recording the storage positions of the Q pieces of historical detection data as D i I is more than or equal to 1 and less than or equal to Q, and the entity is marked as O j J is more than or equal to 1 and less than or equal to P, and the entity is marked as O j Having m j According to historical detection data, the entity O j Index vector A of S（Oj） Characterized in that:
wherein D is 1（Oj） Represents the entity O j 1 st storage location of historical probe data, D 2（Oj） Represents the entity O j The storage location of the 2 nd historical probe data, and so on, D mj（Oj） Represents the entity O j M of j A storage location of the bar of historical probe data; obtaining the index vector of each entity, arranging the index vector of each entity according to rows to obtain the index matrix A of each entity in the network S The length of the index vector of each of the entities is not exactly equal.
In some embodiments, the second processing unit is configured to: to the entity O j According to the index vector A S（Oj） Reading m stored in the network entity database D j The historical detection data is read, and the entity O is determined based on the attribute values in the historical detection data j A corresponding predictive probability model; characterizing the entity O based on read historical probe data j Data of active state in the history period, training and the entity O j Corresponding predictive probability model X j To obtain said entity O j Predicted probability model parameter vector C of j (ii) a The predictive probability model parameter vector C j Comprising a state probability prediction calculation parameter C j1 And state probability threshold C j2 Obtaining the prediction probability model parameter vectors of each entity, and arranging the prediction probability model parameter vectors of each entity in rows to obtain the prediction probability model parameter matrix A C 。
In some embodiments, the third processing unit is configured to: extracting a prediction time point from the externally input prediction time, and forming a prediction time point parameter N t Based on said predicted time-point parameter N t And the parameter matrix A of the predictive probability model C Included predictive probability model parameter vectors C for said respective entities j State probability prediction in (1) calculating parameter C j1 Using said respective entitiesCalculating the predicted state probability P of each entity by using the trained predictive probability model j Arranging the predicted state probabilities for each of the entities in rows to form a predicted state probability matrix A P (ii) a The prediction state probability matrix A is divided into P Predicted state probability P in j Predictive probability model parameter vector C with corresponding entity j State probability threshold C in j2 Comparing the predicted state probability P j Not less than the state probability threshold C j2 Then, the entity O is determined j Being active at the predicted time; reading attribute values contained in historical detection data stored in the network entity database D according to the index vector of the entity in the active state, determining an incidence relation between the attribute values and the entity in the active state, and generating a network topology link based on the incidence relation to form the network topology view.
Detailed description of the preferred embodiment 1
The present embodiment provides a feasible embodiment of a dynamic network topology prediction system based on historical data probability analysis, including: the network data access analysis module: the method mainly comprises the steps of completing the access of network detection data, the classified storage, classification and statistical analysis of the data, and providing a training sample for network topology prediction; the network entity prediction training module: training a network entity probability analysis model by mainly utilizing a network entity training sample to obtain a model prediction parameter of each network entity; a predicted network entity data generation module: the method mainly comprises the steps that for entity elements of a network universe, a network entity prediction model is utilized to calculate the prediction probability of each network entity, and a network topology prediction data set at a corresponding moment is generated through probability screening; the predicted network topology view display module: the method mainly utilizes the network topology prediction data set to generate a network topology prediction view, and provides support for network behavior decision.
The network object targeted by the embodiment is a relatively mature network, the network entity state and topology have time domain regularity and dynamic evolution characteristics, the entity state mainly embodied in the network takes periodic dynamic change as a main body, and the real-time topology of the network can dynamically evolve along with the entity state in the network; the network perception system carries out global and periodic detection on the network and forms network entity detection data in real time.
The network entity detection data comprises necessary attribute elements including at least detection time, an associated entity list and an entity type. The detection time attribute is used for training entity probability prediction model parameters by a system, the associated entity list is used for predicting the generation of the topological relation of the network topology, and the entity type is used for selecting the entity model when the network topology view is generated.
Specific example 2
The present embodiment provides a feasible embodiment of a dynamic network topology prediction method based on historical data probability analysis.
Step 1, a network entity probability model training process; the network entity probability model training process is realized by a network data access analysis module and a network entity state prediction training module, and comprises the following steps.
Step 11, acquiring full network entity detection data of the object network from various channels through a network detection data access module, and storing all acquired original data in a full network entity database D by adopting an accumulation method to realize the storage of the network entity detection data; the storage position of each piece of data is recorded as Di; each entity number in the network is marked as O according to the network entity number rule j ，O j With uniqueness, different probing of the same network entity, O j And is not changed.
Step 12, classifying and analyzing the stored network entity data to generate a network entity retrieval matrix A S As an input of the network entity state prediction, a specific method is as follows.
a. In the data classification analysis, the network entity is numbered O j As index key, searching network entity database D to obtain all O j Is detected as a data storage position D mj（Oj） Generated network entity O j Detection data retrieval set:
b. finishing the data classification index set of all network entities in D, and generating a network entity index matrix A from all index sets S ：
Wherein, the network entity index matrix A S The number of rows of (A) is identical to the total number of network entities S Is identical to the number of records in D for each network entity, and the number of records in D for different network entities is different, so that the matrix a S The data length of each row is different.
Step 13, in the network entity state prediction training module, different network entity prediction probability training models can be prefabricated according to different entity types, model training is respectively carried out on all network entities in the network entity index matrix, a prediction probability calculation model parameter of each network entity is obtained, and a network entity prediction probability model parameter base is generated.
a. Indexing matrix A according to network entity S Entity O in (1) j And reading the network entity type attribute value from the D.
b. Entity of basis O j Selecting a predictive probability training model by using the type attribute value and the set data A S（Oj） Developing a predictive probability computation model training to obtain an entity O j Corresponding set of prediction probability calculation model parameters C j The method comprises the steps of calculating a state probability prediction parameter and a state probability threshold value;
c. completion matrix A S Training the predictive probability calculation model of all network entities to generate a parameter matrix A of the predictive probability model C ：
Step 2, a network topology prediction display process; the network topology prediction display process is realized by a network entity prediction generation module and a prediction network topology view display module.
In some embodiments, according to a predicted time of an external input, an active state of each entity in the network at the predicted time is predicted by using a prediction probability model of each entity, a data set at the predicted time is generated based on the entities in the network in the active state, and a predicted topological view of the network at the predicted time is constructed. The method comprises the following steps:
step 21, the prediction time parameter processing module extracts the prediction point parameters from the external input prediction time and converts the prediction point parameters into a prediction point parameter set N t Using A C Each network entity O in the matrix j State probability prediction algorithm parameter C j Introducing a network entity state probability prediction calculation function module to calculate O j Predicted state probability P of j ：
. Thereby obtaining a matrix A S The probability matrix of the predicted state of all network entities:
step 22, for each matrix A S Network entity O in j Which corresponds to C j The state probability threshold in (1) is PO j (i.e., C) j2 ) Then predict O at the time j State T j The judgment rule is as follows:
T j =1 denotes a predicted point network entity O j The state is active and is visible in the prediction network, otherwise, the state is invisible;
step 23, all determined to be active O j Network entityConstructing a network topology prediction entity library E; the predicted network topology view display module is mainly used for reading related attribute values of corresponding network entities in the network entity database D through the data extraction module according to actively predicted network entities in the network topology prediction entity database E, performing entity association analysis on an associated entity list and generating predicted network topology link information; through the matching of network entity type data, a corresponding network entity view model can be selected from a view model library;
and 24, generating a network topology view at the prediction moment on a view interface by adopting a corresponding network layout algorithm through a prediction network topology view generation module.
A third aspect of the invention discloses an electronic device. The electronic device comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the dynamic network topology prediction method based on historical data probability analysis according to any one of the first aspect of the disclosure when executing the computer program.
Fig. 2 is a block diagram of an electronic device according to an embodiment of the present invention, and as shown in fig. 2, the electronic device includes a processor, a memory, a communication interface, a display screen, and an input device, which are connected by a system bus. Wherein the processor of the electronic device is configured to provide computing and control capabilities. The memory of the electronic equipment comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The communication interface of the electronic device is used for communicating with an external terminal in a wired or wireless mode, and the wireless mode can be realized through WIFI, an operator network, near Field Communication (NFC) or other technologies. The display screen of the electronic equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the electronic equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the electronic equipment, an external keyboard, a touch pad or a mouse and the like.
It will be understood by those skilled in the art that the structure shown in fig. 2 is only a partial block diagram related to the technical solution of the present disclosure, and does not constitute a limitation of the electronic device to which the solution of the present application is applied, and a specific electronic device may include more or less components than those shown in the drawings, or combine some components, or have a different arrangement of components.
A fourth aspect of the invention discloses a computer-readable storage medium. The computer readable storage medium has a computer program stored thereon, and when executed by a processor, the computer program implements the steps in a dynamic network topology prediction method based on historical data probability analysis according to any one of the first aspect of the present disclosure.
In conclusion, the network planning is developed through the network prediction topology view generated by the invention, so that the timeliness and the accuracy of the network planning can be effectively improved, and the network topology for planning implementation is consistent with the network at the future moment. The network space view constructed based on the network structure develops network planning, so that the timeliness and the accuracy of the network planning can be effectively improved, support is provided for network behavior decision based on dynamic topology, and the effectiveness of the network behavior decision is improved. The technical scheme of the invention is a system for predicting the network topology at the future moment based on the historical network detection data aiming at the network with the dynamic change characteristic. The problem of insufficient timeliness for developing network planning, management and other operations on the existing known network topology view is solved.
It should be noted that the technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, however, as long as there is no contradiction between the combinations of the technical features, the scope of the present description should be considered. The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (10)
1. A dynamic network topology prediction method based on historical data probability analysis is characterized by comprising the following steps:
s1, acquiring historical detection data of each entity in a network to form an index matrix of each entity in the network; the entity is a physical node in the network, and the historical detection data comprises data which are obtained by detection and are used for representing the active state of the entity in a historical period and attribute values of the entity;
s2, training the prediction probability model of each entity based on the index matrix to determine a prediction probability model parameter matrix of each entity; each entity corresponds to a prediction probability model, and the prediction probability model parameter matrix is a matrix formed by prediction probability model parameters related to a period;
and S3, according to the prediction time of external input, predicting the active state of each entity at the prediction time by using the prediction probability model of each entity and the corresponding prediction probability model parameter matrix, and establishing a network topology view of the entity in the active state of the network at the prediction time based on the attribute value of the entity in the active state.
2. The dynamic network topology prediction method based on historical data probability analysis according to claim 1, wherein in the step S1:
storing Q pieces of historical detection data of P entities in a network entity database D, and recording the storage positions of the Q pieces of historical detection data as D i I is more than or equal to 1 and less than or equal to Q, and the entity is marked as O j J is more than or equal to 1 and less than or equal to P, and the entity is marked as O j Having m j According to historical detection data, the entity O j Index vector A of S（ O j） Characterized in that:
wherein D is 1（ O j） Represents the entity O j Storage location of the 1 st historical probe data, D 2（ O j） Represents the entity O j The storage location of the 2 nd historical probe data, and so on, D mj（ O j） Represents the entity O j M of j A storage location of the bar of historical probe data;
obtaining the index vector of each entity, arranging the index vector of each entity according to rows to obtain the index matrix A of each entity in the network S The length of the index vector of each of the entities is not exactly equal.
3. The method for predicting the dynamic network topology based on the historical data probability analysis as claimed in claim 2, wherein in the step S2:
for the entity O j According to the index vector A S（ O j） Reading m stored in the network entity database D j The historical detection data is read, and the entity O is determined based on the attribute values in the historical detection data j A corresponding predictive probability model;
characterizing the entity O based on the read historical probe data j Data of active state in the history period, training and the entity O j Corresponding predictive probability model X j To obtain said entity O j Predicted probability model parameter vector C of j ；
The predictive probability model parameter vector C j Including state probability prediction calculation parameters C j1 Sum state probability threshold C j2 Obtaining the prediction probability model parameter vectors of each entity, and arranging the prediction probability model parameter vectors of each entity according to rows to obtain the prediction probability model parameter matrix A C 。
4. The dynamic network topology prediction method based on historical data probability analysis according to claim 3, wherein in the step S3:
extracting a prediction time point from the externally input prediction time, and forming a prediction time point parameter N t Based on said predicted time-point parameter N t And the prediction probability model parameter matrix A C Included predictive probability model parameter vectors C for said respective entities j State probability prediction calculation parameter C in (1) j1 Calculating the predicted state probability P of each entity by using the trained predicted probability model of each entity j Arranging the predicted state probabilities of each of the entities in rows to form a predicted state probability matrix A P ；
The prediction state probability matrix A is divided into P Predicted state probability P in j Predictive probability model parameter vector C corresponding to entity j State probability threshold C in j2 Comparing the predicted state probability P j Not less than the state probability threshold C j2 Then, the entity O is determined j Being active at the predicted time;
reading attribute values contained in historical detection data stored in the network entity database D according to the index vector of the entity in the active state, determining an incidence relation between the attribute values and the entity in the active state, and generating a network topology link based on the incidence relation to form the network topology view.
5. A dynamic network topology prediction system based on historical data probability analysis, the system comprising:
a first processing unit configured to: acquiring historical detection data of each entity in a network to form an index matrix of each entity in the network; the entity is a physical node in the network, and the historical detection data comprises data which are obtained by detection and are used for representing the active state of the entity in a historical period and attribute values of the entity;
a second processing unit configured to: training a predictive probability model of each entity based on the index matrix to determine a predictive probability model parameter matrix of each entity; each entity corresponds to a prediction probability model, and the prediction probability model parameter matrix is a matrix formed by prediction probability model parameters related to a period;
a third processing unit configured to: according to the prediction time of external input, predicting the active state of each entity at the prediction time by using the prediction probability model of each entity and the corresponding prediction probability model parameter matrix, and establishing a network topology view of the entity in the active state of the network at the prediction time based on the attribute value of the entity in the active state.
6. The system according to claim 5, wherein the first processing unit is configured to:
storing Q pieces of historical detection data of P entities in a network entity database D, and recording the storage positions of the Q pieces of historical detection data as D i I is more than or equal to 1 and less than or equal to Q, and the entity is marked as O j J is more than or equal to 1 and less than or equal to P, and the entity is marked as O j Having m j The historical probe data, then the entity O j Index vector A of S（Oj） Characterized in that:
wherein D is 1（Oj） Represents the entity O j Storage location of the 1 st historical probe data, D 2（Oj） Represents the entity O j The storage location of the 2 nd historical probe data, and so on, D mj（Oj） Represents the entity O j M of j A storage location of the bar of historical probe data;
obtaining the index vector of each entity, arranging the index vector of each entity according to rows to obtain the index matrix A of each entity in the network S The length of the index vector of each of the entities is not exactly equal.
7. The dynamic network topology prediction system based on historical data probability analysis of claim 6, wherein the second processing unit is configured to:
for the entity O j According to the index vector A S（Oj） Reading m stored in the network entity database D j The historical detection data is read, and the entity O is determined based on the attribute values in the historical detection data j A corresponding predictive probability model;
characterizing the entity O based on read historical probe data j Data of active state in the history period, training and the entity O j Corresponding predictive probability model X j To obtain said entity O j Predicted probability model parameter vector C of j ；
The predictive probability model parameter vector C j Including state probability prediction calculation parameters C j1 And state probability threshold C j2 Obtaining the prediction probability model parameter vectors of each entity, and arranging the prediction probability model parameter vectors of each entity according to rows to obtain the prediction probability model parameter matrix A C 。
8. The dynamic network topology prediction system based on historical data probability analysis of claim 7, wherein the third processing unit is configured to:
extracting a prediction time point from the externally input prediction time, and forming a prediction time point parameter N t Based on said predicted time-point parameter N t And the prediction probability model parameter matrix A C Included predictive probability model parameter vectors C for said respective entities j State probability predictor in (1)Calculating parameter C j1 Calculating the predicted state probability P of each entity by using the trained predicted probability model of each entity j Arranging the predicted state probabilities of each of the entities in rows to form a predicted state probability matrix A P ；
The prediction state probability matrix A is divided into P Predicted state probability P in (1) j Predictive probability model parameter vector C with corresponding entity j State probability threshold C in j2 Comparing the predicted state probability P j Not less than the state probability threshold C j2 When it is determined that the entity O is j Being active at the predicted time;
reading attribute values contained in historical detection data stored in the network entity database D according to the index vector of the entity in the active state, determining an incidence relation between the attribute values and the entity in the active state, and generating a network topology link based on the incidence relation to form the network topology view.
9. An electronic device, comprising a memory storing a computer program and a processor, wherein the processor implements the steps of the method for dynamic network topology prediction based on historical data probability analysis according to any one of claims 1 to 4 when executing the computer program.
10. A computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, and when being executed by a processor, the computer program implements the steps of any one of claims 1 to 4 in a dynamic network topology prediction method based on historical data probability analysis.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211629149.7A CN115622894B (en) | 2022-12-19 | 2022-12-19 | Dynamic network topology prediction method and system based on historical data probability analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211629149.7A CN115622894B (en) | 2022-12-19 | 2022-12-19 | Dynamic network topology prediction method and system based on historical data probability analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115622894A true CN115622894A (en) | 2023-01-17 |
| CN115622894B CN115622894B (en) | 2023-04-18 |
Family
ID=84880884
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211629149.7A Active CN115622894B (en) | 2022-12-19 | 2022-12-19 | Dynamic network topology prediction method and system based on historical data probability analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115622894B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104134159A (en) * | 2014-08-04 | 2014-11-05 | 中国科学院软件研究所 | Method for predicting maximum information spreading range on basis of random model |
| US20180196893A1 (en) * | 2017-01-09 | 2018-07-12 | Facebook, Inc. | Stochastic network traffic modeling |
| CN111884867A (en) * | 2020-08-17 | 2020-11-03 | 南昌航空大学 | Opportunistic network topology prediction method and device based on cycle generation type countermeasure network |
| CN112116138A (en) * | 2020-09-09 | 2020-12-22 | 山东科技大学 | Power system prediction state estimation method and system based on data driving |
| CN114997036A (en) * | 2022-02-25 | 2022-09-02 | 中国人民解放军国防科技大学 | Network topology reconstruction method, device and equipment based on deep learning |
-
2022
- 2022-12-19 CN CN202211629149.7A patent/CN115622894B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104134159A (en) * | 2014-08-04 | 2014-11-05 | 中国科学院软件研究所 | Method for predicting maximum information spreading range on basis of random model |
| US20180196893A1 (en) * | 2017-01-09 | 2018-07-12 | Facebook, Inc. | Stochastic network traffic modeling |
| CN111884867A (en) * | 2020-08-17 | 2020-11-03 | 南昌航空大学 | Opportunistic network topology prediction method and device based on cycle generation type countermeasure network |
| CN112116138A (en) * | 2020-09-09 | 2020-12-22 | 山东科技大学 | Power system prediction state estimation method and system based on data driving |
| CN114997036A (en) * | 2022-02-25 | 2022-09-02 | 中国人民解放军国防科技大学 | Network topology reconstruction method, device and equipment based on deep learning |
Non-Patent Citations (3)
| Title |
|---|
| LIANG ZHANG ETAL: "Probabilistic Network Topology Prediction for Active Planning: An Adaptive Algorithm and Application" * |
| 宋光鑫等: "利用矩阵补全优化模型进行动态网络链接预测" * |
| 陈莎等: "一种基于混合相似性指标的网络动态链路预测方法" * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115622894B (en) | 2023-04-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112910710B (en) | Network flow space-time prediction method and device, computer equipment and storage medium | |
| JP2015082259A (en) | Time series data prediction device, time series data prediction method, and program | |
| CN115659177A (en) | Method and device for generating data recommendation model and computer equipment | |
| Mustafa et al. | A Time Monte Carlo method for addressing uncertainty in land-use change models | |
| Amin et al. | Towards resolving unidentifiability in inverse reinforcement learning | |
| Liang et al. | Learning social relations and spatiotemporal trajectories for next check-in inference | |
| Chowdhury | Supervised machine learning and heuristic algorithms for outlier detection in irregular spatiotemporal datasets | |
| JP6658507B2 (en) | Load estimation system, information processing device, load estimation method, and computer program | |
| CN115545103A (en) | Abnormal data identification method, label identification method and abnormal data identification device | |
| JP4088218B2 (en) | Data extraction apparatus, data extraction method, and data extraction program | |
| CN115622894B (en) | Dynamic network topology prediction method and system based on historical data probability analysis | |
| CN116388864B (en) | Quantum network device performance prediction method and device, electronic device and storage medium | |
| CN110503296B (en) | Test method, test device, computer equipment and storage medium | |
| CN109800887B (en) | Generation method and device of prediction process model, storage medium and electronic equipment | |
| WO2021193931A1 (en) | Machine learning device, learning model generation method, and program | |
| CN115758271A (en) | Data processing method, data processing device, computer equipment and storage medium | |
| CN112581250B (en) | Model generation method, device, computer equipment and storage medium | |
| CN111737319B (en) | User cluster prediction method, device, computer equipment and storage medium | |
| CN114881521A (en) | Service evaluation method, device, electronic equipment and storage medium | |
| CN115455276A (en) | Method and device for recommending object, computer equipment and storage medium | |
| JP6726312B2 (en) | Simulation method, system, and program | |
| CN116405323B (en) | Security situation awareness attack prediction method, device, equipment, medium and product | |
| CN117875737A (en) | Resource prediction method, apparatus, device, storage medium, and program product | |
| CN116150341B (en) | Method for detecting claim event, computer device and storage medium | |
| CN117891811B (en) | Customer data acquisition and analysis method and device and cloud server |
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 |