Heterogeneous database fusion access systemTechnical Field
The invention relates to the technical field of network databases, in particular to a heterogeneous database fusion access system.
Background
With the rapid development of network data architecture, network database technology is gradually developed, and a considerable number of enterprises and scientific institutions currently accumulate a large amount of data which is stored in different forms and depends on different database management systems. Based on the rapid development of computer technology, various heterogeneous data information exists in different data layers, in the big data age, informationized data is explosively increased, and the challenges are difficult to comprehensively and efficiently face by the traditional relational database and the emerging NoSQL database. How to provide the data information is a problem to be solved by further development of enterprises.
In the prior art, although the integration of multi-source heterogeneous data can be realized, the shared access management method of the heterogeneous database is completed aiming at the modes of heterogeneous data source registration, virtual database table object registration, virtual resource registration and the like, and a uniform data access interface is provided, the integration degree of the multi-source heterogeneous data is poor, the data communication performance is poor, the resource intercommunication of all external third party application systems is difficult to realize, and the uniform management capability of the heterogeneous data is poor.
In the prior art, an XML-based heterogeneous data source data fusion system is also adopted, and the technology solves the problem of distributed heterogeneous characteristics generated by heterogeneous data sources, so that a user does not need to know the physical distribution, composition and operation method of each data source, does not need to carry out complex result summarization, and can obtain a desired result only through simple global access. But the processing and calculation methods of the data are not mentioned yet, and the interactive capability of the data is poor.
Disclosure of Invention
Aiming at the defects of the technology, the invention discloses a heterogeneous database fusion access system, which can realize fusion, calculation and processing of various heterogeneous data information and improve the management capability of a database.
In order to solve the technical problems, the research adopts the following technical scheme: a heterogeneous database fusion access system, comprising:
At least one database interface; the database interface is a database interface of different data models in a network architecture, and is at least an Oracle data interface, a SQLServer data interface, a Unix data interface, a Window NT data interface or a Linux data interface;
An information distribution module; the information distribution module comprises a communication protocol mapping module, a multi-data interface and an information receiving and transmitting module, wherein the output end of the communication protocol mapping module is connected with the input end of the multi-source data interface, the output end of the multi-source data interface is connected with the input end of the central processing unit, and the output end of the central processing unit is connected with the input end of the information receiving and transmitting module;
a data preprocessing module; the method is used for filtering, clearing or processing the received heterogeneous data information of different networks and is used for realizing pure output of the input data information;
at least one data processing module; the method comprises the steps of classifying acquired disordered different heterogeneous data information so as to enable the data information to be acquired by a user quickly;
a fusion module; the method is used for optimizing different types of multi-source heterogeneous data information so as to improve the communication quality of heterogeneous networks; and
A client application; the method is used for applying the processed data information;
The output end of the at least one database interface is connected with the input end of the information distribution module, the output end of the information distribution module is connected with the input end of the data preprocessing module, the output end of the data preprocessing module is connected with the input end of the at least one data processing module, the output end of the at least one data processing module is connected with the input end of the fusion module, and the output end of the fusion module is connected with the input end of the client application end.
As a further technical scheme of the invention, the information transceiver module is at least a Modbus transceiver module, an HTTP transceiver module, an XMPP transceiver module, a WIA-PA transceiver module, a PLC transceiver module or an RS485 transceiver module; and the multi-data interface is at least connected with a Modbus data interface, an HTTP data interface, an XMPP data interface, a WIA-PA data interface, a PLC data interface or an RS485 data interface.
As a further technical scheme of the invention, the central processing unit is an ARM processor.
As a further technical scheme of the present invention, the communication protocol mapping module includes a physical layer, a driving layer, an application layer and a virtual bus implementation layer, wherein an output end of the physical layer is connected with an input end of the driving layer, an output end of the driving layer is connected with an input end of the application layer, and an output end of the application layer is connected with an input end of the virtual bus implementation layer.
As a further technical scheme of the invention, the data processing module is a data processing module based on a computer program algorithm, and the data processing module comprises a data feature extraction module, a feature receiving module, a feature data dimension reduction module, a feature calculating module and a processed data output module, wherein the output end of the data feature extraction module is connected with the input end of the feature receiving module, the output end of the feature receiving module is connected with the input end of the feature data dimension reduction module, the output end of the feature data dimension reduction module is connected with the input end of the feature calculating module, and the output end of the feature calculating module is connected with the input end of the processed data output module.
As a further technical scheme of the invention, the method for extracting data by the data characteristic extraction module comprises the following steps:
Setting the data category of the multi-source heterogeneous data information as n, and representing the fused multi-source heterogeneous data function by an i multiplied by j matrix:
after all the data information is constructed into the data matrix, any row and column data information in the multi-source heterogeneous data information can be marked as alphaij(ωf), and then:
In the formula (2), when the data characteristics are extracted, by extracting multi-source heterogeneous data information, assuming that the extracted characteristic attribute is marked as j, under the action of j, the extracted multi-source heterogeneous data information function is alphaij (·), i represents the frequency response of the output multi-source heterogeneous data information function, wherein i, j=1, 2, … …, n; phiik is the i-th characteristic data signal element of the k-th information characteristic phik in dynamic response of multi-source heterogeneous data, phijk is the j-th characteristic data signal element of the k-th information characteristic phik in dynamic response of multi-source heterogeneous data, omegak is the k-th natural frequency of multi-source heterogeneous data output, and xik is the k-th data purity coefficient detected by multi-source heterogeneous data;
When abnormal phenomena occur in the multi-source heterogeneous data information, according to the inherent information frequency function of the multi-source heterogeneous data information, outputting a frequency response difference value, wherein the information characteristic of the multi-source heterogeneous data information output database, and the difference value formula between the retrieved data information and the data retrieval standard set by a user is expressed as follows:
In the formula (3), ωp represents the frequency of the multi-source heterogeneous data output when the frequency response function of the multi-source heterogeneous data output is in the p-order mode, and Δλp is the ratio of the difference value existing in the information spectrum of the multi-source heterogeneous data sent by the multi-source heterogeneous data to the different heterogeneous data information frequency when the multi-source heterogeneous data is in the p-order mode and the information characteristic of the p-order data.
As a further technical scheme of the invention, the method for realizing multi-source heterogeneous data dimension reduction by the characteristic data dimension reduction module comprises the following steps:
Step one: setting multi-source heterogeneous data information, and recording multi-source heterogeneous data reconstruction data as a matrix A, wherein the multi-source heterogeneous data reconstruction data comprises:
in the formula (4), n represents the number of multi-source heterogeneous data, wherein k is the data dimension of multi-source heterogeneous data information, the dimension data information is represented as k=n- (u-1) r through the formula, wherein r is the time delay for acquiring the data dimension, and u is the embedding dimension set in the data dimension reduction process;
Step two: then calculating the values of k and r, and calculating the interaction information quantity O after dimension reduction in the delay time during data acquisition, wherein the interaction information quantity O is expressed by the following formula:
In the formula (5), m and n represent different data element libraries in the multi-source heterogeneous database, P is the distribution probability of m and n element libraries in the input multi-source heterogeneous data information, when O (r) is equal to 0, the data characteristics of a (t+r) and a (t) have no correlation, and when O (r) is 0, the data characteristics of a (t+r) and a (t) represent obvious data fusion;
Step three: in the heterogeneous dimensionality reduction of multi-source data, assuming that the difference between two different dimensions is represented by the a (n) in a database element in the u dimension, there are:
A(n)＝{a(n),a(n+r),…,a[n+(u-1)r]} (6)
When the distance between the multi-source heterogeneous data information a (n) and a (n+r) is larger than 10, the multi-source heterogeneous data information a (n) and a (n+r) are expressed as no relation, and when the distance between a (n) and a (n+r) is larger than 10, the multi-source heterogeneous data information is removed, and the rest data information is reserved.
As a further technical scheme of the invention, the method for realizing the dimension reduction of the multi-source heterogeneous data by the characteristic data dimension reduction module is a characteristic pair measurement method, and the dimension reduction of the multi-source heterogeneous data information is realized by the method, wherein the dimension reduction formula is as follows:
In the formula (7), reconstructing a data set from data information with a (n) and a (n+r) being greater than 10, wherein the optimal data dimension is u, and after reconstruction, an original dimension matrix Am of the building design database is represented by the following relational expression:
Am＝(am,am+r,…,am+(u-1)r)T (8)
in formula (8), T represents transposed matrices of different information dimensions of the multi-source heterogeneous data information, and it is assumed that two center points of the multi-source heterogeneous data information are am and an, and the dimensionality reduction formula is as follows:
In equation (9), where H is a herveliedel function, li is the distance from the data dimension i to the center point dimension, and the reduced dimension data function D (L) is expressed as:
in the formula (10), the relationship between different data is calculated by a given distance standard value, and then there is:
|am-(v-1)r-an-(v-1)r|1≤v≤u (11)
in formula (11), v represents a range of dimensions that meet the information requirements of the multi-source heterogeneous data.
As a further technical scheme of the invention, the characteristic calculation module comprises a calculation chip, a data input interface, a display module, a data output interface, an external expansion circuit, a relay assembly, a relay switch circuit, a computer management system and a data classification model, wherein the data input interface, the display module, the data output interface, the external expansion circuit, the relay assembly, the relay switch circuit, the computer management system and the data classification model are connected with the calculation chip; the computing chip is an ARM 9-based embedded single-chip microcomputer computing system, and the data classification model is a classification algorithm model.
As a further technical scheme of the invention, the method for realizing data fusion by the characteristic calculation module comprises the following steps:
(1) Data selection, namely, assuming multi-source heterogeneous network communication data output from a database interface to be recorded as a large sample, recording target functions of the data as Y to (alpha, beta2), and recording expected values and standard deviations as alpha, beta, wherein the value ranges of the alpha and the beta are respectively between [0,1];
When data fusion calculation is carried out, setting data information to be fused as m different data types, recording as a data set B= (B1,B2,...,Bm), and recording an output heterogeneous network data source as a data set Z= (Z1,Z2,...,Zn) after n data types are converted, wherein the confidence level of the communication data sources of different heterogeneous networks is set to be qj;
(2) The heterogeneous network communication data sources converted through data operation are collected into a data set:
Zji＝(bji,cji,dji) (12)
In the formula (1), the conversion degree of the jth heterogeneous network communication data source and the ith decision object is recorded as Zji; the conversion degree of the multi-source heterogeneous data information is expressed by a triangle fuzzy number function, and the function formula is recorded as (bji,cji,dji) and comprises the following steps: bji≤cji≤dji is more than or equal to 0 and less than or equal to 1;
(2) In order to balance a data conversion function and improve the data computing capacity, OWA operator weight vectors are set in a function formula, reasonable fuzzy semantic quantization standards are selected according to the preference of a decision maker, and OWA operator weight vectors are set; the degree of data conversion is divided into 'complete conversion', 'unconverted' and 'in-conversion', 'complete conversion', 'unconverted' data parameter set to (0.45,0.79) 'unconverted' data parameter set to (0.16,0.32) 'in-conversion', 'in-conversion' data parameter set to (0.78,1.03), and g (y) can be recorded by acquiring fuzzy semantic quantization operator functions according to parameters; according to the set function g (y), calculating an OWA operator weight vector and recording the vector asWherein m represents the number of communication data sources input into the heterogeneous network;
(3) After setting the data parameters, carrying out data conversion on the confidence qj of each heterogeneous network communication data source and the approval level value zji;
in the method of utilizing and increasing the calculated amount of OWA weight vectors, setting each decision value of qj and zji, and then carrying out size arrangement on the decision values in a data conversion mode, wherein a data conversion formula can be recorded as:
zji-min＝qj×zji (13)
zji-max＝qj+zji-qj×zji (14)
(4) And realizing heterogeneous network communication data fusion according to the OWA operator weight vector and the converted acceptance level, and simultaneously calculating the final decision value of each decision. Then there are:
Where the j-th maximum element is set to cji.
(5) And then continuously inputting network heterogeneous data information, and fusing the data information according to the set decision value, so as to realize heterogeneous network communication data fusion.
Has the positive beneficial effects that:
The invention comprises a database interface, an information distribution module, a data preprocessing module, a data processing module and a client application end, and can distribute, preprocess, extract characteristic information, fuse data and the like for heterogeneous data information in various different forms, thereby realizing the fusion processing of heterogeneous data information of multi-source data.
The information transceiver module is at least a Modbus transceiver module, an HTTP transceiver module, an XMPP transceiver module, a WIA-PA transceiver module, a PLC transceiver module or an RS485 transceiver module; and the multiple data interfaces are at least connected with a Modbus data interface, an HTTP data interface, an XMPP data interface, a WIA-PA data interface, a PLC data interface or an RS485 data interface, so that multiple data communication can be supported, and the data communication capacity is improved.
The data processing module adopted by the invention is a data processing module based on a computer program algorithm, and comprises a data feature extraction module, a feature receiving module, a feature data dimension reduction module, a feature calculation module and a processed data output module, wherein the output end of the data feature extraction module is connected with the input end of the feature receiving module, the output end of the feature receiving module is connected with the input end of the feature data dimension reduction module, the output end of the feature data dimension reduction module is connected with the input end of the feature calculation module, and the output end of the feature calculation module is connected with the input end of the processed data output module. By extracting the characteristic data information, data information analysis can be realized.
The invention adopts the characteristic calculation module to comprise a calculation chip, a data input interface, a display module, a data output interface, an external expansion circuit, a relay assembly, a relay switch circuit, a computer management system and a data classification model which are connected with the calculation chip; the computing chip is an ARM 9-based embedded single-chip microcomputer computing system, and the data classification model is a classification algorithm model, so that data information can be processed and calculated.
Drawings
For a clearer description of embodiments of the invention or of solutions in the prior art, the drawings that are necessary for the description of the embodiments or of the prior art will be briefly described, it being apparent that the drawings in the description below are only some embodiments of the invention, from which, without inventive faculty, other drawings can be obtained for a person skilled in the art, in which:
FIG. 1 is a schematic diagram of a hardware architecture of a heterogeneous database fusion access system according to the present invention;
FIG. 2 is a schematic diagram of an information distribution module architecture in a heterogeneous database fusion access system according to the present invention;
FIG. 3 is a schematic diagram of a communication protocol mapping module in a heterogeneous database fusion access system according to the present invention;
FIG. 4 is a schematic diagram of a data processing module in a heterogeneous database fusion access system according to the present invention;
fig. 5 is a schematic diagram of a hardware architecture of a feature calculation module in a heterogeneous database fusion access system according to the present invention.
Detailed Description
The preferred embodiments of the present invention will be described below with reference to the accompanying drawings, it being understood that the embodiments described herein are for illustration and explanation of the present invention only, and are not intended to limit the present invention.
Example 1
As shown in fig. 1, a heterogeneous database fusion access system includes:
At least one database interface; the database interface is a database interface of different data models in a network architecture, and is at least an Oracle data interface, a SQLServer data interface, a Unix data interface, a Window NT data interface or a Linux data interface; through the arrangement, interaction of various data information can be realized, and various information interactions with external equipment can be realized.
An information distribution module; the information distribution module comprises a communication protocol mapping module, a multi-data interface and an information receiving and transmitting module, wherein the output end of the communication protocol mapping module is connected with the input end of the multi-source data interface, the output end of the multi-source data interface is connected with the input end of the central processing unit, and the output end of the central processing unit is connected with the input end of the information receiving and transmitting module; in this way, interaction and allocation of different types of data channels can be achieved.
A data preprocessing module; the method is used for filtering, clearing or processing the received heterogeneous data information of different networks and is used for realizing pure output of the input data information; by the method, calculation and processing of heterogeneous data of different networks can be realized, and the data information processing capability is improved.
At least one data processing module; the method comprises the steps of classifying acquired disordered different heterogeneous data information so as to enable the data information to be acquired by a user quickly; in this way, the cluttered different heterogeneous data information can be made regular.
A fusion module; the method is used for optimizing different types of multi-source heterogeneous data information so as to improve the communication quality of heterogeneous networks; and
A client application; the method is used for applying the processed data information;
The output end of the at least one database interface is connected with the input end of the information distribution module, the output end of the information distribution module is connected with the input end of the data preprocessing module, the output end of the data preprocessing module is connected with the input end of the at least one data processing module, the output end of the at least one data processing module is connected with the input end of the fusion module, and the output end of the fusion module is connected with the input end of the client application end.
As shown in fig. 2, the information transceiver module is at least a Modbus transceiver module, an HTTP transceiver module, an XMPP transceiver module, a WIA-PA transceiver module, a PLC transceiver module, or an RS485 transceiver module; and the multi-data interface is at least connected with a Modbus data interface, an HTTP data interface, an XMPP data interface, a WIA-PA data interface, a PLC data interface or an RS485 data interface.
In the present invention, the central processor is an ARM processor.
In the invention, the communication protocol mapping module comprises a physical layer, a driving layer, an application layer and a virtual bus implementation layer, wherein the output end of the physical layer is connected with the input end of the driving layer, the output end of the driving layer is connected with the input end of the application layer, and the output end of the application layer is connected with the input end of the virtual bus implementation layer.
In the invention, the data processing module is a data processing module based on a computer program algorithm, and the data processing module comprises a data feature extraction module, a feature receiving module, a feature data dimension reduction module, a feature calculation module and a processed data output module, wherein the output end of the data feature extraction module is connected with the input end of the feature receiving module, the output end of the feature receiving module is connected with the input end of the feature data dimension reduction module, the output end of the feature data dimension reduction module is connected with the input end of the feature calculation module, and the output end of the feature calculation module is connected with the input end of the processed data output module.
Example two
In the invention, the method for extracting data by the data characteristic extraction module comprises the following steps:
Setting the data category of the multi-source heterogeneous data information as n, and representing the fused multi-source heterogeneous data function by an i multiplied by j matrix:
All data information can be constructed into a data matrix by the formula (1) so as to process multiple forms of multi-source heterogeneous data information. To put forward data computing power, the data information of any row and column in the multi-source heterogeneous data information can be denoted as αij(ωf), then there are:
In the formula (2), when the data characteristics are extracted, by extracting multi-source heterogeneous data information, assuming that the extracted characteristic attribute is marked as j, under the action of j, the extracted multi-source heterogeneous data information function is alphaij (·), i represents the frequency response of the output multi-source heterogeneous data information function, wherein i, j=1, 2, … …, n; phiik is the i-th characteristic data signal element of the k-th information characteristic phik in dynamic response of multi-source heterogeneous data, phijk is the j-th characteristic data signal element of the k-th information characteristic phik in dynamic response of multi-source heterogeneous data, omegak is the k-th natural frequency of multi-source heterogeneous data output, and xik is the k-th data purity coefficient detected by multi-source heterogeneous data; by the formula, specific data information in the multi-source heterogeneous data information can be dissociated.
When abnormal phenomena occur in the multi-source heterogeneous data information, according to the inherent information frequency function of the multi-source heterogeneous data information, outputting a frequency response difference value, wherein the information characteristic of the multi-source heterogeneous data information output database, and the difference value formula between the retrieved data information and the data retrieval standard set by a user is expressed as follows:
In the formula (3), ωp represents the frequency of the multi-source heterogeneous data output when the frequency response function of the multi-source heterogeneous data output is in the p-order mode, and Δλp is the ratio of the difference value existing in the information spectrum of the multi-source heterogeneous data sent by the multi-source heterogeneous data to the different heterogeneous data information frequency when the multi-source heterogeneous data is in the p-order mode and the information characteristic of the p-order data. Through the formula, the relation between the data information and the data set by the user can be calculated, and the multi-source heterogeneous data characteristic can be directly output through comparison discovery.
Example III
In the invention, the method for realizing multi-source heterogeneous data dimension reduction by the characteristic data dimension reduction module comprises the following steps:
Step one: setting multi-source heterogeneous data information, and recording multi-source heterogeneous data reconstruction data as a matrix A, wherein the multi-source heterogeneous data reconstruction data comprises:
In the formula (4), n represents the number of multi-source heterogeneous data, wherein k is the data dimension of multi-source heterogeneous data information, the dimension data information is represented as k=n- (u-1) r through the formula, wherein r is the time delay for acquiring the data dimension, and u is the embedding dimension set in the data dimension reduction process; by representing the multisource heterogeneous data reconstruction data in a matrix mode, aggregation of different data groups can be achieved, data research is facilitated, and data aggregation capability is improved.
Step two: then calculating the values of k and r, and calculating the interaction information quantity O after dimension reduction in the delay time during data acquisition, wherein the interaction information quantity O is expressed by the following formula:
In the formula (5), m and n represent different data element libraries in the multi-source heterogeneous database, P is the distribution probability of m and n element libraries in the input multi-source heterogeneous data information, when O (r) is equal to 0, the data characteristics of a (t+r) and a (t) have no correlation, and when O (r) is 0, the data characteristics of a (t+r) and a (t) represent obvious data fusion;
Step three: in the heterogeneous dimensionality reduction of multi-source data, assuming that the difference between two different dimensions is represented by the a (n) in a database element in the u dimension, there are:
A(n)＝{a(n),a(n+r),…,a[n+(u-1)r]} (6)
When the distance between the multi-source heterogeneous data information a (n) and a (n+r) is larger than 10, the multi-source heterogeneous data information a (n) and a (n+r) are expressed as no relation, and when the distance between a (n) and a (n+r) is larger than 10, the multi-source heterogeneous data information is removed, and the rest data information is reserved.
In a further embodiment of the present invention, the method for implementing dimension reduction of multi-source heterogeneous data by the feature data dimension reduction module is a feature pair measurement method, and dimension reduction of multi-source heterogeneous data information is implemented by the method, where a dimension reduction formula is as follows:
In the formula (7), reconstructing a data set from data information with a (n) and a (n+r) being greater than 10, wherein the optimal data dimension is u, and after reconstruction, an original dimension matrix Am of the building design database is represented by the following relational expression:
Am＝(am,am+r,…,am+(u-1)r)T (8)
in the formula (8), T represents transposed matrixes of different information dimensions of the multi-source heterogeneous data information, and simple processing of complex data can be realized through the formula, so that the data processing capability is improved.
Assuming that two central points of the multi-source heterogeneous data information are am and an, the dimensionality reduction formula is as follows:
In equation (9), where H is a herveliedel function, li is the distance from the data dimension i to the center point dimension, and the reduced dimension data function D (L) is expressed as:
in the formula (10), the relationship between different data is calculated by a given distance standard value, and then there is:
|am-(v-1)r-an-(v-1)r|1≤v≤u (11)
In formula (11), v represents a range of dimensions that meet the information requirements of the multi-source heterogeneous data. The formula can realize the simplified processing of the multi-source heterogeneous data information, and the user has outstanding efficiency when in application.
In the invention, the characteristic calculation module comprises a calculation chip, a data input interface connected with the calculation chip, a display module, a data output interface, an external expansion circuit, a relay assembly, a relay switch circuit, a computer management system and a data classification model; the computing chip is an ARM 9-based embedded single-chip microcomputer computing system, and the data classification model is a classification algorithm model.
By calculating in this way, the processing of various information features of heterogeneous data can be realized.
Example IV
In the invention, the method for realizing data fusion by the characteristic calculation module comprises the following steps:
(1) Data selection, namely, assuming multi-source heterogeneous network communication data output from a database interface to be recorded as a large sample, recording target functions of the data as Y to (alpha, beta2), and recording expected values and standard deviations as alpha, beta, wherein the value ranges of the alpha and the beta are respectively between [0,1]; in this way, the data information can be quickly acquired, so that different data information can be selected in a special range for use by a user.
When data fusion calculation is carried out, setting data information to be fused as m different data types, recording as a data set B= (B1,B2,...,Bm), and recording an output heterogeneous network data source as a data set Z= (Z1,Z2,...,Zn) after n data types are converted, wherein the confidence level of the communication data sources of different heterogeneous networks is set to be qj; by this arrangement, identification of different data information can be achieved.
(2) The heterogeneous network communication data sources converted through data operation are collected into a data set:
Zji＝(bji,cji,dji) (12)
In the formula (1), the conversion degree of the jth heterogeneous network communication data source and the ith decision object is recorded as Zji; the conversion degree of the multi-source heterogeneous data information is expressed by a triangle fuzzy number function, and the function formula is recorded as (bji,cji,dji) and comprises the following steps: bji≤cji≤dji is more than or equal to 0 and less than or equal to 1;
(2) In order to balance a data conversion function and improve the data computing capacity, OWA operator weight vectors are set in a function formula, reasonable fuzzy semantic quantization standards are selected according to the preference of a decision maker, and OWA operator weight vectors are set; the degree of data conversion is divided into 'complete conversion', 'unconverted' and 'in-conversion', 'complete conversion', 'unconverted' data parameter set to (0.45,0.79) 'unconverted' data parameter set to (0.16,0.32) 'in-conversion', 'in-conversion' data parameter set to (0.78,1.03), and g (y) can be recorded by acquiring fuzzy semantic quantization operator functions according to parameters; according to the set function g (y), calculating an OWA operator weight vector and recording the vector asWherein m represents the number of communication data sources input into the heterogeneous network;
(3) After setting the data parameters, carrying out data conversion on the confidence qj of each heterogeneous network communication data source and the approval level value zji;
in the method of utilizing and increasing the calculated amount of OWA weight vectors, setting each decision value of qj and zji, and then carrying out size arrangement on the decision values in a data conversion mode, wherein a data conversion formula can be recorded as:
zji-min＝qj×zji (13)
zji-max＝qj+zji-qj×zji (14)
(4) And realizing heterogeneous network communication data fusion according to the OWA operator weight vector and the converted acceptance level, and simultaneously calculating the final decision value of each decision. Then there are:
Where the j-th maximum element is set to cji.
(5) And then continuously inputting network heterogeneous data information, and fusing the data information according to the set decision value, so as to realize heterogeneous network communication data fusion.
While specific embodiments of the present invention have been described above, it will be understood by those skilled in the art that these specific embodiments are by way of example only, and that various omissions, substitutions, and changes in the form and details of the methods and systems described above may be made by those skilled in the art without departing from the spirit and scope of the invention. For example, it is within the scope of the present invention to combine the above-described method steps to perform substantially the same function in substantially the same way to achieve substantially the same result. Accordingly, the scope of the invention is limited only by the following claims.