A communication complex network service end-to-end intelligent diagnosis analysis method

Abstract

This invention discloses an end-to-end intelligent diagnostic analysis method for complex communication network services, comprising the following steps: S1, establishing the analytical processing foundation for end-to-end source tracing analysis of complex communication network services, including: establishing a complex network knowledge graph model; dynamic routing analysis based on graph computation; and establishing a multi-source data integration and operation engine; S2, when a link fault is detected through dynamic routing analysis based on graph computation, the multi-source data integration and operation engine is used in conjunction with the complex network knowledge graph model to perform end-to-end source tracing analysis of network services, checking each node through which the service data passes, and accurately locating the node where performance failure occurs; S3, based on the end-to-end diagnostic link of the network service, the end-to-end link nodes are dynamically drawn, graphically displaying the location of the faulty node in the system's network topology and the relevant performance of the specific device port; this method solves the problem of automated diagnostic analysis and fault location in complex scenarios.

Description

Technical Field

[0001] This invention relates to the field of operation and maintenance of multiple systems in complex rail transit networks, and particularly to an end-to-end intelligent diagnostic and analysis method for complex communication network services. Background Technology

[0002] The reliability of communication systems and networks is a fundamental condition for the safe operation of rail transit. They typically consist of transmission communication systems supporting the backbone network and service communication systems supporting the business network. Transmission communication systems include information transmission systems, LTE wireless systems, and high-speed data networks; service communication systems include communication power supply systems, telephone systems, video surveillance systems, and clock and time synchronization systems.

[0003] Currently, although each professional communication system is equipped with a corresponding network management system for specialized monitoring and maintenance, the independent setup of these network management systems results in the inability to effectively centralize and share data. When a fault involves a complex network and spans multiple business systems, it is only possible to achieve preliminary fault location through the manual analysis of experienced experts by multiple professional maintenance personnel accessing the relevant network management terminals. Timeliness and accuracy are greatly constrained. Summary of the Invention

[0004] The purpose of this invention is to improve the detection and fault diagnosis analysis capabilities in complex communication networks.

[0005] To achieve the above objectives, this invention proposes an end-to-end intelligent diagnostic analysis method for complex communication network services, comprising the following steps:

[0006] S1. Establish the analytical processing foundation for end-to-end source tracing analysis of complex network services, including: establishing a complex network knowledge graph model; using graph computing for dynamic routing analysis to monitor service links in real time; and establishing a multi-source data integration and operation engine.

[0007] S2. When a link failure is detected through graph-based dynamic routing analysis, a multi-source data integration and operation engine is used, combined with a complex network knowledge graph model, to perform end-to-end source analysis of network services, check each node through which the service data passes, and accurately locate the node where the performance failure occurred.

[0008] S3. Based on the end-to-end diagnostic link of network services, dynamically draw the end-to-end link nodes and graphically display the location of the faulty node in the network topology of the system and the relevant performance of the specific device port.

[0009] Specifically, the establishment of the complex network knowledge graph model involves: building a knowledge graph database based on the network knowledge graph, storing multiple sets of network knowledge graphs in the knowledge graph database; and after completing the storage of the knowledge graph database, performing knowledge extraction, knowledge fusion, and knowledge computation to construct the complex network knowledge graph.

[0010] The complex network knowledge graph incorporates factors such as business network, technical design, backbone network, product defects, network topology and business ports, construction and implementation, business terminal equipment, human operation, power supply factors, environmental factors, and construction and implementation factors as fault influencing factors.

[0011] The dynamic routing analysis using graph computation for real-time monitoring of service links involves: calculating dynamic routes based on graph computation, accessing routing tables in the MIB library by calling SNMP primitives, extracting topology-related information, continuously and dynamically expanding the topology discovery range using a breadth-first search algorithm, and locating the range of service links that may be affected by faulty nodes based on the service type queried from the OID information, thereby achieving real-time monitoring of service links.

[0012] The graph-based dynamic routing analysis and real-time monitoring of business links also includes constructing a vectorized model between nodes through a graph database: dividing entity resource nodes into devices, interfaces, links, and services, and obtaining the final working path and protection path based on a pre-set list of interconnected working paths and protection paths of each site, and then using the site name and connection relationship in the site attributes.

[0013] The establishment of the multi-source data integration and operation engine includes: standard normalization of multi-source data acquisition and diagnostic analysis driven by multiple engines.

[0014] The standardization of multi-source data acquisition is as follows: unify the research and sorting out of data integration specifications, and formulate a complete set of data integration technical methods to solve the problem of data uniformity in heterogeneous systems.

[0015] The diagnostic analysis of the multiple engine-related drivers is as follows: combining waveform analysis, inference analysis, and fault tree analysis of multiple engine-related drivers, and adopting corresponding diagnostic analysis techniques for different communication network services to achieve the fusion of multiple engines for complex communication network services.

[0016] The end-to-end fault tracing analysis involves: establishing a knowledge base based on key influencing factors for fault tracing methods using a complex network knowledge graph; converting logical fault tracing rules into identifiable information rule tables; combining this with an alarm storage buffer pool to achieve alarm tracing analysis; and finally generating result alarms with relational expression capabilities.

[0017] When a communication alarm occurs, the process is as follows: the diagnostic analysis method is used to obtain the analysis results, which include the equipment affected by the fault and the end-to-end communication link status; an emergency work order for the fault is triggered; the system assesses the impact of the fault and guides the fault handling process through a process-oriented approach; on-site personnel quickly restore transmission, the fault is eliminated, and the system is restored.

[0018] Compared with existing technologies, this invention studies an end-to-end intelligent diagnostic analysis method for complex network services, solving the problems of automated diagnostic analysis and fault location in complex scenarios, thereby achieving efficient intelligent maintenance;

[0019] This method includes: modeling with complex network knowledge graphs to help achieve accurate fault location; graph computing for dynamic routing to facilitate real-time monitoring of business links; a multi-source data integration and operation engine to achieve the integration of multiple professional systems to form comprehensive operation and maintenance decisions; an end-to-end network service tracing and analysis algorithm to achieve end-to-end fault diagnosis and tracing; and dynamic visualization of tracing results to graphically display the location of fault nodes. Attached Figure Description

[0020] Figure 1 A schematic diagram illustrating the steps of an end-to-end intelligent diagnostic and analysis method for complex communication network services.

[0021] Figure 2 It is a type of network knowledge graph;

[0022] Figure 3 This is a schematic diagram of an end-to-end complex network.

[0023] Figure 4 This is a schematic diagram of an end-to-end link for a network service.

[0024] Figure 5 This is a schematic diagram of a visual topology graph;

[0025] Figure 6 This is a scenario for tracing the source of a fixed-station fault. Detailed Implementation

[0026] The following will be combined with the embodiments of the present invention. Figures 1-6 The technical solutions, structural features, objectives and effects achieved in the embodiments of the present invention will be described in detail.

[0027] It should be noted that the accompanying drawings are in a very simplified form and use non-precise proportions. They are only used to facilitate and clarify the purpose of illustrating the embodiments of the present invention, and are not intended to limit the implementation conditions of the present invention. Therefore, they have no substantial technical significance. Any modifications to the structure, changes in the proportional relationship, or adjustments to the size should still fall within the scope of the technical content disclosed in the present invention, provided that they do not affect the effects and objectives that the present invention can produce.

[0028] It should be noted that, in this invention, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only the elements expressly listed, but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus.

[0029] This embodiment discloses an end-to-end intelligent diagnostic analysis method for complex communication network services, such as... Figure 1 As shown, it includes the following steps:

[0030] S1. Establish the analytical processing foundation for end-to-end source tracing analysis of complex network services, including: establishing a complex network knowledge graph model; utilizing graph computing for dynamic routing analysis and real-time monitoring of service links; and establishing a multi-source data integration and operation engine, specifically including the following:

[0031] S11. Establish a knowledge graph model for complex networks;

[0032] Considering the lack of multi-dimensional and multi-level analytical capabilities in traditional fault diagnosis and localization, this method, based on the application scenario of complex communication networks, introduces a domain-specific knowledge graph to compensate for this shortcoming. Specifically: a data model is established based on the network knowledge graph, i.e., a knowledge graph database is built, storing multiple sets of network knowledge graphs in the knowledge graph database; after completing the storage of the knowledge graph database, further knowledge extraction, knowledge fusion, and knowledge computation are performed to construct the complex network knowledge graph. Figure 2 It demonstrates a network knowledge graph.

[0033] In this embodiment, factors such as business network, technical design, backbone network, product defects, network topology and service ports, construction and implementation, service-end equipment, human operation, power supply factors, environmental factors, and construction and implementation are incorporated as fault influencing factors into the constructed network knowledge graph for diagnosing and locating complex network service faults. This established complex network knowledge graph combines business data and real-world scenarios, leveraging the coupling effect of data correlation to address the needs of complex queries and root cause tracing, thereby achieving precise fault location across systems, down to the board and port of the alarm device.

[0034] S12. Dynamic routing analysis based on graph computing, real-time monitoring of the transmission route of business data and the operating status of related nodes;

[0035] To address the performance degradation issues that occur when the system performs a large number of join queries, such as traditional relational database joins, it is necessary to improve end-to-end routing analysis capabilities. Therefore, this method utilizes dynamic routing analysis based on graph computing to monitor the transmission routes of business data and the operational status of related nodes in real time. Specifically, it includes the following:

[0036] S121. The graph calculation dynamic routing is as follows: by calling the SNMP (Simple Network Management Protocol) primitive to access the routing table in the MIB (Management Information Base), topology-related information is extracted, and the breadth-first search algorithm is used to continuously and dynamically expand the topology discovery range. Based on the service type queried by OID information, the range of business links that may be affected by the faulty node is located, thereby realizing real-time monitoring of business links.

[0037] S122. Construct a vectorized model between nodes using a graph database. Specifically, divide entity resource nodes into devices, interfaces, links, and services. Based on a pre-set list of working paths and protection paths for each site to connect to each other, obtain the final working path and protection path from the site name and connection relationship in the site attributes. The site name and connection relationship in the site attributes are obtained from the graph database.

[0038] The processing order of steps S121 and S122 above is not important. By combining dynamic routing analysis and vectorization model based on graph computing, not only can end-to-end performance monitoring of various professional systems be closed-loop, but after diagnosing and analyzing end-to-end faults, routing adjustment strategies and network optimization suggestions can also be provided to improve the stability of services.

[0039] S13. Normalize the standards for multi-source data acquisition and establish diagnostic analysis techniques driven by multiple engines:

[0040] Since communication systems are built separately, each with its own independent architecture and data system, achieving cross-disciplinary collaborative analysis and automated intelligent diagnosis in complex networks requires standardization of multi-source data acquisition and diagnostic analysis driven by multiple engines.

[0041] The standardization of multi-source data acquisition refers to: unifying the research and sorting out of data integration specifications, and formulating a complete set of data integration technical methods to solve the problem of data uniformity in heterogeneous systems.

[0042] The aforementioned multi-engine-driven diagnostic analysis refers to the fact that for complex communication network services, single-category diagnostic analysis techniques cannot meet the needs of automated diagnostic analysis and accurate fault location. Therefore, this method combines waveform analysis, inference analysis, fault tree analysis, and other multi-engine-driven approaches. For different communication network services, corresponding diagnostic analysis techniques are adopted. By integrating multiple engines, automatic early warning can be achieved for complex communication network services, providing cross-professional maintenance suggestions, rapid and accurate fault location, and efficient operation and maintenance.

[0043] The processing order of steps S11, S12, and S13 is not important.

[0044] S2. When link failures are detected through graph-based dynamic routing analysis, diagnostic analysis techniques driven by multiple engines are used in conjunction with a complex network knowledge graph model to perform source analysis on end-to-end network service failures. Through a vectorized model, the nodes through which the service data passes are investigated one by one in reverse to accurately locate the nodes where performance failures occur.

[0045] like Figure 3 The diagram illustrates an end-to-end representation of a complex network, as shown below. Figure 4 The diagram illustrates an end-to-end link of one of the network services. The end-to-end fault tracing analysis refers to: based on an established complex network knowledge graph, conducting in-depth research on the correlation between various alarms and key influencing factors in the end-to-end link of the network service, identifying valid alarms, and forming alarm logic links; specifically, based on the established complex network knowledge graph, a knowledge base for fault tracing methods based on key influencing factors is established, the logical fault tracing rules are converted into identifiable information rule tables, and combined with an alarm storage buffer pool, alarm tracing analysis is realized, ultimately forming result alarms with relational expression capabilities.

[0046] S3. Dynamically visualize the results of the source tracing analysis in step S2;

[0047] Based on the end-to-end fault diagnosis link of network services, the end-to-end link nodes are dynamically drawn, and the location of the fault-affected nodes in the system's network topology and the relevant performance of specific device ports are graphically displayed. Figure 5 A visual topology diagram is presented.

[0048] Figure 6 This paper demonstrates a scenario for tracing faults in fixed base stations. In this scenario, the primary and backup transmission network cables of the base station BBU main control board are disconnected through an operational exercise. Using this method for intelligent diagnostic analysis, the following service network management alarms are obtained:

[0049]

[0050] When using the method of this invention, if a fault alarm occurs in the system, for example, a communication interruption alarm for fixed station equipment caused by a fault in the LTE link of the base station BBU, the system processing flow is as follows: The system uses this method to diagnose and analyze to obtain analysis results, which include the fault-affected equipment and the end-to-end communication link status; triggers a fault emergency work order; the system assesses the impact of the fault and guides the fault handling process through a process-oriented approach; on-site personnel quickly restore the base station BBU connection transmission network cable, the fault is eliminated, and the system is restored.

[0051] Although the present invention has been described in detail through the preferred embodiments above, it should be understood that the above description should not be considered as a limitation of the present invention. Various modifications and substitutions to the present invention will be apparent to those skilled in the art after reading the above description. Therefore, the scope of protection of the present invention should be defined by the appended claims.

Claims

1. A communication complex network service end-to-end intelligent diagnosis analysis method, characterized in that, Includes the following steps: S1. Establish the analytical processing foundation for end-to-end source tracing analysis of complex network services, including: establishing a complex network knowledge graph model; using dynamic routing analysis based on graph computing to monitor service links in real time; and establishing a multi-source data integration and operation engine. S2. When a link fault is detected through dynamic routing analysis based on graph computing, the multi-source data integration and operation engine, combined with the complex network knowledge graph model, is used to perform end-to-end source tracing analysis of network services, checking each node through which the service data passes to accurately locate the node where performance failure occurs. S3. Based on the end-to-end diagnostic links of network services, dynamically draw end-to-end link nodes, graphically displaying the location of the faulty node in the system's network topology and the relevant performance of specific device ports. The dynamic routing analysis using graph computation for real-time monitoring of service links involves: Based on graph computation, dynamic routing is achieved by accessing routing tables in the MIB database using SNMP primitives to extract topology-related information. A breadth-first search algorithm is used to continuously and dynamically expand the topology discovery scope. The service type queried based on OID information is used to locate the scope of service links affected by faulty nodes, thus achieving real-time monitoring of service links. A vectorized model between nodes is constructed using a graph database: entity resource nodes are divided into devices, interfaces, links, and services. Based on a pre-set list of interconnected working paths and protection paths for each site, and using the site name and connection relationships in the site attributes, the final working path and protection path are obtained.

2. The method of claim 1, wherein the method further comprises: The establishment of the complex network knowledge graph model specifically involves: building a knowledge graph database based on the network knowledge graph, storing multiple sets of network knowledge graphs in the knowledge graph database; and after completing the storage of the knowledge graph database, performing knowledge extraction, knowledge fusion, and knowledge computation to construct the complex network knowledge graph.

3. The end-to-end intelligent diagnostic analysis method for complex communication network services as described in claim 2, characterized in that, The business network, technical design, backbone network, product defects, network topology and business ports, construction and implementation, business terminal equipment, human operation, power supply factors, environmental factors, and various factors in construction and implementation are incorporated as fault influencing factors into the constructed complex network knowledge graph.

4. The method of claim 1, wherein the method further comprises: determining a plurality of network performance metrics for each of the plurality of network paths; and determining a plurality of network performance metrics for each of the plurality of network paths. The establishment of a multi-source data integration and operation engine includes: standard normalization of multi-source data acquisition and diagnostic analysis driven by multiple engines.

5. The method of claim 4, wherein the method further comprises: The standardization of multi-source data acquisition is as follows: unify the research and sorting out of data integration specifications, and formulate a complete set of data integration technical methods to solve the problem of data uniformity in heterogeneous systems.

6. The method of claim 5, wherein the method further comprises: The diagnostic analysis of the multi-engine associated drive is as follows: combining waveform analysis, inference analysis, and fault tree analysis of multiple engine associated drives, and adopting corresponding diagnostic analysis techniques for different communication network services to achieve the fusion of multiple engines for complex communication network services.

7. The end-to-end intelligent diagnostic analysis method for complex communication network services as described in claim 6, characterized in that, The end-to-end source tracing analysis is as follows: Based on the established complex network knowledge graph, a knowledge base for fault source tracing methods based on key influencing factors is established. Logical fault source tracing rules are converted into identifiable information rule tables. Combined with an alarm storage buffer pool, alarm source tracing analysis is realized, and finally, result alarms with relational expression capabilities are formed.

8. The method of claim 7, wherein the method further comprises: When a communication alarm occurs, the process is as follows: the diagnostic analysis method is used to obtain the analysis results, which include the equipment affected by the fault and the end-to-end communication link status; a fault emergency work order is triggered; the system assesses the impact of the fault and guides the fault handling process through a process-oriented approach; On-site personnel quickly restored transmission, the fault was eliminated, and the system was restored.

Images