Multi-element network equipment topology optimization method based on clustering analysis

Abstract

The invention relates to the technical field of multi-element network equipment topological optimization, in particular to a multi-element network equipment topological optimization method based on clustering analysis. Comprising the following steps: setting a plurality of network nodes based on historical network topology data, and generating a communication evaluation value of each network node; setting a plurality of center nodes according to all the communication evaluation values, and establishing a communication topological structure according to all the center nodes; the method comprises the following steps: acquiring a feedback data packet according to a preset update time node, judging whether to update a communication topological structure according to the feedback data packet, establishing a plurality of network nodes based on historical network topological data, analyzing communication path information among the network nodes, effectively clustering the network nodes, and updating the communication topological structure according to the network nodes. According to the method, a large-scale network is divided into a plurality of small areas and clusters, and each cluster is iteratively optimized to construct a communication topological structure, so that an optimal communication path between network nodes is generated, and the communication efficiency between the network nodes is improved.

Description

Technical Field

[0001] The present application relates to the technical field of multi-element network device topology optimization, and in particular to a multi-element network device topology optimization method based on cluster analysis. Background Art

[0002] Network topology refers to the physical layout of various network devices interconnected by transmission media. An unreasonable topology can easily lead to low communication efficiency or poor communication quality between network nodes. Currently, with the increasing scale and complexity of network topologies, the requirements for network topology layout are becoming increasingly stringent. Therefore, a multivariate network device topology optimization method based on cluster analysis is urgently needed to optimize the network topology and improve the communication efficiency and quality between network devices. Summary of the Invention

[0003] The purpose of this application is: to solve the above technical problems, this application provides a multi-network device topology optimization method based on cluster analysis, aiming to improve the communication efficiency between various network devices.

[0004] In some embodiments of the present application, multiple network nodes are established based on historical network topology data, and by analyzing the communication path information between each network node, the network nodes are effectively clustered, and the large network is divided into several smaller areas and clusters. The communication topology structure is constructed by iteratively optimizing each cluster, thereby generating the optimal communication path between each network node and improving the communication efficiency between each network node.

[0005] In some embodiments of the present application, by establishing an update cycle, the communication topology structure is periodically updated to avoid interference with the overall communication efficiency due to fluctuations in network equipment and changes in communication requirements, thereby ensuring the communication quality between various network nodes.

[0006] In some embodiments of the present application, a multivariate network device topology optimization method based on cluster analysis is provided, including: Setting multiple network nodes based on historical network topology data and generating a communication evaluation value for each network node; Setting multiple central nodes according to all communication evaluation values, and establishing a communication topology structure based on all central nodes; Obtaining feedback data packets according to a preset update time node, and determining whether to update the communication topology structure based on the feedback data packets; The setting of network nodes includes: Establish a network node sequence A, A=(a1,a2…a i …a n ), where a iis the i-th network node, and n is the number of network nodes.

[0007] In some embodiments of the present application, generating a communication evaluation value of each network node includes: According to the network node sequence A, set ai as the node to be evaluated; Generate the initial evaluation value p of the node to be evaluated based on the historical network topology data; p= (β i *v i ); in, is the number of communication evaluation indicators; β i is the impact factor of the i-th communication evaluation index; v i is the reference value of the i-th communication evaluation index of the node to be evaluated; Generate a communication mapping table of the node to be evaluated; Generate initial evaluation values ​​and communication mapping tables for each network node in sequence; Establish the initial evaluation value sequence P, P=(p1,p2…p i …p n ), where p i is the initial evaluation value of the i-th network node; The communication evaluation value of each network node is generated according to all communication mapping tables and the initial evaluation value sequence P.

[0008] In some embodiments of the present application, when generating the communication evaluation value of each network node, the following steps are further included: According to the network node sequence A, set ai as the target network node in sequence; Generate a communication evaluation value b of the target network node; b=e1*Q1*p'+e2*Q2*[ w i *p 1i ]; Wherein, e1 is the preset first weight coefficient; e2 is the preset second weight coefficient; Q1 is the preset first fixed coefficient; Q2 is the preset second fixed coefficient; p' is the initial evaluation value of the target network node; θ2 is the number of associated nodes of the target network node obtained based on the communication mapping table of the target network node; w i is the impact factor of the i-th associated node of the target network node; p 1i is the initial evaluation value of the i-th associated node of the target network node; Generate communication evaluation values ​​of each network node in sequence; Establish a communication evaluation value sequence B, B=(b1,b2…b i …b n ), where bi It is the communication evaluation value of the i-th network node.

[0009] In some embodiments of the present application, multiple central nodes are set according to all communication evaluation values, including: Preset a communication evaluation value threshold B1; If bi > B1, set the i-th network node as a central node; If bi < B1, set the i-th network node as an edge node; Obtain all central nodes and establish a central node sequence A1, A1=(a 11 , a 12 … a 1i … a 1n1 ), where a 1i is the i-th central node; n1 is the number of central nodes; Obtain all edge nodes and establish an edge node sequence A2, A2=(a 21 , a 22 … a 2i … a 2n2 ), where a 2i is the i-th edge node; n2 is the number of edge nodes, where n2 > n1.

[0010] In some embodiments of the present application, when establishing a communication topology structure according to all central nodes, it includes: Set a 2i as the target edge node in sequence according to the edge node sequence A2; Generate the association evaluation value between the target edge node and each central node; Establish an association evaluation value sequence C of the target edge node, C=(c1, c2… c i … c n1 ), where c i is the association evaluation value between the target edge node and the i-th central node; Establish the association relationship between the central node corresponding to the maximum value c max in the association evaluation value sequence C and the target edge node; Establish an association mapping table based on all association relationships; Set a 1i as the target central node in sequence according to the central node sequence A1; Obtain all associated edge nodes of the target central node according to the association mapping table and generate the initial sub-communication structure of the target central node; Establish the initial sub-communication structures of each central node in sequence and generate the operation evaluation values of each initial sub-communication structure; Generate a correction strategy according to all operation evaluation values and establish a communication topology structure according to the correction result.

[0011] In some embodiments of the present application, generating an initial sub-communication structure of a target central node includes: Generate multiple sub-topologies based on the target central node and all associated edge nodes; Establish a sub-topological structure sequence D, D=(d1, d2…d i …d m ), where d i is the i-th sub-topology structure established based on the target central node; m is the number of sub-topology structures; Set di as the target sub-topology in turn; Generate the running evaluation value f of the target sub-topology; Generate the operation evaluation value of each sub-topology structure in turn; Establish the running evaluation value series F, F=(f1, f2…f i …f m ), where f i is the running evaluation value of the i-th sub-topology structure; Set the maximum value f in the running evaluation value sequence F max The corresponding sub-topology structure is the initial sub-communication structure of the target central node.

[0012] In some embodiments of the present application, generating an operation evaluation value f of a target sub-topology structure includes: f=e3*Q3*[ (η i *j i )]+e4*Q4*[ s i ]; Among them, e3 is the preset third weight coefficient; e4 is the preset fourth weight coefficient; Q3 is the preset third fixed coefficient; Q4 is the preset fourth fixed coefficient; θ3 is the number of structural evaluation indicators; η i is the influencing factor of the i-th structural evaluation index; j i is the reference value of the i-th structural evaluation index in the target sub-topology structure; n3 is the number of associated edge nodes of the central node in the target sub-topology structure; s i is the expected efficiency value of the i-th associated edge node in the target sub-topology.

[0013] In some embodiments of the present application, generating a correction strategy based on all operational evaluation values ​​includes: Preset operation evaluation value threshold f '; Obtain the operation evaluation value of each initial sub-communication structure and establish the operation evaluation value sequence F1, F1=(f 11 ,f 12 ,f 1i…f 1n1 ), where f 1i is the running evaluation value of the initial sub-communication structure corresponding to the i-th central node; n1 is the number of initial sub-communication structures; If f 1i < f', generate a correction instruction for the initial communication sub-structure corresponding to the i-th central node; If f 1i > f', do not generate a correction instruction for the initial communication sub-structure corresponding to the i-th central node; Generate multiple first-level sub-communication structures according to all correction instructions; Establish a communication topology structure according to the sequence H of all first-level sub-communication structures.

[0014] In some embodiments of the present application, judging whether to correct the communication topology structure according to the feedback data packet includes: Preset multiple monitoring periods, and set the end time node of each monitoring period as the update time node; Obtain the feedback data packet at the current update time node; Establish the sequence H of first-level sub-communication structures at the current update time node, H = (h1, h2…h i …h n4 ), where h i is the i-th first-level sub-communication structure; n4 is the number of first-level sub-communication structures at the current update time node; Generate the communication efficiency values of each first-level sub-communication structure at the current update time node; Establish the sequence G of communication efficiency values, G = (g1, g2…g i …g n4 ), where g i is the communication efficiency value of the i-th first-level sub-communication structure at the current update time node; Preset the communication efficiency value threshold G1; If g i < G1, generate an update instruction for the i-th first-level sub-communication structure; Generate an update strategy at the current update time node according to all update instructions.

[0015] In some embodiments of the present application, generating the communication efficiency values of each first-level sub-communication structure at the current update time node includes: Set h i as the target first-level sub-communication structure in sequence according to the sequence H of first-level sub-communication structures; Generate the communication efficiency value g of the target first-level sub-communication structure; g = r i * t i ; Among them, u1 is the number of efficiency evaluation indicators; r i is the influencing factor of the i-th efficiency evaluation index; t i is the reference value of the i-th efficiency evaluation index; Generate the communication efficiency value of each first-level sub-communication structure in turn.

[0016] Based on historical network topology data, multiple network nodes are established. By analyzing the communication path information between each network node, the network nodes are effectively clustered, and the large network is divided into several smaller areas and clusters. The communication topology structure is constructed by iteratively optimizing each cluster, thereby generating the optimal communication path between each network node and improving the communication efficiency between each network node.

[0017] By establishing an update cycle, the communication topology structure is periodically updated to avoid interference with the overall communication efficiency caused by fluctuations in network equipment and changes in communication needs, and to ensure the communication quality between each network node.

[0018] Compared with the prior art, the multivariate network device topology optimization method based on cluster analysis in the embodiment of the present application has the following advantages: Based on historical network topology data, multiple network nodes are established. By analyzing the communication path information between each network node, the network nodes are effectively clustered, and the large network is divided into several smaller areas and clusters. The communication topology structure is constructed by iteratively optimizing each cluster, thereby generating the optimal communication path between each network node and improving the communication efficiency between each network node.

[0019] By establishing an update cycle, the communication topology structure is periodically updated to avoid interference with the overall communication efficiency caused by fluctuations in network equipment and changes in communication needs, and to ensure the communication quality between each network node. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a flow chart of a multi-network device topology optimization method based on cluster analysis in a preferred embodiment of the present application. DETAILED DESCRIPTION

[0021] The following embodiments are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0022] In the description of this application, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on this application.

[0023] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature specified as "first" or "second" may explicitly or implicitly include one or more of such features. Throughout this application, unless otherwise specified, "plurality" means two or more.

[0024] In the description of this application, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they can refer to fixed connections, detachable connections, or integral connections; mechanical connections or electrical connections; direct connections or indirect connections through an intermediate medium; and internal connections between two components. Those skilled in the art will understand the specific meanings of the above terms in this application based on the specific circumstances.

[0025] like Figure 1 As shown, a multivariate network device topology optimization method based on cluster analysis in a preferred embodiment of the present application is characterized by comprising: S101: setting a plurality of network nodes based on historical network topology data, and generating a communication evaluation value of each network node; S102: setting a plurality of central nodes according to all communication evaluation values, and establishing a communication topology structure according to all central nodes; S103: Obtaining a feedback data packet according to a preset update time node, and determining whether to update the communication topology structure according to the feedback data packet; The setting of network nodes includes: Establish a network node sequence A, A=(a1,a2…a i …a n ), where a i is the i-th network node, and n is the number of network nodes.

[0026] Specifically, it uses SNMP, ICMP, sockets, PING, Traceroute, MIB and other technologies to automatically discover and collect network topology information from near to far, in an expansive and user-interactive manner, thereby generating historical network topology data.

[0027] Specifically, the communication evaluation value of each network node is generated, including: According to the network node sequence A, set ai as the node to be evaluated; Generate the initial evaluation value p of the node to be evaluated based on the historical network topology data; p= (β i *v i ); in, is the number of communication evaluation indicators; β i is the impact factor of the i-th communication evaluation index; v i is the reference value of the i-th communication evaluation index of the node to be evaluated; Generate a communication mapping table of the node to be evaluated; Generate initial evaluation values ​​and communication mapping tables for each network node in sequence; Establish the initial evaluation value sequence P, P=(p1,p2…p i …p n ), where p i is the initial evaluation value of the i-th network node; The communication evaluation value of each network node is generated according to all communication mapping tables and the initial evaluation value sequence P.

[0028] Specifically, the communication evaluation indicators include but are not limited to parameters such as the data interaction volume, data output volume and data reception volume of network nodes. The larger the communication evaluation value, the more frequent the data interaction of the current network nodes.

[0029] Specifically, whether there is communication interaction between each network node and the node to be evaluated is determined based on the historical communication data of the node to be evaluated, and a communication mapping table of the node to be evaluated is established based on all communication interaction relationships.

[0030] Specifically, when generating the communication evaluation value of each network node, it also includes: According to the network node sequence A, set ai as the target network node in sequence; Generate a communication evaluation value b of the target network node; b=e1*Q1*p'+e2*Q2*[ w i *p 1i ]; Wherein, e1 is the preset first weight coefficient; e2 is the preset second weight coefficient; Q1 is the preset first fixed coefficient; Q2 is the preset second fixed coefficient; p' is the initial evaluation value of the target network node; θ2 is the number of associated nodes of the target network node obtained based on the communication mapping table of the target network node; w iis the influence factor of the i-th associated node of the target network node; p 1i is the initial evaluation value of the i-th associated node of the target network node; Generate the communication evaluation values of each network node in sequence; Establish a communication evaluation value sequence B, B = (b1, b2…b i …b n ), where b i is the communication evaluation value of the i-th network node.

[0031] Specifically, normalize all output parameters in the model by presetting the first fixed coefficient and the second fixed coefficient, so that each parameter in the model is within the same value range.

[0032] Specifically, the higher its communication evaluation value, the greater the data communication interaction volume and the number of network nodes that can be mapped by the current network node. The higher the reliability of the current network node as the cluster center, the higher its overall interaction efficiency.

[0033] It can be understood that in the above embodiments, by analyzing historical network topology data, multiple network nodes are constructed, and each network node is separated by a clustering algorithm to construct multiple clusters, and each cluster is optimized and iterated to generate the optimal communication path between each network node, improving the communication efficiency between each network node.

[0034] In the preferred embodiment of the present application, set multiple central nodes according to all communication evaluation values, including: Preset a communication evaluation value threshold B1; If bi > B1, set the i-th network node as a central node; If bi < B1, set the i-th network node as an edge node; Obtain all central nodes and establish a central node sequence A1, A1 = (a 11 , a 12 …a 1i …a 1n1 ), where a 1i is the i-th central node; n1 is the number of central nodes; Obtain all edge nodes and establish an edge node sequence A2, A2 = (a 21 , a 22 …a 2i …a 2n2 ), where a 2i is the i-th edge node; n2 is the number of edge nodes, where n2 > n1.

[0035] Specifically, the communication evaluation value threshold B1 can be set according to historical parameters, and the number of central nodes is much smaller than the number of edge nodes.

[0036] Specifically, when establishing a communication topology based on all central nodes, it includes: According to the edge node sequence A2, set a 2i is the target edge node; Generate the correlation evaluation value between the target edge node and each central node; Establish the associated evaluation value sequence C of the target edge node, C=(c1,c2…c i …c n1 ), where c i is the association evaluation value between the target edge node and the i-th center node; Establish the maximum value c in the associated evaluation value sequence C max The association relationship between the corresponding central node and the target edge node; Establish an association mapping table based on all association relationships; According to the central node sequence A1, set a 1i is the target center node; Obtain all associated edge nodes of the target central node according to the association mapping table, and generate an initial sub-communication structure of the target central node; Establishing the initial sub-communication structure of each central node in turn, and generating the operation evaluation value of each initial sub-communication structure; Generate a correction strategy based on all operation evaluation values, and establish a communication topology structure based on the correction results.

[0037] Specifically, by separating all network nodes through each central node, edge nodes within a single cluster can communicate with each other, and edge nodes in different clusters establish communication paths through their respective central nodes, thereby generating point-to-point communication paths between each network node.

[0038] Specifically, a communication path is built between the two central nodes, and the edge nodes in different clusters communicate through the central node.

[0039] It can be understood that in the above embodiments, by analyzing the communication path information between each network node, the network nodes are effectively clustered, the large network is divided into several smaller areas and clusters, and the communication topology structure is constructed by iteratively optimizing each cluster, thereby generating the optimal communication path between each network node and improving the communication efficiency between each network node.

[0040] In a preferred embodiment of the present application, generating an initial sub-communication structure of a target central node includes: Generate multiple sub-topologies based on the target central node and all associated edge nodes; Establish a sub-topological structure sequence D, D=(d1, d2…d i …d m ), where d i is the i-th sub-topology structure established based on the target central node; m is the number of sub-topology structures; Set di as the target sub-topology in turn; Generate the running evaluation value f of the target sub-topology; Generate the operation evaluation value of each sub-topology structure in turn; Establish the running evaluation value series F, F=(f1, f2…f i …f m ), where fi is the operation evaluation value of the i-th sub-topology structure; Set the maximum value f in the running evaluation value sequence F max The corresponding sub-topology structure is the initial sub-communication structure of the target central node.

[0041] Specifically, after establishing the communication paths between the central node and each edge node, communication paths between the edge nodes are randomly established, and multiple sub-topology structures are established according to the generated results, wherein the overall communication structure in each sub-topology structure is different.

[0042] Specifically, generating the running evaluation value f of the target sub-topology structure includes: f=e3*Q3*[ (η i *j i )]+e4*Q4*[ s i ]; Among them, e3 is the preset third weight coefficient; e4 is the preset fourth weight coefficient; Q3 is the preset third fixed coefficient; Q4 is the preset fourth fixed coefficient; θ3 is the number of structural evaluation indicators; η i is the influencing factor of the i-th structural evaluation index; j i is the reference value of the i-th structural evaluation index in the target sub-topology structure; n3 is the number of associated edge nodes of the central node in the target sub-topology structure; s i is the expected efficiency value of the i-th associated edge node in the target sub-topology.

[0043] Specifically, the structure evaluation indicators include, but are not limited to, multiple parameters such as the number of network nodes within the structure, the amount of communication data, the load pressure on the central node, the number of internal communication paths, and path security. The expected efficiency value refers to the total efficiency value when the associated edge node communicates and interacts with each edge node within the current sub-topology structure. The larger the expected efficiency value, the higher the overall communication efficiency of the current associated edge node.

[0044] Specifically, all parameters in the model are normalized by presetting a third fixed coefficient and a fourth fixed coefficient, so that each parameter in the model is within the same value range.

[0045] Specifically, the larger the operation evaluation value, the higher the adaptability of the target sub-topology structure to the current cluster, and the higher the overall communication efficiency within the cluster.

[0046] Specifically, a correction strategy is generated based on all operation evaluation values, including: Presetting an operation evaluation value threshold f'; Obtain the operation evaluation values of each initial sub-communication structure, and establish an operation evaluation value sequence F1, F1=(f 11 , f 12 , f 1i … f 1n1 ), where f 1i is the operation evaluation value of the initial sub-communication structure corresponding to the i-th central node; n1 is the number of initial sub-communication structures; If f 1i < f', generate a correction instruction for the initial communication sub-structure corresponding to the i-th central node; If f 1i > f', do not generate a correction instruction for the initial communication sub-structure corresponding to the i-th central node; Generate multiple first-level sub-communication structures according to all correction instructions; Establish a communication topology structure according to the sequence H of all first-level sub-communication structures.

[0047] Specifically, select the initial communication sub-structure to be corrected according to the correction instruction, and judge whether to remove the edge node or解除当前初始通信子结构 by iteratively optimizing its internal structure, and re-add each edge node to the remaining initial communication sub-structures, or establish a new central node and initial communication sub-structure.

[0048] Specifically, obtain all initials according to the correction result Specifically, by periodically analyzing each initial communication sub-structure, the sub-topology structures within each cluster are continuously optimized, the adaptability of each sub-topology structure to the cluster is improved, and thus the overall communication efficiency and security are improved.

[0049] In the preferred embodiment of the embodiment of the present application, judging whether to correct the communication topology according to the feedback data packet includes: Preset a plurality of monitoring periods, and set the end time node of each monitoring period as the update time node; Obtain the feedback data packet at the current update time node; Establish a first-level sub-communication structure sequence H at the current update time node, H = (h1, h2... h i …h n4 ), where h i is the i-th first-level sub-communication structure; n4 is the number of first-level sub-communication structures at the current update time node; Generate the communication efficiency value of each first-level sub-communication structure at the current update time node; Establish a communication efficiency value sequence G, G = (g1, g2... g i …g n4 ), where g i is the communication efficiency value of the i-th first-level sub-communication structure at the current update time node; Preset a communication efficiency value threshold G1; If g i < G1, generate an update instruction for the i-th first-level sub-communication structure; Generate an update strategy at the current update time node according to all the update instructions.

[0050] Specifically, generating the communication efficiency value of each first-level sub-communication structure at the current update time node includes: Set h i as the target first-level sub-communication structure in sequence according to the first-level sub-communication structure sequence H; Generate the communication efficiency value g of the target first-level sub-communication structure; g = r i *t i ; Where u1 is the number of efficiency evaluation indicators; r i sis the influence factor of the i-th efficiency evaluation indicator; t i is the reference value of the i-th efficiency evaluation indicator; Generate the communication efficiency values of each first-level sub-communication structure in sequence.

[0051] Specifically, the efficiency evaluation indicators include but are not limited to multiple parameters such as the load pressure of the central node, the data interaction frequency, and the actual communication efficiency of each edge node.

[0052] Specifically, the larger the efficiency evaluation value, the better the overall operation of the current first-level sub-communication structure, and the less likely it is that there are network nodes with communication efficiency issues within it. The smaller the communication efficiency value, the less able the current first-level sub-communication structure is to meet current communication needs.

[0053] Specifically, the communication efficiency value threshold can be set according to historical parameters.

[0054] Specifically, the update command analyzes all nodes within the first-level sub-communication structure based on current changes in communication needs, optimizing the communication paths of network nodes with low communication efficiency to avoid affecting overall communication efficiency. If a central node has problems with low communication efficiency or low communication volume, the first-level sub-communication node needs to be split, some edge nodes need to be merged into other first-level sub-communication nodes, or a new first-level sub-communication structure needs to be generated.

[0055] According to the first concept of the present application, multiple network nodes are established based on historical network topology data, and by analyzing the communication path information between each network node, the network nodes are effectively clustered, and the large network is divided into several smaller areas and clusters. The communication topology structure is constructed by iteratively optimizing each cluster, thereby generating the optimal communication path between each network node and improving the communication efficiency between each network node.

[0056] According to the second concept of this application, by establishing an update cycle, the communication topology is periodically updated to avoid interference with the overall communication efficiency due to fluctuations in network equipment and changes in communication requirements, and to ensure the communication quality between each network node. The above is only a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and replacements can be made without departing from the technical principles of the present application. These improvements and replacements should also be regarded as the scope of protection of the present application.

Claims

1. A multivariate network device topology optimization method based on cluster analysis, characterized in that: including: setting multiple network nodes based on historical network topology data and generating communication evaluation values for each network node; setting multiple central nodes according to all communication evaluation values and establishing a communication topology structure based on all central nodes; obtaining feedback data packets according to a preset update time node and determining whether to update the communication topology structure according to the feedback data packets; wherein, setting network nodes includes: Establish a network node sequence A, A=(a1,a2…a i …a n ), where a i is the i-th network node, and n is the number of network nodes.

2. The multivariate network device topology optimization method based on cluster analysis according to claim 1, characterized in that: generating communication evaluation values for each network node, including: successively setting ai as a node to be evaluated according to the network node sequence A; generating an initial evaluation value p of the node to be evaluated according to historical network topology data; p= (b i *v i ); in, is the number of communication evaluation indicators; β i is the impact factor of the i-th communication evaluation index; v i is the reference value of the i-th communication evaluation index of the node to be evaluated; generating a communication mapping table of the node to be evaluated; successively generating initial evaluation values and communication mapping tables for each network node; Establish the initial evaluation value sequence P, P=(p1,p2…p i …p n ), where p i is the initial evaluation value of the i-th network node; generating communication evaluation values for each network node according to all communication mapping tables and the initial evaluation value sequence P.

3. The multivariate network device topology optimization method based on cluster analysis according to claim 2, characterized in that: When generating communication evaluation values for each network node, it further includes: Set a in sequence according to the number of network nodes A i is the target network node; generating a communication evaluation value b of the target network node; b=e1*Q1*p'+e2*Q2*[ w i *p 1i ]; Wherein, e1 is the preset first weight coefficient; e2 is the preset second weight coefficient; Q1 is the preset first fixed coefficient; Q2 is the preset second fixed coefficient; p' is the initial evaluation value of the target network node; θ2 is the number of associated nodes of the target network node obtained based on the communication mapping table of the target network node; w i is the impact factor of the i-th associated node of the target network node; p 1i is the initial evaluation value of the i-th associated node of the target network node; successively generating communication evaluation values for each network node; Establish a communication evaluation value sequence B, B=(b1,b2…b i …b n ), where b i is the communication evaluation value of the i-th network node.

4. The multivariate network device topology optimization method based on cluster analysis according to claim 3, characterized in that: setting multiple central nodes according to all communication evaluation values, including: presetting a communication evaluation value threshold B1; if bi > B1, setting the i-th network node as a central node; if bi < B1, setting the i-th network node as an edge node; Get all the central nodes and establish the central node sequence A1, A1=(a 11 ,a 12 …a 1i …a 1n1 ), where a 1i is the i-th central node; n1 is the number of central nodes; Get all edge nodes and create edge node array A2, A2=(a 21 , a 22 …a 2i …a 2n2 ), where a 2i is the i-th edge node; n2 is the number of edge nodes, where n2>n1.

5. The multivariate network device topology optimization method based on cluster analysis according to claim 4, characterized in that: when establishing a communication topology structure based on all central nodes, it includes: According to the edge node sequence A2, set a 2i is the target edge node; generating an association evaluation value between the target edge node and each central node; Establish the associated evaluation value sequence C of the target edge node, C=(c1,c2…c i …c n1 ), where c i is the association evaluation value between the target edge node and the i-th center node; Establish the maximum value c in the associated evaluation value sequence C max The association relationship between the corresponding central node and the target edge node; establishing an association mapping table based on all association relationships; According to the central node sequence A1, set a 1i is the target center node; obtaining all associated edge nodes of the target central node according to the association mapping table and generating an initial sub-communication structure of the target central node; successively establishing initial sub-communication structures for each central node and generating operation evaluation values for each initial sub-communication structure; generating a correction strategy according to all operation evaluation values and establishing a communication topology structure according to the correction result.

6. The multivariate network device topology optimization method based on cluster analysis according to claim 5, characterized in that: Generating an initial sub-communication structure of the target central node includes: generating multiple sub-topology structures according to the target central node and all associated edge nodes; Establish a sub-topological structure sequence D, D=(d1, d2…d i …d m ), where d i is the i-th sub-topology structure established based on the target central node; m is the number of sub-topology structures; successively setting di as the target sub-topology structure; generating an operation evaluation value f of the target sub-topology structure; successively generating operation evaluation values for each sub-topology structure; Establish the running evaluation value series F, F=(f1, f2…f i …f m ), where f i is the running evaluation value of the i-th sub-topology structure; Set the maximum value f in the running evaluation value sequence F max The corresponding sub-topology structure is the initial sub-communication structure of the target central node.

7. The multivariate network device topology optimization method based on cluster analysis according to claim 6, characterized in that: generating an operation evaluation value f of the target sub-topology structure, including: f=e3*Q3*[ (η i *j i )]+e4*Q4*[ s i ]; Among them, e3 is the preset third weight coefficient; e4 is the preset fourth weight coefficient; Q3 is the preset third fixed coefficient; Q4 is the preset fourth fixed coefficient; θ3 is the number of structural evaluation indicators; η i is the influencing factor of the i-th structural evaluation index; j i is the reference value of the i-th structural evaluation index in the target sub-topology structure; n3 is the number of associated edge nodes of the central node in the target sub-topology structure; s i is the expected efficiency value of the i-th associated edge node in the target sub-topology.

8. The multivariate network device topology optimization method based on cluster analysis according to claim 7, characterized in that: generating a correction strategy according to all operation evaluation values, including: presetting an operation evaluation value threshold f'; Obtain the operation evaluation value of each initial sub-communication structure and establish the operation evaluation value sequence F1, F1=(f 11 ,f 12 ,f 1i …f 1n1 ), where f 1i is the operational evaluation value of the initial sub-communication structure corresponding to the i-th central node; n1 is the number of initial sub-communication structures; If f 1i < f', generate a correction instruction for the initial communication substructure corresponding to the i-th central node; If f 1i >f', does not generate the correction instruction of the initial communication substructure corresponding to the i-th central node; generating multiple first-level sub-communication structures according to all correction instructions; establishing a communication topology structure according to all first-level sub-communication structure sequences H.

9. The multivariate network device topology optimization method based on cluster analysis according to claim 8, characterized in that: Determining whether to correct the communication topology structure according to the feedback data packet includes: presetting multiple monitoring periods and setting the end time node of each monitoring period as the update time node; obtaining the feedback data packet of the current update time node; Establish the first-level sub-communication structure sequence H of the current update time node, H=(h1, h2…h i …h n4 ), where h i is the i-th first-level sub-communication structure; n4 is the number of first-level sub-communication structures at the current update time node; generating a communication efficiency value of each first-level sub-communication structure at the current update time node; Establish a communication efficiency value sequence G, G=(g1,g2…g i …g n4 ), where g i is the communication efficiency value of the i-th first-level sub-communication structure at the current update time node; presetting a communication efficiency value threshold G1; If g i <G1, generate an update instruction for the i-th first-level sub-communication structure; generating an update strategy at the current update time node according to all update instructions.

10. The multivariate network device topology optimization method based on cluster analysis according to claim 9, characterized in that: Generating a communication efficiency value of each first-level sub-communication structure at the current update time node includes: According to the first-level sub-communication structure sequence H, h is set in sequence i It is the target first-level sub-communication structure; generating a communication efficiency value g of the target first-level sub-communication structure; g=[ r i *t i ]; Among them, u1 is the number of efficiency evaluation indicators; r i is the influencing factor of the i-th efficiency evaluation index; t i is the reference value of the i-th efficiency evaluation index; successively generating communication efficiency values for each first-level sub-communication structure.

Images