An interconnected communication system based on multiple mesh circuits

Abstract

The present invention provides an interconnected communication system based on multiple mesh circuits, belonging to the technical field of communication network management, and comprising: a tensor acquisition module: collecting basic parameters of network nodes, and then constructing a multidimensional tensor, and expanding the multidimensional tensor to obtain an extended multidimensional tensor; a model construction module: decomposing the extended multidimensional tensor, and establishing a global network topology model based on the decomposition result; a model analysis module: analyzing the global network topology model, and determining an initial path candidate set based on the analysis result; a path determination module: selecting an optimal path from the initial path candidate set based on a preset algorithm, and using the optimal path as a main path for communication resource transmission; a fault handling module: deploying a preset monitoring tool at each network node to monitor the operating parameters of each link in real time, and when the agent detects that the link is abnormal, triggering a fault reporting mechanism and starting a fault handling process. The intelligence and reliability of network management are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of communication network management, and in particular to an interconnected communication system based on multiple mesh circuits. Background Art

[0002] As the scale and complexity of communication networks continue to expand, network performance optimization and fault handling have become key issues. Most current communication networks rely on traditional path selection algorithms and centralized management methods, which are difficult to efficiently cope with complex network topologies and dynamically changing communication needs.

[0003] In the prior art, a typical path planning method selects the optimal path based on a single-dimensional network parameter. Although this method is simple to calculate, it is difficult to comprehensively consider the impact of multi-dimensional parameters on communication performance. In addition, network fault handling mostly uses static rules, which cannot quickly adapt to dynamic fault conditions in complex network environments, resulting in extended communication interruption time and reduced network resource utilization.

[0004] Therefore, the present invention provides an interconnected communication system based on multiple mesh circuits. Summary of the invention

[0005] The present invention provides an interconnected communication system based on multiple mesh circuits, which is used to perform network modeling and analysis by utilizing tensor technology, construct a global topology through tensor mapping and expansion, and implement a path performance score calculation formula that integrates multi-dimensional link parameters to achieve optimization under dynamic network conditions. A hierarchical fault handling mechanism is designed, and the path is dynamically adjusted in combination with a tensor model and a link monitoring tool to improve fault recovery efficiency, ensure efficient transmission of communication resources and network stability, and significantly improve the intelligence and reliability of network management.

[0006] The present invention provides an interconnection communication system based on multiple mesh circuits, comprising:

[0007] Tensor acquisition module: collects basic parameters of network nodes, constructs multidimensional tensors, and expands the multidimensional tensors to obtain extended multidimensional tensors;

[0008] Model building module: decomposes the extended multi-dimensional tensor and builds a global network topology model based on the decomposition results;

[0009] Model analysis module: analyzes the global network topology model and determines the initial path candidate set based on the analysis results;

[0010] Path determination module: selects the best path from the initial path candidate set based on a preset algorithm, and uses the best path as the main path for communication resource transmission;

[0011] Fault handling module: Deploy preset monitoring tools at each network node to monitor the operating parameters of each link in real time. When the agent detects a link abnormality, it triggers the fault reporting mechanism and starts the fault handling process.

[0012] The present invention provides an interconnection communication system based on multiple mesh circuits, a tensor acquisition module, comprising:

[0013] A parameter acquisition unit: acquires basic parameters from a preset management information base of the mesh circuit based on a preset protocol, wherein the basic parameters include: node parameters and link parameters;

[0014] Tensor mapping unit: analyzes basic parameters to determine several tensor dimensions, and maps the basic parameters into multi-dimensional tensors based on preset mapping rules;

[0015] Characteristic acquisition unit: analyzing the characteristics of the mesh circuit to acquire several characteristics of the mesh circuit;

[0016] A dimension determination unit: determining a plurality of expansion dimensions based on a plurality of characteristics of the mesh circuit and a preset characteristic-dimension database;

[0017] Dimension expansion unit: determines the expansion method corresponding to each expansion dimension based on a preset dimension-extension method data table, and then expands the multidimensional tensor to obtain an expanded multidimensional tensor.

[0018] The present invention provides an interconnected communication system based on multiple mesh circuits, wherein the tensor dimension comprises: a first dimension and a second dimension, wherein the first dimension is related to the node pair, and the second dimension is related to the link parameter.

[0019] The present invention provides an interconnection communication system based on multiple mesh circuits, with preset mapping rules, including: node pair mapping rules and link parameter mapping rules.

[0020] The present invention provides an interconnection communication system based on multiple mesh circuits, a model building module, comprising:

[0021] Tensor decomposition unit: decomposes the extended multidimensional tensor based on a preset decomposition method, and then obtains the factor matrix and core tensor corresponding to the extended multidimensional tensor;

[0022] Model building unit: extract network structure data from the factor matrix, and then combine it with the core tensor to build a global network topology model.

[0023] The present invention provides an interconnection communication system based on multiple mesh circuits, a model analysis module, comprising:

[0024] Data acquisition unit: analyzes the global network topology model and obtains relevant data of each link on each path;

[0025] Score determination unit: determines the performance score of each path based on the relevant data of each link on each path:

[0026] ;in, is the performance score of the ith path, n is the number of links in the ith path, is the link weight of the jth path of the ith path, is the link bandwidth of the jth path of the ith path, is the link delay of the jth path of the ith path, is the preset adjustment factor, is the historical stability coefficient of the link of the jth path of the i-th path, is the load of the link of the jth path of the i-th path, is the congestion coefficient of the link of the jth path of the i-th path, is the weight of the congestion impact of the link on the jth path of the i-th path, is the preset weight corresponding to the dynamic attribute of the link, is the preset dynamic conversion coefficient, is the preset static conversion coefficient.

[0027] The present invention provides an interconnection communication system based on multiple mesh circuits, a fault processing module, comprising:

[0028] Parameter collection unit: deploys a preset monitoring tool at each node of the mesh circuit to collect the operating parameters of the node in real time based on a preset monitoring frequency;

[0029] Node judgment unit: judging whether a node fails based on a preset abnormal mode and a preset parameter threshold corresponding to each preset abnormal mode combined with a preset judgment algorithm;

[0030] Fault handling unit: If a fault occurs, the corresponding preset path reconstruction operation is performed according to the fault type and impact scope;

[0031] Fault judgment unit: collects the operating parameters of the node after the execution path reconstruction and judges whether the fault is solved. If it is solved, the normal working mode is restored. If it is not solved, the corresponding preset path reconstruction operation is repeatedly executed until the fault is solved.

[0032] The present invention provides an interconnection communication system based on multiple mesh circuits, a path reconstruction unit, comprising:

[0033] Parameter collection subunit: if the fault is not resolved, the dynamic operation parameters of the node are collected in real time from the monitoring tool of the preset network node based on the preset algorithm;

[0034] Optimization subunit: dynamically adjusts link weights based on real-time collected dynamic operation parameters, combined with the global network topology model and the performance score of each path;

[0035] Path reselection subunit: When the performance score of the path after dynamically adjusting the link weight is lower than the preset threshold, the path reselection process is triggered to optimize the main path based on the preset reinforcement algorithm;

[0036] Fault judgment subunit: Communication resources are transmitted based on the optimized main path, and whether a node fails is determined. If a failure occurs, the link weight is repeatedly adjusted to retransmit communication resources until the failure is resolved.

[0037] The present invention provides an interconnection communication system based on multiple mesh circuits, optimizing subunits, comprising:

[0038] Based on the dynamic operating parameters collected in real time, combined with the global network topology model and the performance score of each path, the link weight is dynamically adjusted:

[0039] ;in, Adjust the link weight of the jth path of the ith path, is the link weight of the jth path of the ith path, is the preset weight coefficient for dynamic adjustment, is the stability coefficient of the link of the jth path of the i-th path, is the preset maximum stability coefficient, is the preset load weight adjustment factor, is the preset stability weight adjustment coefficient, is the link utilization of the jth path of the ith path, is the preset link utilization adjustment coefficient, is the link bandwidth of the jth path of the ith path, is the load of the link of the jth path of the i-th path, It is the preset link utilization weight adjustment coefficient.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] By using tensor technology to model and analyze the network, building a global topology through tensor mapping and expansion, and integrating the path performance score calculation formula of multi-dimensional link parameters, we can achieve optimization under dynamic network conditions, design a hierarchical fault handling mechanism, and combine the tensor model and link monitoring tools to dynamically adjust the path, improve fault recovery efficiency, ensure the efficient transmission of communication resources and the stability of the network, and significantly improve the intelligence and reliability of network management. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0043] Figure 1 It is a structural diagram of an interconnected communication system based on multiple mesh circuits provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0045] Example 1

[0046] The embodiment of the present invention provides an interconnection communication system based on multiple mesh circuits, such as Figure 1 As shown, including:

[0047] Tensor acquisition module: collects basic parameters of network nodes, constructs multidimensional tensors, and expands the multidimensional tensors to obtain extended multidimensional tensors;

[0048] Model building module: decomposes the extended multi-dimensional tensor and builds a global network topology model based on the decomposition results;

[0049] Model analysis module: analyzes the global network topology model and determines the initial path candidate set based on the analysis results;

[0050] Path determination module: selects the best path from the initial path candidate set based on a preset algorithm, and uses the best path as the main path for communication resource transmission;

[0051] Fault handling module: Deploy preset monitoring tools at each network node to monitor the operating parameters of each link in real time. When the agent detects a link abnormality, it triggers the fault reporting mechanism and starts the fault handling process.

[0052] In this embodiment, the basic parameters of the network node refer to the basic information and status data of each node in the network, and these parameters are used to describe the characteristics of the node and its relationship with other nodes;

[0053] In this embodiment, a tensor is a mathematical representation used to describe the relationship between multi-dimensional data. In a network, a tensor can represent multiple characteristics of nodes and links and their interactions. For example, a three-dimensional tensor: dimension 1 represents nodes, dimension 2 represents link characteristics (such as bandwidth, delay), and dimension 3 represents time changes. The data structure is as follows: T[i,j,k] represents a certain attribute value of the i-th node and the j-th node at time k;

[0054] In this embodiment, the extended multidimensional tensor adds more dimensions or data content to the original multidimensional tensor to express more complex network information. For example, a new dimension is added to the three-dimensional tensor to represent security (such as link encryption level) to obtain a four-dimensional tensor. The extended tensor more comprehensively represents multidimensional characteristics such as network performance, security, and energy consumption.

[0055] In this embodiment, the decomposition result is the output of mathematical decomposition (such as matrix decomposition, tensor decomposition) of the extended multidimensional tensor, which is used to extract key features. For example, CP decomposition is used to extract the main patterns in the tensor, such as the optimal node set or core link combination. The result obtained after decomposition can be used to reduce the computational complexity and extract the key characteristics of the global network.

[0056] In this embodiment, the global network topology model is a network structure diagram established based on the tensor decomposition results, which is used to describe the nodes of the entire network and the connection relationships between them. The topology diagram contains node and link weights (such as delay, bandwidth, etc.), which intuitively reflects the global state of the network. Through the model, it is easier to identify network bottlenecks or optimize path selection.

[0057] In this embodiment, the initial path candidate set is a set of potential communication paths obtained by analysis based on the global network topology model, which provides a basis for subsequent path optimization, for example, a set of all possible paths from the source node to the target node, such as path 1 (A->B->C), path 2 (A->D->C), each path has different performance indicators (such as delay, reliability).

[0058] In this embodiment, the optimal path is a path selected from the initial path candidate set based on a preset algorithm (such as a shortest path algorithm or a multi-objective optimization algorithm) and has the best overall performance. For example, considering the minimum delay and maximum bandwidth, path 2 (A->D->C) is selected as the optimal path.

[0059] In this embodiment, link abnormality refers to abnormal deviation or failure of performance indicators of links in the network, which may cause communication interruption or quality degradation. For example, the bandwidth of the link suddenly drops to zero, or the delay increases sharply. The reasons may be physical failure, congestion, attack, etc.

[0060] In this embodiment, the fault handling process is a systematic process triggered when the monitoring tool detects a link abnormality, which is used to quickly locate the problem and restore network functions, for example, detecting an abnormality -> sending a fault report -> reselecting an alternative path -> updating the network topology model -> restoring communication.

[0061] The beneficial effects of the above technical solution are: by utilizing tensor technology for network modeling and analysis, building a global topology through tensor mapping and expansion, integrating the path performance score calculation formula of multi-dimensional link parameters, achieving optimization under dynamic network conditions, designing a hierarchical fault handling mechanism, combining tensor models and link monitoring tools to dynamically adjust paths, improving fault recovery efficiency, ensuring efficient transmission of communication resources and network stability, and significantly improving the intelligence and reliability of network management.

[0062] Example 2

[0063] The embodiment of the present invention provides an interconnection communication system based on multiple mesh circuits, a tensor acquisition module, including:

[0064] A parameter acquisition unit: acquires basic parameters from a preset management information base of the mesh circuit based on a preset protocol, wherein the basic parameters include: node parameters and link parameters;

[0065] Tensor mapping unit: analyzes basic parameters to determine several tensor dimensions, and maps the basic parameters into multi-dimensional tensors based on preset mapping rules;

[0066] Characteristic acquisition unit: analyzing the characteristics of the mesh circuit to acquire several characteristics of the mesh circuit;

[0067] A dimension determination unit: determining a plurality of expansion dimensions based on a plurality of characteristics of the mesh circuit and a preset characteristic-dimension database;

[0068] Dimension expansion unit: determines the expansion method corresponding to each expansion dimension based on a preset dimension-extension method data table, and then expands the multidimensional tensor to obtain an expanded multidimensional tensor.

[0069] In this embodiment, the preset protocol refers to a standardized communication rule defined for acquiring and exchanging basic parameters of network nodes and links, ensuring that the tensor acquisition module can effectively extract data from the management information base of the mesh circuit, for example, SNMP (Simple Network Management Protocol): used to collect node and link status information, NETCONF (Network Configuration Protocol): used to extract configuration parameters of network devices.

[0070] In this embodiment, mesh circuit characteristics refer to key attributes that reflect the structure, behavior or performance of nodes and links in the network, and are used to analyze network topology and communication performance, including: link characteristics: bandwidth, delay, packet loss rate; node characteristics: processing power, energy consumption, storage capacity; topology characteristics: node connectivity and clustering coefficient.

[0071] In this embodiment, the extended dimension is a further extension of the original multidimensional tensor, which is used to represent additional attributes not included in the mesh circuit characteristics and improve the completeness of network modeling, including: time dimension: introducing time-varying characteristics to analyze dynamic network behavior, regional dimension: representing the geographical distribution of nodes and links in the network, load dimension: representing the link load distribution in different time periods or scenarios.

[0072] In this embodiment, the expansion method refers to the process of selecting a specific algorithm or strategy to expand the original multidimensional tensor according to the nature and characteristics of the expansion dimension to generate an expanded multidimensional tensor. For example: linear expansion: adding a time dimension to each node according to time series data, hierarchical expansion: grouping nodes in different regions and expanding the geographic dimension, and nested expansion: adding a role dimension for node functional roles (such as core, access).

[0073] The beneficial effects of the above technical solution are: through parameter acquisition, tensor mapping and dimensional expansion, efficient mapping and dynamic expansion of basic parameters to extended multidimensional tensors are achieved, and combined with characteristic analysis and dimensional expansion, the flexibility of network characteristic expression and the accuracy of data processing are improved, supporting the intelligent management and optimization of complex networks.

[0074] Example 3

[0075] An embodiment of the present invention provides an interconnected communication system based on multiple mesh circuits, wherein the tensor dimension includes: a first dimension and a second dimension, wherein the first dimension is related to the node pair, and the second dimension is related to the link parameter.

[0076] In this embodiment, the first dimension represents all possible node pair combinations in the network, which is one of the core characteristics of the network topology. There may be a direct connection or indirect association between every two nodes in the network. Through the first dimension, all node pairs can be arranged in order to form the basic unit of the tensor. If there are N nodes in the network, the size of the first dimension is C(N,2) (the number of node pair combinations). Assuming that there are 4 nodes in the network: A, B, C, D, all possible node pairs are AB, AC, AD, BC, BD, CD. The first dimension will contain 6 elements corresponding to these node pairs;

[0077] In this embodiment, the second dimension records the link parameters corresponding to each node pair, which are used to describe the properties of the communication link between the pair of nodes. The link parameters are used to quantify the quality of communication between nodes, including but not limited to the following indicators: Bandwidth: the maximum rate at which data can be transmitted (unit: Mbps / Gbps). Delay: the time it takes for data to be transmitted from one node to another (unit: ms). Packet Loss Rate: the proportion of data packets lost during transmission (unit: %). Stability: the reliability of the link running for a long time, usually a coefficient of 0-1. If the link parameter type is M, the size of the second dimension is M.

[0078] The beneficial effect of the above technical solution is: by dividing the tensor dimension into a first dimension related to node pairs and a second dimension related to link parameters, a unified representation of network topology and link characteristic data is achieved, thereby improving the expression ability of multi-dimensional data and the accuracy of network analysis.

[0079] Example 4

[0080] The embodiment of the present invention provides an interconnection communication system based on multiple mesh circuits, with preset mapping rules, including: node pair mapping rules and link parameter mapping rules.

[0081] In this embodiment, the node pair mapping rule is as follows: according to the adjacency matrix (node ​​connection relationship) of the network, a node pair index (i, j) is generated. If there is a link between the nodes, the parameter value is recorded; otherwise, it is filled with 0 or NaN (depending on the specific situation);

[0082] In this embodiment, the link parameter mapping rule is: collect characteristic data of each link and map them to the second dimension of the tensor respectively: link bandwidth→first layer, link delay→second layer, link packet loss rate→third layer.

[0083] The beneficial effects of the above technical solution are: through node pair mapping rules and link parameter mapping rules, the network connection relationship and link characteristics are efficiently mapped into a multi-dimensional tensor structure, realizing the structured representation and multi-dimensional characteristic analysis of complex network data, and improving the accuracy of network modeling and data processing efficiency.

[0084] Example 5

[0085] The embodiment of the present invention provides an interconnection communication system based on multiple mesh circuits, and a model building module, including:

[0086] Tensor decomposition unit: decomposes the extended multidimensional tensor based on a preset decomposition method, and then obtains the factor matrix and core tensor corresponding to the extended multidimensional tensor;

[0087] Model building unit: extract network structure data from the factor matrix, and then combine it with the core tensor to build a global network topology model.

[0088] In this embodiment, the preset decomposition method is an algorithm for mathematically decomposing the extended multidimensional tensor, including CP decomposition (CANDECOMP / PARAFAC decomposition) and Tucker decomposition;

[0089] In this embodiment, the factor matrix is ​​a low-dimensional matrix obtained during the decomposition process, each matrix corresponds to a dimension of the tensor and contains the main feature information of the dimension;

[0090] In this embodiment, the core tensor is the central data structure retained after tensor decomposition, which represents the interaction relationship between different dimensions and is usually small in scale. Assume that there is a three-dimensional tensor T[i,j,k] representing: i: node pair (first dimension, representing the combination of two nodes in the network, such as AB, AC), j: link parameter (second dimension, representing bandwidth, delay, etc.), k: time (third dimension, representing the network status at different time points). Through Tucker decomposition, the tensor T is decomposed into: factor matrix: U1: feature matrix of node pairs, indicating which node pairs are most important in the global network, U2: feature matrix of link parameters, which extracts key link performance indicators (such as the principal components of bandwidth or delay), U3: feature matrix of time, which describes the trend of the time dimension (such as the pattern of peak hours and idle hours), core tensor: represents the interaction intensity between node pairs, link parameters and time.

[0091] In this embodiment, extracting network structure data in the factor matrix is ​​to convert the factor matrix obtained by decomposition into the structural information of the actual network. For example, by analyzing the node pair feature matrix, the core nodes of the network and their relationships can be identified, and the key paths and regional divisions of the network topology can be determined. For example, extracting node pair data: the rows of U1 (node ​​pair feature matrix) represent different node pairs, and the columns represent the characteristic vectors of these node pairs. Through cluster analysis or characteristic weight sorting, important node pairs (such as high-bandwidth, low-latency key links) can be identified. Result: Determine the backbone links of the network, such as AB and CD are high-priority paths. Extract link parameter data: U2 (link parameter feature matrix) represents the weights of different link parameters. For example, the bandwidth has the highest weight, and the delay has the second highest weight. Result: It is used to optimize the measurement standard of link performance in the network, such as giving priority to high-bandwidth, low-latency links. Extract time-related data: U3 (time feature matrix) describes the changing rules of network performance in different time periods, such as congested nodes during peak hours, and is used to formulate dynamic routing strategies and adjust data transmission paths according to time.

[0092] In this embodiment, the global network topology model includes: a weighted graph representation of network nodes and links: G=(V,E,W) is an edge weight matrix, which is obtained by combining A(i,r) and B(j,r), a dynamic network mode: a dynamic network topology map is generated through the time-varying weights provided by D(t,r), key node and link identification: based on the principal component analysis of the factor matrix, key nodes and links in the network are identified, and global pattern extraction: global patterns are extracted based on core tensors to identify high communication frequency areas and network bottlenecks.

[0093] The beneficial effects of the above technical solution are: by decomposing the extended multi-dimensional tensor, extracting the factor matrix and core tensor, and applying high-dimensional data analysis to network topology modeling, it is possible to accurately extract network structure characteristics and efficiently construct a global network topology model, which significantly improves the system's adaptability and analysis capabilities for complex network structures.

[0094] Example 6

[0095] The embodiment of the present invention provides an interconnection communication system based on multiple mesh circuits, a model analysis module, including:

[0096] Data acquisition unit: analyzes the global network topology model and obtains relevant data of each link on each path;

[0097] Score determination unit: determines the performance score of each path based on the relevant data of each link on each path:

[0098] ;in, is the performance score of the ith path, n is the number of links in the ith path, is the link weight of the jth path of the ith path, is the link bandwidth of the jth path of the ith path, is the link delay of the jth path of the ith path, is the preset adjustment factor, is the historical stability coefficient of the link of the jth path of the i-th path, is the load of the link of the jth path of the i-th path, is the congestion coefficient of the link of the jth path of the i-th path, is the weight of the congestion impact of the link on the jth path of the i-th path, is the preset weight corresponding to the dynamic attribute of the link, is the preset dynamic conversion coefficient, is the preset static conversion coefficient.

[0099] In this embodiment, the historical stability coefficient is used to measure the stability of a link in long-term operation, and is usually obtained by counting the historical failure rate, fluctuation range, etc. of the link. The coefficient range is generally between 0 and 1, and the higher the value, the more stable the link. For example, in the statistics of a link in the past year, the normal operation time is 99.5%, and its historical stability coefficient is 0.995. Due to frequent maintenance or interruption of another link, its historical stability coefficient is only 0.8. High-stability links are usually given priority for critical communication paths;

[0100] In this embodiment, the dynamic conversion coefficient is used to adjust the impact strength of dynamic link attributes (such as bandwidth and delay) to adapt to real-time network changes. For example, when the network load is high, the importance of increasing bandwidth can be adjusted to balance the calculation results of the performance score, for example,

[0101] In high-load network scenarios, increase the value of the dynamic conversion coefficient so that the performance score focuses more on bandwidth and real-time response, thereby giving priority to high-bandwidth, low-latency links. In stable network scenarios, reduce the value of the dynamic conversion coefficient so that the performance score is more comprehensive and avoids over-reliance on dynamic attributes.

[0102] In this embodiment, the static conversion coefficient is used to adjust the impact strength of the static characteristics of the link (such as load, congestion), which is often determined based on the network infrastructure and historical statistical data. By controlling the static conversion coefficient, the contribution of static and dynamic factors to the performance score can be balanced. In the early stage of network design or when the topology is fixed, the value of the static conversion coefficient is increased so that the performance score pays more attention to the congestion and load level of the link, and the overall path selection is optimized. In a rapidly changing network environment, the static conversion coefficient is reduced to reduce the impact of static properties and make dynamic adjustment more flexible.

[0103] The beneficial effects of the above technical solution are: through global topology analysis and a performance score calculation model based on multi-dimensional link attributes, accurate quantification and dynamic evaluation of path performance are achieved, and dynamic and static factors such as link weight, bandwidth, delay, and historical stability are comprehensively considered, which significantly improves the scientific nature of path optimization and the stability of system operation.

[0104] Example 7

[0105] The embodiment of the present invention provides an interconnection communication system based on multiple mesh circuits, and a fault processing module, including:

[0106] Parameter collection unit: deploys a preset monitoring tool at each node of the mesh circuit to collect the operating parameters of the node in real time based on a preset monitoring frequency;

[0107] Node judgment unit: judging whether a node fails based on a preset abnormal mode and a preset parameter threshold corresponding to each preset abnormal mode combined with a preset judgment algorithm;

[0108] Fault handling unit: If a fault occurs, the corresponding preset path reconstruction operation is performed according to the fault type and impact scope;

[0109] Fault judgment unit: collects the operating parameters of the node after the execution path reconstruction and judges whether the fault is solved. If it is solved, the normal working mode is restored. If it is not solved, the corresponding preset path reconstruction operation is repeatedly executed until the fault is solved.

[0110] In this embodiment, the preset abnormal mode is a series of problem modes defined in advance according to the common fault types and characteristics in network operation, which is used to quickly identify the abnormalities that may occur in network nodes or links. Each abnormal mode is bound to a specific parameter threshold, such as high latency, bandwidth utilization abnormality, etc. For example, in network monitoring, it is found that the delay of node A suddenly rises to 150 ms, exceeding the preset high latency mode threshold (100 ms). Therefore, the system determines that node A may be in "high latency mode" and triggers the fault handling process.

[0111] In this embodiment, the preset judgment algorithm is a set of algorithms used to comprehensively analyze whether the node operation parameters conform to the abnormal mode. It combines the parameter data and the threshold value to realize automatic fault judgment through logical operation or statistical model, including: a simple threshold comparison algorithm is used for single parameter abnormality judgment, rule: if the monitored value > the preset threshold value, it is judged as abnormal, a multi-parameter weighted judgment algorithm is used for joint fault judgment of multiple parameters, rule: assign weights to each parameter, and a time series anomaly detection algorithm is used to monitor the trend changes of parameters, rule: use sliding average or machine learning model to judge whether the parameters deviate from the historical normal value.

[0112] In this embodiment, the preset path reconstruction operations include: minor faults (such as link performance degradation): dynamically adjust link parameters (such as priority, resource allocation) to alleviate the problem, and redistribute traffic to avoid overload. For serious faults (such as link interruption), start the path reconstruction process and reselect an alternative path in the global network topology; notify the relevant nodes or administrators to repair the physical link. For multiple link failures: use a distributed fault recovery algorithm to prioritize the recovery of the path with the largest impact range.

[0113] The beneficial effects of the above technical solution are: through real-time parameter collection, abnormal mode judgment and hierarchical path reconstruction operations, efficient fault location and processing are achieved, combined with preset judgment algorithms and dynamic path reconstruction strategies, the targeted and continuous fault processing is ensured, and the system's fault repair efficiency and network communication stability are significantly improved.

[0114] Example 8

[0115] An embodiment of the present invention provides an interconnection communication system based on multiple mesh circuits, a path reconstruction unit, including:

[0116] Parameter collection subunit: if the fault is not resolved, the dynamic operation parameters of the node are collected in real time from the monitoring tool of the preset network node based on the preset algorithm;

[0117] Optimization subunit: dynamically adjusts link weights based on real-time collected dynamic operation parameters, combined with the global network topology model and the performance score of each path;

[0118] Path reselection subunit: When the performance score of the path after dynamically adjusting the link weight is lower than the preset threshold, the path reselection process is triggered to optimize the main path based on the preset reinforcement algorithm;

[0119] Fault judgment subunit: Communication resources are transmitted based on the optimized main path, and whether a node fails is determined. If a failure occurs, the link weight is repeatedly adjusted to retransmit communication resources until the failure is resolved.

[0120] In this embodiment, the preset algorithm refers to the calculation method or logic rule set in advance according to the network structure and operation requirements when designing the system, which is used to analyze and process the collected node or link parameters. The algorithm can be used to monitor, evaluate and adjust the network status, such as calculating the path weight, judging the link status, etc., including: the shortest path algorithm (Dijkstra): used to calculate the optimal path from the starting node to the ending node, the weight is usually the delay or cost of the link, the link quality evaluation algorithm: the path weight is updated in real time according to the dynamic operation parameters of the link (such as delay, bandwidth, etc.), the distributed fault detection algorithm: the nodes collaborate to analyze the operation parameters and quickly detect network faults. Assuming that there are two paths between network node A and node B, the system uses the preset algorithm to evaluate the path weight: Path 1: delay 50 ms, bandwidth 100 Mbps, packet loss rate 0.01%. The weight is 25.0302, path 2: delay 30 ms, bandwidth 50 Mbps, packet loss rate 0.02%, weight is 15.0604, according to the weight, the system selects path 2 for communication.

[0121] In this embodiment, the dynamic operation parameter refers to the node or link performance index that changes in real time during the actual operation of the network, reflects the current state of the network, and is usually used to evaluate link quality, path selection and fault judgment. Common dynamic operation parameters include delay, bandwidth, packet loss rate, and node load. For example, during a communication process, the system monitors the following dynamic operation parameters: Node A to Node B: delay 60 ms, bandwidth 90 Mbps, packet loss rate 0.02%, Node A to Node C: delay 40 ms, bandwidth 80 Mbps, and packet loss rate 0.01%. The system re-evaluates the path performance based on these parameters and adjusts the link weight to adapt to the current network state.

[0122] In this embodiment, the preset reinforcement algorithm is an optimization algorithm based on the idea of ​​reinforcement learning, which can continuously adjust the strategy through the "trial-feedback" process in a dynamic network environment to maximize the path performance score or minimize the impact of communication failures. Reinforcement learning works through a "state-action-reward" model: state: current network topology and dynamic operating parameters, action: adjust path or link weights, reward: give feedback based on path performance scores or fault recovery effects. The preset reinforcement algorithm updates the strategy through historical experience to optimize main path selection and link adjustment.

[0123] The beneficial effects of the above technical solution are: through parameter collection, dynamic adjustment of link weights and enhanced path reselection, efficient path optimization and resource transmission in a fault environment are achieved, a fault judgment and repeated optimization mechanism is introduced, and the global network topology and performance score are dynamically adjusted, which significantly improves the system's fault recovery capability and communication reliability, ensuring the stable operation of the network under complex conditions.

[0124] Example 9

[0125] The embodiment of the present invention provides an interconnection communication system based on multiple mesh circuits, an optimization subunit, including:

[0126] Based on the dynamic operating parameters collected in real time, combined with the global network topology model and the performance score of each path, the link weight is dynamically adjusted:

[0127] ;in, Adjust the link weight of the jth path of the ith path, is the link weight of the jth path of the ith path, is the preset weight coefficient for dynamic adjustment, is the stability coefficient of the link of the jth path of the i-th path, is the preset maximum stability coefficient, is the preset load weight adjustment factor, is the preset stability weight adjustment coefficient, is the link utilization of the jth path of the ith path, is the preset link utilization adjustment coefficient, is the link bandwidth of the jth path of the ith path, is the load of the link of the jth path of the i-th path, It is the preset link utilization weight adjustment coefficient.

[0128] In this embodiment, link weight adjustment refers to dynamically adjusting the weight value of each link according to the dynamic operation parameters (such as link utilization, bandwidth, load, etc.) collected in real time during the path optimization process, and re-evaluating the priority of the link by combining the preset weight coefficient (such as load coefficient, stability coefficient, etc.), thereby optimizing the path selection and ensuring the communication performance;

[0129] In this embodiment, the preset maximum stability coefficient is a maximum reference value that defines the stability of the link operation, indicating the upper limit of the stability of the link under ideal conditions, and is usually used to evaluate whether the actual stability of the link is close to the theoretical optimal value, thereby affecting the weight adjustment of the path.

[0130] The beneficial effects of the above technical solution are: by dynamically adjusting link weights, comprehensively considering multi-dimensional parameters such as link stability, utilization and load, combining the global network topology model, path optimization is achieved, and preset weight adjustment coefficients and maximum stability coefficients are introduced. Through precise dynamic weight calculation, the real-time and reliability of path selection are improved, and the network's fault resistance and communication efficiency are significantly enhanced.

[0131] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An interconnected communication system based on multiple mesh circuits, characterized in that: include: Tensor acquisition module: collects basic parameters of network nodes, constructs multidimensional tensors, and expands the multidimensional tensors to obtain extended multidimensional tensors; Model building module: decomposes the extended multi-dimensional tensor and builds a global network topology model based on the decomposition results; Model analysis module: analyzes the global network topology model and determines the initial path candidate set based on the analysis results; Path determination module: selects the best path from the initial path candidate set based on a preset algorithm, and uses the best path as the main path for communication resource transmission; Fault handling module: Deploy preset monitoring tools on each network node to monitor the operating parameters of each link in real time. When the agent detects a link anomaly, it triggers the fault reporting mechanism and starts the fault handling process. Among them, the tensor acquisition module includes: A parameter acquisition unit: acquires basic parameters from a preset management information base of the mesh circuit based on a preset protocol, wherein the basic parameters include: node parameters and link parameters; Tensor mapping unit: analyzes basic parameters to determine several tensor dimensions, and maps the basic parameters into multi-dimensional tensors based on preset mapping rules; Characteristic acquisition unit: analyzing the characteristics of the mesh circuit to acquire several characteristics of the mesh circuit; A dimension determination unit: determining a plurality of expansion dimensions based on a plurality of characteristics of the mesh circuit and a preset characteristic-dimension database; Dimension expansion unit: determines the expansion method corresponding to each expansion dimension based on a preset dimension-extension method data table, and then expands the multidimensional tensor to obtain an expanded multidimensional tensor; Among them, the model building module includes: Tensor decomposition unit: decomposes the extended multidimensional tensor based on a preset decomposition method, and then obtains the factor matrix and core tensor corresponding to the extended multidimensional tensor; Model building unit: extract network structure data from the factor matrix, and then combine it with the core tensor to build a global network topology model.

2. An interconnection communication system based on multiple mesh circuits according to claim 1, characterized in that: The tensor dimensions include: a first dimension and a second dimension, wherein the first dimension is related to the node pair, and the second dimension is related to the link parameter.

3. The interconnection communication system based on multiple mesh circuits according to claim 1, characterized in that: The preset mapping rules include: node pair mapping rules and link parameter mapping rules.

4. The interconnection communication system based on multiple mesh circuits according to claim 1, characterized in that: Model analysis module, including: Data acquisition unit: analyzes the global network topology model and obtains relevant data of each link on each path; Score determination unit: determines the performance score of each path based on the relevant data of each link on each path: ;in, is the performance score of the ith path, n is the number of links in the ith path, is the link weight of the jth path of the ith path, is the link bandwidth of the jth path of the ith path, is the link delay of the jth path of the ith path, is the preset adjustment factor, is the historical stability coefficient of the link of the jth path of the i-th path, is the load of the link of the jth path of the i-th path, is the congestion coefficient of the link of the jth path of the i-th path, is the weight of the congestion impact of the link on the jth path of the i-th path, is the preset weight corresponding to the dynamic attribute of the link, is the preset dynamic conversion coefficient, is the preset static conversion coefficient.

5. The interconnection communication system based on multiple mesh circuits according to claim 1, characterized in that: Fault handling module, including: Parameter collection unit: deploys a preset monitoring tool at each node of the mesh circuit to collect the operating parameters of the node in real time based on a preset monitoring frequency; Node judgment unit: judging whether a node fails based on a preset abnormal mode and a preset parameter threshold corresponding to each preset abnormal mode combined with a preset judgment algorithm; Fault handling unit: If a fault occurs, the corresponding preset path reconstruction operation is performed according to the fault type and impact scope; Fault judgment unit: collects the operating parameters of the node after the execution path reconstruction and judges whether the fault is solved. If so, restores the normal working mode; Path reconstruction unit: If the fault is not resolved, the weight of the link of each path is adjusted and the corresponding preset path reconstruction operation is repeatedly performed until the fault is resolved.

6. The interconnection communication system based on multiple mesh circuits according to claim 5, characterized in that: Path reconstruction unit, including: Parameter collection subunit: if the fault is not resolved, the dynamic operation parameters of the node are collected in real time from the monitoring tool of the preset network node based on the preset algorithm; Optimization subunit: dynamically adjusts link weights based on real-time collected dynamic operation parameters, combined with the global network topology model and the performance score of each path; Path reselection subunit: When the performance score of the path after dynamically adjusting the link weight is lower than the preset threshold, the path reselection process is triggered to optimize the main path based on the preset reinforcement algorithm; Fault judgment subunit: Communication resources are transmitted based on the optimized main path, and whether a node fails is determined. If a failure occurs, the link weight is repeatedly adjusted to retransmit communication resources until the failure is resolved.

7. The interconnection communication system based on multiple mesh circuits according to claim 6, characterized in that: Optimization subunits, including: Based on the dynamic operating parameters collected in real time, combined with the global network topology model and the performance score of each path, the link weight is dynamically adjusted: ;in, Adjust the link weight of the jth path of the ith path, is the link weight of the jth path of the ith path, is the preset weight coefficient for dynamic adjustment, is the stability coefficient of the link of the jth path of the i-th path, is the preset maximum stability coefficient, is the preset load weight adjustment factor, is the preset stability weight adjustment coefficient, is the link utilization of the jth path of the ith path, is the preset link utilization adjustment coefficient, is the link bandwidth of the jth path of the ith path, is the load of the link of the jth path of the i-th path, It is the preset link utilization weight adjustment coefficient.

Images