Core network root cause alarm positioning method, device, equipment, storage medium and product
Through Bayesian network and first-order predicate logic, the fault model is constructed, and the alarm information is automatically analyzed, which solves the problem of inefficient fault positioning in the existing technology, and realizes intelligent fault positioning of the cloud-based vIMS core network, improving maintenance efficiency and positioning accuracy.
Patent Information
- Application Number
- CN202411089264.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-09
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2044-08-09
AI Technical Summary
In the prior art, core network fault analysis relies on manual experience, resulting in low fault positioning efficiency, uncontrollable processing time, affecting user service experience.
Using Bayesian network and first-order predicate logic, network element-level and network-level fault models are constructed, alarm information is automatically analyzed, root cause alarm is determined, and intelligent fault location of cloud-based vIMS core network is realized.
It improves the maintenance efficiency of the core network, quickly finds the root cause alarms among massive alarms, improves the accuracy and efficiency of fault location, and reduces the dependence of manual analysis.
Smart Images

Figure CN119052067A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of core network technology, and in particular to a core network root cause alarm locating method, device, equipment, storage medium and product. Background Art
[0002] With the development of communication technology and digital transformation, the cloud-based virtual Internet Protocol Multimedia Subsystem (vIMS) core network has become an important part of voice services. In the cloud-based vIMS core network, the connection relationship between various virtual network function devices (VNF) is relatively complex, which brings great challenges to fault analysis and location.
[0003] Currently, the core network alarm mainly collects alarm information of various network elements through the network management platform, and analyzes a large number of alarms one by one. In the process of troubleshooting and locating, maintenance personnel need to rely on experience to analyze and judge a large amount of alarm data, and finally locate the root cause alarm. However, since fault analysis relies entirely on manual experience, there are problems such as difficulty in fault location, extended processing time, and uncontrollable service recovery time, which makes the efficiency of troubleshooting the root cause of the fault low, which in turn affects the user service experience. Summary of the invention
[0004] The present invention provides a core network root cause alarm positioning method, device, equipment, storage medium and product, which are used to solve the defect of low efficiency in locating the root cause of fault investigation in the prior art, realize automatic sorting of the causal relationship between alarms under the complex network structure of vIMS, dig out the root cause of alarms, and realize intelligent demarcation and positioning of cloud-based vIMS faults, thereby significantly improving the maintenance efficiency of the core network.
[0005] The present invention provides a core network root cause alarm positioning method, comprising: Get current alarm information; Analyze the current alarm information according to the pre-built network element fault Bayesian network and network element first-order predicate logic to determine the network element level root cause alarm location result; Analyze the current alarm information according to the pre-built network fault Bayesian network and network first-order predicate logic to determine the network-level root cause alarm location result; Based on the network element level root cause alarm locating result and the network level root cause alarm locating result, a final root cause alarm locating result is output.
[0006] According to a core network root cause alarm locating method provided by the present invention, the network element fault Bayesian network is constructed in the following manner: Constructing a core network device topology relationship; the core network device topology relationship includes network elements, domain relationships and network element connection relationships; The historical alarm information is grouped based on the network element as a granularity to obtain historical alarm information of multiple network elements; For each network element, a scalable sliding time window algorithm is used to divide the historical alarm information of the network element into multiple items, and the earliest alarms of different types in each item are retained to obtain an item set for each network element; According to the item set of the same type of network elements, a first two-dimensional matrix is constructed for describing the relationship between the number of occurrences of alarms under the same type of network elements; Based on the first two-dimensional matrix, the relationship between alarms is evaluated, and a network element fault Bayesian network for describing the relationship between alarms under the same type of network elements is constructed.
[0007] According to a core network root cause alarm locating method provided by the present invention, the scalable sliding time window algorithm is used to divide the historical alarm information of the network element into multiple items, including: Arrange the alarms generated by the network element in chronological order to obtain a timeline of alarm occurrence; For each main window, according to the alarms on the alarm occurrence time axis and the placement position of the main window, the alarms within the main window delay range are determined and placed in the target item; Based on the placement position of the main window, a corresponding extended window is placed, and alarms within the delay range of the extended window are determined and placed in the target item; After each alarm on the alarm occurrence time axis is placed into a corresponding item, a plurality of items are obtained.
[0008] According to a core network root cause alarm locating method provided by the present invention, the first two-dimensional matrix for describing the relationship between the number of occurrences of alarms under the same type of network elements is constructed according to the item set of the same type of network elements, including: Traversing the item set of the same type of network elements, counting the number of times the first alarm type occurs before the second alarm type, and the number of times the first alarm type occurs and the second alarm type does not occur; wherein the first alarm type is any alarm type, and the second alarm type is one of other alarm types except the any alarm type; Based on the number of times the first alarm type occurs before the second alarm type and the number of times the first alarm type occurs but the second alarm type does not occur, a first two-dimensional matrix is constructed to describe the relationship between the number of occurrences of alarms in network elements of the same type.
[0009] According to a core network root cause alarm locating method provided by the present invention, the relationship between alarms is evaluated based on the first two-dimensional matrix, and a network element fault Bayesian network for describing the relationship between alarms under the same type of network elements is constructed, including: Calculate, based on the number of times the first alarm type occurs before the second alarm type and the total number of occurrences between alarms, a first evaluation value of whether the first alarm type is a factor causing the second alarm type; Calculate, based on the number of times the second alarm type occurs before the first alarm type and the total number of occurrences between alarms, a second evaluation value of the second alarm type as a triggering factor of the first alarm type; Calculate the correlation evaluation value of the first alarm type and the second alarm type according to the number of times the first alarm type occurs before the second alarm type, the number of times the second alarm type occurs before the first alarm type, and the total number of occurrences between the alarms; Determine, according to the first evaluation value and the correlation evaluation value, a relationship between the first alarm type and the second alarm type; the relationship includes a parent-child relationship and an association relationship; Based on the relationship between the first alarm type and the second alarm type and the constructed core network device topology relationship, a network element fault Bayesian network for describing the relationship between alarms under the same type of network elements is constructed.
[0010] According to a core network root cause alarm locating method provided by the present invention, the current alarm information is analyzed according to the pre-built network element fault Bayesian network and network element first-order predicate logic to determine the network element level root cause alarm locating result, including: Taking network element as granularity, obtain the first initial root cause alarm set within the preset range of the fault time point; Traversing all alarm types in the first initial root cause alarm set, when determining according to the network element failure Bayesian network that there is an alarm triggering the current traversal alarm in the first initial root cause alarm set, removing the current traversal alarm from the first initial root cause alarm set to obtain a first final root cause alarm set; The first final root cause alarm set is grouped to obtain multiple groups of root cause alarms; wherein the root cause alarms in each group are associated with each other, and the root cause alarms between the groups are not associated with each other; For the root cause alarm in each group, the network element first-order predicate logic is used to analyze and determine the network element level root cause alarm location result.
[0011] According to a core network root cause alarm locating method provided by the present invention, the network fault Bayesian network is constructed in the following manner: The historical alarm information of multiple network elements associated with the same domain network elements, the same type of network elements and the same protocol is divided into the same group, and the historical alarm information of multiple groups of associated network elements is obtained; For each group of associated network elements, a scalable sliding time window algorithm is used to divide the historical alarm information of the network elements into multiple items, and the earliest alarms of different types in each item are retained to obtain an item set of each group of associated network elements; According to the item set of each group of associated network elements, construct a second two-dimensional matrix for describing the alarm occurrence relationship under the same group of associated network elements; Based on the second two-dimensional matrix, the parent-child relationship between alarms is evaluated, and a network fault Bayesian network for describing the relationship between alarms in the same group of associated network elements is constructed.
[0012] According to a core network root cause alarm locating method provided by the present invention, the current alarm information is analyzed based on a pre-constructed network fault Bayesian network and a network first-order predicate logic to determine a network-level root cause alarm locating result, including: Using domain relationships and connection relationships as granularity, obtain the second initial root cause alarm set within the preset range of the fault time point; Traversing all alarm types in the second initial root cause alarm set, and when determining according to the network fault Bayesian network that there is an alarm triggering the currently traversed alarm in the second initial root cause alarm set, removing the currently traversed alarm from the second initial root cause alarm set to obtain a second final root cause alarm set; The second final root cause alarm set is grouped to obtain multiple groups of root cause alarms; wherein the root cause alarms in each group are associated with each other, and the root cause alarms between the groups are not associated with each other; For the root cause alarm in each group, the network first-order predicate logic is used to analyze and determine the network-level root cause alarm location result.
[0013] The present invention also provides a core network root cause alarm positioning device, comprising: Information acquisition module, used to obtain current alarm information; A first positioning module is used to analyze the current alarm information according to a pre-built network element fault Bayesian network and network element first-order predicate logic to determine a network element level root cause alarm positioning result; A second positioning module is used to analyze the current alarm information according to a pre-built network fault Bayesian network and a network first-order predicate logic to determine a network-level root cause alarm positioning result; The output module is used to output a final root cause alarm locating result based on the network element level root cause alarm locating result and the network level root cause alarm locating result.
[0014] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it implements any of the core network root cause alarm locating methods described above.
[0015] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the core network root cause alarm locating method as described in any one of the above-mentioned methods is implemented.
[0016] The present invention also provides a computer program product, including a computer program, which, when executed by a processor, implements any of the core network root cause alarm locating methods described above.
[0017] The core network root cause alarm locating method, device, equipment, storage medium and product provided by the present invention can automatically sort out the causal relationship between alarms under the complex network structure of vIMS by troubleshooting the root cause alarm results at the network element level and troubleshooting the root cause alarm results at the network level based on the correlation alarms, without relying on manual analysis of alarm data, and can quickly find the root cause alarm from massive alarm information, thereby improving the maintenance efficiency of the cloud-based vIMS core network; by constructing Bayesian network models at the network element level and the network level, using Bayesian networks and first-order predicate logic, the network element level and network level fault root cause positioning of the cloud-based vIMS network is realized, thereby filtering the influence of invalid alarms and non-root cause alarms on troubleshooting, and improving the accuracy of troubleshooting root cause positioning. BRIEF DESCRIPTION OF THE DRAWINGS
[0018] 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.
[0019] Figure 1 It is a flow chart of a core network root cause alarm locating method provided by an embodiment of the present invention.
[0020] Figure 2 It is a schematic diagram of the construction process of the network element failure Bayesian network provided by an embodiment of the present invention.
[0021] Figure 3 It is a schematic diagram of sub-items provided by an embodiment of the present invention.
[0022] Figure 4 It is a schematic diagram of a first two-dimensional matrix provided by an embodiment of the present invention.
[0023] Figure 5It is an example schematic diagram of implementing root cause location based on first-order predicate logic provided by an embodiment of the present invention.
[0024] Figure 6 It is a functional structure diagram of a core network root cause alarm locating method provided by an embodiment of the present invention.
[0025] Figure 7 It is a structural diagram of a core network root cause alarm locating device provided by an embodiment of the present invention.
[0026] Figure 8 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0027] 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.
[0028] Related root cause alarm location technologies mainly perform manual analysis on alarms generated by a single network element, which has the following disadvantages: 1. Dependence on personnel skill level: Operation and maintenance personnel need to collect a large number of network element alarms from the network management system, and analyze the alarm data one by one based on personal experience, find anomalies and determine possible root causes.
[0029] 2. Low efficiency in root cause alarm investigation: The mobile communication network has complex layers and architectures. When a single network element fails, the network element itself and surrounding related network elements will generate a large amount of related alarm data. In the massive amount of alarm data, important information is often submerged. If each alarm is investigated one by one, there is a problem of low efficiency in locating the root cause of the fault.
[0030] In view of the above problems, the embodiments of the present invention realize automatic sorting of causal relationships between alarms under the complex network structure of vIMS by troubleshooting the root cause alarm results at the network element level and troubleshooting the root cause alarm results at the network level based on the correlation alarms. This eliminates the need to rely on manual analysis of alarm data, and enables rapid finding of root cause alarms from massive alarm information, thereby improving the maintenance efficiency of the cloud-based vIMS core network. By constructing Bayesian network models at the network element level and the network level, and using Bayesian networks and first-order predicate logic, the root cause location of network element-level and network-level faults in the cloud-based vIMS network is realized, thereby filtering out the impact of invalid alarms and non-root cause alarms on troubleshooting, and improving the accuracy of troubleshooting root cause location.
[0031] Figure 11 is a flow chart of a core network root cause alarm location method provided by an embodiment of the present invention. Figure 1 The embodiment of the present invention provides a core network root cause alarm location method, and the method may specifically include the following steps: Step 110, obtaining current alarm information.
[0032] It should be noted that the execution subject of the core network root cause alarm positioning method provided in the embodiment of the present invention may be an electronic device, a component in an electronic device, an integrated circuit or a chip. The electronic device may be a mobile electronic device or a non-mobile electronic device. Exemplarily, the mobile electronic device may be a mobile phone, a tablet computer, a laptop computer, a PDA, a wearable device, an ultra-mobile personal computer (UMPC), a netbook or a personal digital assistant (PDA), etc., and the non-mobile electronic device may be a server, a network attached storage (NAS), a personal computer (PC), a television (TV), a teller machine or a self-service machine, etc., which is not specifically limited in the embodiment of the present invention.
[0033] Specifically, when a fault occurs, the current alarm information can be collected through the network management platform, and the root cause location analysis of the current alarm information can be automatically performed to output the root cause alarm location result.
[0034] Step 120: Analyze the current alarm information according to the pre-built network element fault Bayesian network and network element first-order predicate logic to determine the network element level root cause alarm location result.
[0035] Among them, the network element fault Bayesian network can be used to describe the relationship between alarms in the same type of network elements. The network element first-order predicate logic can be used to define the reasoning logic for inferring the root cause of the fault based on the relationship between alarms in the same type of network elements.
[0036] In the embodiment of the present invention, there may be multiple alarms under the same type of network element. By using the relationship between the alarms under the same type of network element to perform root cause analysis, the root cause alarm location result can be determined from the network element level. Specifically, the current alarm information can be analyzed according to the pre-built network element level fault Bayesian network and the pre-defined network element level first-order predicate logic, so as to determine the root cause alarm location result from the network element level.
[0037] The embodiment of the present invention can realize the root cause location of network element-level faults in the cloudified vIMS network by utilizing the network element fault Bayesian network and the network element first-order predicate logic, thereby filtering the influence of invalid alarms and non-root cause alarms on troubleshooting and improving the accuracy of troubleshooting and locating the root cause of faults.
[0038] Step 130: Analyze the current alarm information according to the pre-constructed network fault Bayesian network and network first-order predicate logic to determine the network-level root cause alarm location result.
[0039] The network fault Bayesian network can be used to describe the relationship between alarms in the same group of associated network elements, where the associated network elements may refer to network elements in the same domain or with protocol associations. The network first-order predicate logic can be used to define the reasoning logic for inferring the root cause of the fault based on the relationship between alarms in the same group of associated network elements.
[0040] In the embodiment of the present invention, there may be multiple alarms under the same group of associated network elements. By using the relationship between the alarms under the same group of associated network elements to perform root cause analysis, the root cause alarm location result can be determined from the network level. Specifically, the current alarm information can be analyzed based on the pre-constructed network-level fault Bayesian network and the pre-defined network-level first-order predicate logic, so as to determine the root cause alarm location result from the network level.
[0041] The embodiments of the present invention can realize network-level fault root cause location of the cloudified vIMS network by utilizing the network fault Bayesian network and the network first-order predicate logic, thereby filtering the influence of invalid alarms and non-root cause alarms on troubleshooting and improving the accuracy of troubleshooting and locating the root cause of the fault.
[0042] Step 140: output a final root cause alarm locating result based on the network element level root cause alarm locating result and the network level root cause alarm locating result.
[0043] In an embodiment of the present invention, after obtaining the network element level root cause alarm locating results and the network level root cause alarm locating results, the final root cause alarm locating results can be output and fed back to the user, thereby eliminating the need to manually analyze a large number of alarms one by one, thereby improving the efficiency of locating the root cause of the fault.
[0044] The embodiment of the present invention realizes automatic sorting of causal relationships between alarms under the complex network structure of vIMS by troubleshooting the root cause alarm results at the network element level and troubleshooting the root cause alarm results at the network level based on the correlation alarms, without relying on manual analysis of alarm data, and can quickly find the root cause alarm from massive alarm information, thereby improving the maintenance efficiency of the cloud-based vIMS core network; by constructing Bayesian network models at the network element level and the network level, using Bayesian networks and first-order predicate logic, the network element level and network level fault root cause positioning of the cloud-based vIMS network is realized, thereby filtering the influence of invalid alarms and non-root cause alarms on troubleshooting, and improving the accuracy of troubleshooting root cause positioning.
[0045] Figure 2 FIG. 1 is a schematic diagram of the construction process of the network element failure Bayesian network provided by an embodiment of the present invention. Figure 2 Based on any of the above embodiments, the network element fault Bayesian network can be constructed by the following steps: Step 210, constructing a core network device topology relationship; the core network device topology relationship includes network elements, domain relationships and network element connection relationships.
[0046] In an embodiment of the present invention, a cloud-based vIMS core network device topology relationship can be constructed for subsequent alarm association. The cloud-based vIMS core network device may include user data management (UDM), call session control function (CSCF), application server (TAS), data router and authorization server (DRA), session border controller (SBC), etc.
[0047] The topological relationship of cloud-based vIMS core network equipment may include network elements, domain relationships, and network element connection relationships. Specifically, network elements are two-tuple information (network element A, network element type), and alarms can be grouped based on network elements as the basic granularity; domain relationships are two-tuple information (network element A, domain name). In the cloud-based core network networking architecture, network functions can be shared by multiple similar devices in a functional domain. The embodiments of the present invention can merge and mine alarms based on functional domains or device pools; connection relationships are three-tuple information (network element A, network element B, connection protocol). Business processes involve the collaboration of multiple network elements. Many faults need to consider the spread of alarms between multiple types of network elements. The embodiments of the present invention can merge and mine alarms based on the connection status of the network.
[0048] Step 220: group the historical alarm information based on the network element as the granularity to obtain the historical alarm information of multiple network elements.
[0049] In an embodiment of the present invention, the full amount of historical alarm data within a preset time period of the clouded vIMS core network can be collected from the network management system and data structure preprocessing can be performed. Specifically, the data structure processing can be completed by parsing the historical alarm data, extracting information such as the data center, network element name, network element type, alarm number, alarm occurrence time, alarm title and detailed description. Among them, the preset time period can be set according to actual needs, for example, the preset time period can be 6 months.
[0050] After completing the data structured preprocessing, the network element failure Bayesian network and the network failure Bayesian network can be constructed according to the impact range. Specifically, when constructing the network element failure Bayesian network, the historical alarms can be grouped based on the network element as the granularity, and each group corresponds to the full amount of historical alarm structured information of the specified network element, that is, the structured information of the historical alarms of multiple network elements is obtained.
[0051] Step 230: for each network element, a scalable sliding time window algorithm is used to divide the historical alarm information of the network element into multiple items, and the earliest alarm of different types in each item is retained to obtain an item set for each network element.
[0052] Specifically, for each group of historical alarm structured information corresponding to a specified network element, a scalable sliding time window algorithm can be used to divide the alarm information into multiple items (cases) to obtain an initial item set for each network element, and alarm information deduplication is performed on each item, retaining the earliest alarms of different types to obtain an item set for each network element.
[0053] Step 240: construct a first two-dimensional matrix for describing the relationship between the occurrence frequencies of alarms in network elements of the same type according to the item set of network elements of the same type.
[0054] Specifically, the item sets of all network elements of the same type can be divided into a group according to the network element type to obtain the item sets of network elements of the same type. Each type of network element can correspond to a first two-dimensional matrix, and according to the item sets of network elements of the same type, a first two-dimensional matrix for describing the relationship between the number of occurrences of alarms under network elements of the same type is constructed.
[0055] Step 250: evaluate the relationship between alarms based on the first two-dimensional matrix, and construct a network element fault Bayesian network for describing the relationship between alarms in network elements of the same type.
[0056] Specifically, the relationship between alarms can be evaluated based on the number of occurrences of alarms under the same type of network elements, and whether the relationship between alarms is a parent-child relationship or an association relationship can be evaluated, so as to supplement the relationship between alarms on the topological relationship of the core network equipment and obtain a network element fault Bayesian network for describing the relationship between alarms under the same type of network elements. Among them, a network element fault Bayesian network model of parent-child relationship / association relationship is maintained for each network element type.
[0057] By constructing a Bayesian network model at the network element level, the embodiment of the present invention can realize the network element-level fault root cause location of the clouded vIMS network through the network element fault Bayesian network and the network element first-order predicate logic, thereby filtering the impact of invalid alarms and non-root cause alarms on troubleshooting and improving the accuracy of troubleshooting root cause location.
[0058] Based on any of the above embodiments, the use of a scalable sliding time window algorithm to divide the historical alarm information of the network element into multiple items may specifically include: Step 231, arranging the alarms generated by the network element in chronological order to obtain a time axis of alarm generation; Step 232, for each main window, according to the alarms on the alarm occurrence time axis and the placement position of the main window, determine the alarms within the delay range of the main window and put them into the target item; Step 233, based on the placement position of the main window, place the corresponding extended window, determine the alarms within the delay range of the extended window and place them in the target item; Step 234, after putting each alarm on the alarm occurrence time axis into a corresponding item, a plurality of items are obtained.
[0059] The sliding time window may include a main window delay and an extended window delay. In an embodiment of the present invention, the alarms occurring in the same network element may be arranged in chronological order to obtain an alarm occurrence timeline corresponding to the network element; the sliding time window may be placed on the timeline, and according to the placement of the alarms on the alarm occurrence timeline and the main window, the alarms within the main window delay range may be placed in the target item corresponding to the main window; based on the placement of the main window, the corresponding extended window may be placed, and the alarms within the extended window delay range may be placed in the target item corresponding to the main window. After all the alarms on the alarm occurrence timeline are placed in the corresponding items, multiple items may be obtained.
[0060] Figure 3 Schematic diagram of the sub-items provided by the embodiment of the present invention. Figure 3 , the sliding time window may include a main window delay ΔT and an extended window delay Δt.
[0061] For example, a1~a8 can be alarms that occur in the same network element, and can be arranged in chronological order. Alarms a1~a8 correspond to time points T0~T7 respectively, and the alarm occurrence time axis of the same network element is obtained. The sliding time window can be placed on the time axis. Based on the placement position of the main window ΔT, it can be known that T2 < T0+ΔT (within the ΔT range), T3 > T0+ΔT (outside the ΔT range), and alarms a1, a2, and a3 can all be placed in item Case 1.
[0062] It is possible to use T2 as the starting point again to determine whether there is a new alarm within the extended window delay range. The occurrence time of alarm a4 is within the extended window delay range Δt, and alarm a4 can be placed in the corresponding item Case 1.
[0063] Taking T3 as a new starting point, it is possible to repeatedly determine whether there are new alarms within the extended window delay range until no new alarms appear within the extended window delay range, thereby constructing Case 1.
[0064] Similarly, we can eventually construct Case 1 (corresponding to alarms a1 to a4), Case 2 (corresponding to alarms a5 and a6), and Case 3 (corresponding to alarms a7 and a8).
[0065] The embodiment of the present invention can classify fault-related alarms into the same item through the main window delay and the corresponding extended window delay, so that the alarms in each item are associated with each other, and the alarms between items are not associated with each other, thereby realizing the aggregation of fault-related alarms and the isolation between different fault alarms.
[0066] The embodiment of the present invention uses a scalable sliding time window algorithm to divide the historical alarm information of the network element into multiple items during the data preprocessing process, realizes alarm sub-items based on the scalable sliding window, and takes into account both the aggregation of fault-related alarms and the isolation between different fault alarms, which is beneficial to reducing the algorithm complexity of the subsequent construction of the network element level fault and network level fault Bayesian network.
[0067] Based on any of the above embodiments, constructing a first two-dimensional matrix for describing the relationship between the number of occurrences of alarms under the same type of network elements according to the item set of the same type of network elements may specifically include: Step 241, traverse the item set of the same type of network elements, and count the number of times the first alarm type occurs before the second alarm type, and the number of times the first alarm type occurs and the second alarm type does not occur; wherein the first alarm type is any alarm type, and the second alarm type is one of other alarm types except the any alarm type; Step 242: construct a first two-dimensional matrix for describing the relationship between the occurrence times of alarms in the same type of network elements based on the number of times the first alarm type occurs before the second alarm type and the number of times the first alarm type occurs but the second alarm type does not occur.
[0068] In the embodiment of the present invention, all item sets of all network elements of the same type can be traversed, the occurrence frequency relationship between two alarms can be counted, and a first two-dimensional matrix N×N matrix can be constructed to describe the pairwise relationship between the alarms. Each type of network element can correspond to a first two-dimensional matrix, and N is the number of all alarm types under the same type of network element.
[0069] Specifically, any alarm type can be used as the first alarm type, and one alarm type other than this alarm type can be used as the second alarm type. The number of times the first alarm type occurs before the second alarm type, and the number of times the first alarm type occurs and the second alarm type does not occur are counted, thereby obtaining the number of times each alarm type occurs before any other alarm type, and the number of times each alarm type occurs and any other alarm type does not occur.
[0070] A first two-dimensional matrix describing the relationship between the occurrence times of alarms of the same type in network elements can be constructed with the number of times the first alarm type occurs before the second alarm type as row elements and the number of times the first alarm type occurs but the second alarm type does not occur as column elements.
[0071] Figure 4 is a schematic diagram of a first two-dimensional matrix provided by an embodiment of the present invention. Figure 4 As an example, 3431 / 21 indicates the number of times that alarm type 1 occurs before alarm type 2. , and the number of times that alarm type 1 occurs and alarm type 2 does not occur ; 0 / 7 indicates the number of times that alarm type 2 occurs before alarm type 1 , and the number of times that alarm type 2 occurs and alarm type 1 does not occur .
[0072] The embodiment of the present invention constructs a two-dimensional matrix and then constructs a network element fault Bayesian network for describing the relationship between alarms under the same type of network elements. The network element fault Bayesian network and the network element first-order predicate logic can be used to realize the network element-level fault root cause location of the clouded vIMS network, thereby filtering the influence of invalid alarms and non-root cause alarms on troubleshooting and improving the accuracy of troubleshooting and locating the root cause of faults.
[0073] Based on any of the foregoing embodiments, the evaluating the relationship between alarms based on the first two-dimensional matrix and constructing a network element fault Bayesian network for describing the relationship between alarms under the same type of network elements may specifically include: Step 251, calculating a first evaluation value of the first alarm type as a factor causing the second alarm type according to the number of times the first alarm type occurs before the second alarm type and the total number of occurrences between alarms; Step 252, calculating a second evaluation value of the second alarm type as a triggering factor of the first alarm type according to the number of times the second alarm type occurs before the first alarm type and the total number of occurrences between alarms; Step 253, calculating a correlation evaluation value between the first alarm type and the second alarm type according to the number of times the first alarm type occurs before the second alarm type, the number of times the second alarm type occurs before the first alarm type, and the total number of occurrences between the alarms; Step 254: determining the relationship between the first alarm type and the second alarm type according to the first evaluation value and the correlation evaluation value; the relationship includes a parent-child relationship and an association relationship; Step 255: Based on the relationship between the first alarm type and the second alarm type and the constructed core network device topology relationship, a network element fault Bayesian network for describing the relationship between alarms in the same type of network elements is constructed.
[0074] Among them, the total number of occurrences between alarms is the sum of the number of times the first alarm type occurs before the second alarm type, the number of times the first alarm type occurs and the second alarm type does not occur, the number of times the second alarm type occurs before the first alarm type, and the number of times the second alarm type occurs and the first alarm type does not occur.
[0075] The condition for determining that the alarms are in a parent-child relationship may be that the reference evaluation value is greater than the first evaluation threshold M, and the condition for determining that the alarms are in an association relationship may be that the correlation evaluation value is greater than the second evaluation threshold K. The specific values of the first evaluation threshold and the second evaluation threshold can be set according to actual needs.
[0076] The following example uses the relationship between alarm type 1 and alarm type 2 as an example. The total number of occurrences between alarm type 1 and alarm type 2 is the number of times that alarm type 1 occurs before alarm type 2. , The number of times that alarm type 1 occurs and alarm type 2 does not occur , The number of times that alarm type 2 occurs before alarm type 1 The number of times that alarm type 2 occurs and alarm type 1 does not occur sum.
[0077] Continue to refer to Figure 4 , we can calculate the ratio of the number of times that alarm type 1 occurs before alarm type 2 and the total number of times alarm type 1 occurs to alarm type 2, and get the first evaluation value of alarm type 1 as the triggering factor of alarm type 2: .
[0078] The ratio of the number of times that alarm type 2 occurs before alarm type 1 and the total number of times alarm type 1 occurs to alarm type 2 can be calculated to obtain the second evaluation value of alarm type 2 as the triggering factor of alarm type 1: .
[0079] The correlation evaluation value between alarm type 1 and alarm type 2 can be obtained by calculating the sum of the number of times alarm type 1 occurs before alarm type 2 and the number of times alarm type 2 occurs before alarm type 1, and the ratio of the total number of occurrences between alarm type 1 and alarm type 2: .
[0080] like , then alarm type 1 is the triggering alarm of alarm type 2 (i.e., the parent alarm); if , , then alarm type 1 is an associated alarm with alarm type 2; otherwise, alarm type 1 and alarm type 2 are unrelated alarms. Similarly, it can be calculated that alarm type N is an alarm triggered by alarm type 1 and an alarm triggered by alarm type 2.
[0081] In the embodiment of the present invention, based on the relationship between the first alarm type and the second alarm type and the construction of the core network device topology relationship, a network element fault Bayesian network for describing the relationship between alarms in the same type of network elements can be constructed.
[0082] The embodiment of the present invention constructs a network element fault Bayesian network for describing the relationship between alarms under the same type of network elements. The network element fault Bayesian network and the network element first-order predicate logic can be used to realize the network element-level fault root cause location of the clouded vIMS network, thereby filtering the impact of invalid alarms and non-root cause alarms on troubleshooting and improving the accuracy of troubleshooting root cause location.
[0083] Based on any of the above embodiments, analyzing the current alarm information according to the pre-constructed network element fault Bayesian network and network element first-order predicate logic to determine the network element level root cause alarm locating result may specifically include: Step 121, taking the network element as the granularity, obtaining a first initial root cause alarm set within a preset range of the fault time point.
[0084] Specifically, based on the scalable sliding time window algorithm, we can first extract all alarm data before and after the network failure time point, parse the alarm, extract information such as data center, network element name, network element type, alarm number, alarm occurrence time, alarm title and detailed description to complete data structuring processing.
[0085] After completing the data structure processing, we can obtain all the alarm sets A={y before and after the fault time point based on the network element granularity. i}. Let the root cause alarm set be C, which can be initialized to C=A.
[0086] The preset range of the fault time point can be set according to actual needs, and can be 10 minutes before and after the fault time point, or 20 minutes, and the present invention does not make specific limitations.
[0087] Step 122, traverse all alarm types in the first initial root cause alarm set, and when it is determined according to the network element failure Bayesian network that there is an alarm triggering the current traversal alarm in the first initial root cause alarm set, remove the current traversal alarm from the first initial root cause alarm set to obtain the first final root cause alarm set.
[0088] Specifically, all alarm types y in the first initial root cause alarm set can be traversed i , when the first initial root cause alarm set is determined to exist according to the network element failure Bayesian network , that is, the alarm triggered by the current traversal alarm y is y`, then y can be removed from the first initial root cause alarm set.
[0089] The final alarm set retains several alarm types without parent-child relationship, that is, there is a correlation relationship or no correlation between any two alarm types in the root cause alarm set.
[0090] Step 123, grouping the first final root cause alarm set to obtain multiple groups of root cause alarms; wherein the root cause alarms within each group are associated with each other, and the root cause alarms between groups are not associated with each other.
[0091] In the embodiment of the present invention, the first final root cause alarm set can be grouped to obtain C1, C2, ..., Ck, that is, k groups of root cause alarms. Among them, the root cause alarms in each group have an associated relationship, and the root cause alarms between groups do not have an associated relationship, and the root cause alarm is one of C1~Ck.
[0092] Step 124: Analyze the root cause alarm in each group using the network element first-order predicate logic to determine the network element level root cause alarm location result.
[0093] In the embodiment of the present invention, for each group Ck, the network element first-order predicate logic can be used to analyze and confirm the network element level root cause alarm location result. The network element first-order predicate logic can include basic logic rules and single network element fault reasoning rules.
[0094] The basic logic rule definition can be: (1) proto(x,y): The alarm number y of network element x indicates that there is a protocol circuit fault; (2) sys(x,y): NE x alarm number y indicates a system failure; (3) hrd(x,y): NE x alarm number y indicates a hardware fault; (4) parent(y1,y2): Alarm y1 is caused by alarm y2. y2 and y1 are parent-child relationships. Whether this is true or not is calculated by the Bayesian network. (5) ancestry(y1,y2): Alarm y1 is directly or indirectly caused by alarm y2, and y2 is the father of y1. Whether this is true or not is calculated by the Bayesian network. (6) relation(y1,y2): Alarm y1 is related to alarm y2, and whether it is true is calculated by the Bayesian network; (7) fault(x,y1,y2): Network element x has fault y1, y1 is caused by y2, and the root cause alarm is y2.
[0095] The single network element fault inference rule definition can be shown in Table 1: Table 1
[0096] When it is determined that y1, y2, and y3 in the case are parent alarms and are in an associated relationship, the final root cause alarm can be determined in the order of priority hardware failure > system failure > protocol failure.
[0097] The embodiment of the present invention can realize the root cause location of network element-level faults in the cloudified vIMS network through the network element fault Bayesian network and the network element first-order predicate logic, thereby filtering the influence of invalid alarms and non-root cause alarms on troubleshooting and improving the accuracy of troubleshooting and locating the root cause of faults.
[0098] Based on any of the above embodiments, the network fault Bayesian network is constructed in the following manner: historical alarm information of multiple network elements associated with the same domain network elements, the same type of network elements and the same protocol is divided into the same group to obtain historical alarm information of multiple groups of associated network elements; for each group of associated network elements, a scalable sliding time window algorithm is used to divide the historical alarm information of the network elements into multiple items, and the earliest alarms of different types in each item are retained to obtain an item set of each group of associated network elements; based on the item set of each group of associated network elements, a second two-dimensional matrix is constructed to describe the relationship between alarms in the same group of associated network elements; based on the second two-dimensional matrix, the parent-child relationship between alarms is evaluated to construct a network fault Bayesian network to describe the relationship between alarms in the same group of associated network elements.
[0099] In an embodiment of the present invention, the process of constructing a Bayesian network for network failures is similar to the process of constructing a Bayesian network for network element failures. The difference is that the alarm grouping is based on the network element granularity, and the alarms of the same type of network elements in the same domain and the alarms of multiple network elements associated with the same protocol are merged based on the cloud network topology relationship. On this basis, the alarms are itemized using the timing relationship and the extended time window algorithm to realize the construction of the Bayesian network for network failures, which will not be repeated in the embodiment of the present invention.
[0100] By constructing a Bayesian network model at the network level, the embodiment of the present invention can realize network-level root cause location of faults in the cloudified vIMS network through the network fault Bayesian network and the network first-order predicate logic, thereby filtering the influence of invalid alarms and non-root cause alarms on troubleshooting and improving the accuracy of troubleshooting and locating the root cause of faults.
[0101] Based on any of the above embodiments, the current alarm information is analyzed according to the pre-constructed network fault Bayesian network and network first-order predicate logic to determine the network-level root cause alarm positioning result, including: obtaining a second initial root cause alarm set within a preset range of the fault time point with domain relationship and connection relationship as granularity; traversing all alarm types in the second initial root cause alarm set, and when it is determined according to the network fault Bayesian network that there is an alarm triggering the current traversal alarm in the second initial root cause alarm set, the current traversal alarm is removed from the second initial root cause alarm set to obtain a second final root cause alarm set; the second final root cause alarm set is grouped to obtain multiple groups of root cause alarms; wherein the root cause alarms in each group are associated with each other, and the root cause alarms between groups are not associated with each other; for the root cause alarms in each group, the network first-order predicate logic is used to analyze and determine the network-level root cause alarm positioning result.
[0102] In the embodiment of the present invention, the multiple root cause alarms obtained by grouping the second final root cause alarm set can be analyzed using network first-order predicate logic to confirm the network-level root cause alarm location result. The network first-order predicate logic can include basic logic rules and network fault reasoning rules.
[0103] The network fault inference rule definition can be shown in Table 2: Table 2
[0104] When a single network element failure cannot be confirmed, the alarm may be caused by other related network elements.
[0105] In an embodiment of the present invention, a set of all alarms before and after the fault time point is obtained as A`={yi} with domain relationship / connection relationship as the granularity, and the process of determining the network-level root cause alarm locating result is similar to the process of determining the network element-level root cause alarm locating result. The difference is that the models used are network Bayesian network and network first-order predicate logic, which will not be repeated in the present invention.
[0106] Figure 5 FIG. 1 is an example schematic diagram of implementing root cause location based on first-order predicate logic provided by an embodiment of the present invention. Figure 5 As an example, after determining the NE-level root cause alarm locating results and the network-level root cause alarm locating results, the final root cause alarm locating results can be displayed and output, including CSCF (diameter alarm, module alarm, hardware alarm), DRA (diameter alarm, Stream Control Transmission Protocol (SCTP) alarm), SBC (node failure alarm), etc.
[0107] When performing single-network element fault location analysis, the following can be found: CSCF (hardware alarm causes diameter alarm), DRA (external cause causes diameter alarm), and SBC (external cause causes node fault alarm).
[0108] When performing network fault location analysis, the following can be analyzed: demarcating the root cause network element (SBC and DRA problems are caused by CSCF), locating the root cause alarm (CSCF hardware alarm is the root cause), and providing processing suggestions (prioritize CSCF hardware alarm).
[0109] The embodiment of the present invention can realize network-level fault root cause location of the cloudified vIMS network through the network fault Bayesian network and the network first-order predicate logic, thereby filtering the influence of invalid alarms and non-root cause alarms on troubleshooting and improving the accuracy of troubleshooting and locating the root cause of the fault.
[0110] Figure 6 1 is a functional structural diagram of a core network root cause alarm location method provided by an embodiment of the present invention. Figure 6 For different device types and alarm numbers in the cloudified vIMS core network, the alarm preprocessing filtering module can be used to count the number of occurrences of various alarms on the device within a historical period, divide the number of each alarm by the total number of alarms of the device type in the same time period, obtain the alarm ratio of each alarm number, and select the alarm type with a ratio lower than the preset threshold as the filtering rule. In addition, the alarm type that needs to be paid attention to can also be manually entered into the filtering rule to realize the screening of the alarms collected in the existing network through the above filtering rules.
[0111] The alarm parsing module can be used to collect the full alarm data of the cloudified vIMS core network from the network management platform regularly (for example, every 10 minutes) and convert the full alarm data into structured alarm data. The data can include fields such as network element name, network element type, alarm number, alarm title, alarm information, occurrence time, and end time.
[0112] The alarm grouping and itemization module can be used to divide alarms into multiple groups of different item sets based on the topological relationship of cloud-based vIMS core network equipment at the network element and network levels, and use a scalable sliding time window to itemize the alarm sequences in the same group to obtain an alarm set for each item for subsequent data statistics and Bayesian network construction.
[0113] The alarm database module can be used to use a relational database for data storage.
[0114] The single-network element root cause alarm reasoning module can be used for the alarm set of each item, based on the Bayesian network, to obtain the ancestor alarm or ancestor alarm set of each item. These alarms or alarm sets can be used as candidate root cause alarms for subsequent root cause location.
[0115] The first-order predicate logic rules at the network element level can determine the primary and secondary relationships between single network element alarms (alarms of hardware type, system type, protocol type, etc.). The single network element root cause delimitation and positioning module can be used to further infer candidate root cause alarms based on the first-order predicate logic of the network element to confirm the single network element fault root cause alarm.
[0116] The network-level Bayesian network stores the parent-child relationship and association relationship between the alarm types of network elements in the same domain or with protocol association. The network root cause alarm reasoning module can be used to obtain the fault root network element and the candidate root cause alarm set for the root cause alarm of each network element in the network based on the network-level fault Bayesian network.
[0117] The network-level first-order predicate logic rules determine the primary and secondary relationships of alarms between different network elements in the network (alarms of hardware type, system type, protocol type, etc.). The network root cause demarcation and positioning module can be used to confirm the root source site and root cause of network-level faults based on the network-level first-order predicate logic logic for candidate root cause alarms through progressive reasoning.
[0118] The delimited positioning summary output can be used to determine the data integration and storage of the fault root cause network element, the fault root cause alarm, and the fault-induced alarm list, so as to present the alarm list to the front desk and enable other applications to perform analysis based on the alarm list.
[0119] The embodiments of the present invention can not only find out the root cause alarm results at the network element level, but also find out the root cause alarm results at the network level based on the correlation alarms, thereby realizing automatic positioning of the root cause alarms without relying on manual analysis of alarm data, and realizing rapid locating of the root cause alarms from massive alarm information, thereby improving the maintenance efficiency of the cloud-based vIMS core network.
[0120] The embodiment of the present invention constructs an alarm relationship Bayesian network based on three dimensions of network element, domain, and associated protocol. Through the Bayesian network at two levels of network element and network and the first-order predicate logic, the root cause location of the fault of the cloudified vIMS network can be realized, thereby filtering the influence of invalid alarms and non-root cause alarms on troubleshooting.
[0121] The embodiment of the present invention realizes alarm itemization based on an extensible sliding window method during data preprocessing, and takes into account both the aggregation of fault-related alarms and the isolation between different fault alarms, which is conducive to reducing the algorithm complexity of the subsequent construction of the network element and network relationship Bayesian network.
[0122] The core network root cause alarm locating device provided by the present invention is described below. The core network root cause alarm locating device described below and the core network root cause alarm locating method described above can refer to each other.
[0123] Figure 7 1 is a schematic diagram of the structure of a core network root cause alarm positioning device provided by an embodiment of the present invention. Figure 7 The embodiment of the present invention provides a core network root cause alarm positioning device, which may specifically include the following modules: Information acquisition module 710, used to obtain current alarm information; The first positioning module 720 is used to analyze the current alarm information according to the pre-built network element fault Bayesian network and network element first-order predicate logic to determine the network element level root cause alarm positioning result; The second positioning module 730 is used to analyze the current alarm information according to the pre-built network fault Bayesian network and network first-order predicate logic to determine the network-level root cause alarm positioning result; The output module 740 is used to output a final root cause alarm locating result based on the network element level root cause alarm locating result and the network level root cause alarm locating result.
[0124] The embodiment of the present invention realizes automatic sorting of causal relationships between alarms under the complex network structure of vIMS by troubleshooting the root cause alarm results at the network element level and troubleshooting the root cause alarm results at the network level based on the correlation alarms, without relying on manual analysis of alarm data, and can quickly find the root cause alarm from massive alarm information, thereby improving the maintenance efficiency of the cloud-based vIMS core network; by constructing Bayesian network models at the network element level and the network level, using Bayesian networks and first-order predicate logic, the network element level and network level fault root cause positioning of the cloud-based vIMS network is realized, thereby filtering the influence of invalid alarms and non-root cause alarms on troubleshooting, and improving the accuracy of troubleshooting root cause positioning.
[0125] Figure 8 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 8 As shown, the electronic device may include: a processor 810, a communication interface 820, a memory 830 and a communication bus 840, wherein the processor 810, the communication interface 820 and the memory 830 complete mutual communication through the communication bus 840. The processor 810 may call the logic instructions in the memory 830 to execute the core network root cause alarm locating method, the method comprising: obtaining current alarm information; analyzing the current alarm information according to the pre-constructed network element fault Bayesian network and the network element first-order predicate logic to determine the network element level root cause alarm locating result; analyzing the current alarm information according to the pre-constructed network fault Bayesian network and the network first-order predicate logic to determine the network level root cause alarm locating result; outputting the final root cause alarm locating result based on the network element level root cause alarm locating result and the network level root cause alarm locating result.
[0126] In addition, the logic instructions in the above-mentioned memory 830 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.
[0127] On the other hand, the present invention also provides a computer program product, which includes a computer program, and the computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the core network root cause alarm locating method provided by the above methods, and the method includes: obtaining current alarm information; analyzing the current alarm information according to a pre-constructed network element fault Bayesian network and network element first-order predicate logic, and determining a network element-level root cause alarm locating result; analyzing the current alarm information according to a pre-constructed network fault Bayesian network and network first-order predicate logic, and determining a network-level root cause alarm locating result; based on the network element-level root cause alarm locating result and the network-level root cause alarm locating result, outputting a final root cause alarm locating result.
[0128] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the core network root cause alarm locating method provided by the above-mentioned methods, the method comprising: obtaining current alarm information; analyzing the current alarm information according to a pre-constructed network element fault Bayesian network and network element first-order predicate logic, and determining a network element-level root cause alarm locating result; analyzing the current alarm information according to a pre-constructed network fault Bayesian network and network first-order predicate logic, and determining a network-level root cause alarm locating result; and outputting a final root cause alarm locating result based on the network element-level root cause alarm locating result and the network-level root cause alarm locating result.
[0129] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.
[0130] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.
[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. A core network root cause alarm location method, characterized in that: include: Get current alarm information; Analyze the current alarm information according to the pre-built network element fault Bayesian network and network element first-order predicate logic to determine the network element level root cause alarm location result; Analyze the current alarm information according to the pre-built network fault Bayesian network and network first-order predicate logic to determine the network-level root cause alarm location result; Based on the network element level root cause alarm locating result and the network level root cause alarm locating result, a final root cause alarm locating result is output.
2. The core network root cause alarm location method according to claim 1, characterized in that: The network element fault Bayesian network is constructed in the following way: Constructing a core network device topology relationship; the core network device topology relationship includes network elements, domain relationships and network element connection relationships; The historical alarm information is grouped based on the network element as a granularity to obtain historical alarm information of multiple network elements; For each network element, a scalable sliding time window algorithm is used to divide the historical alarm information of the network element into multiple items, and the earliest alarms of different types in each item are retained to obtain an item set for each network element; According to the item set of the same type of network elements, a first two-dimensional matrix is constructed for describing the relationship between the number of occurrences of alarms under the same type of network elements; Based on the first two-dimensional matrix, the relationship between alarms is evaluated, and a network element fault Bayesian network for describing the relationship between alarms under the same type of network elements is constructed.
3. The core network root cause alarm location method according to claim 2, characterized in that: The scalable sliding time window algorithm is used to divide the historical alarm information of the network element into multiple items, including: Arrange the alarms generated by the network element in chronological order to obtain a timeline of alarm occurrence; For each main window, according to the alarms on the alarm occurrence time axis and the placement position of the main window, the alarms within the main window delay range are determined and placed in the target item; Based on the placement position of the main window, a corresponding extended window is placed, and alarms within the delay range of the extended window are determined and placed in the target item; After each alarm on the alarm occurrence time axis is placed into a corresponding item, a plurality of items are obtained.
4. The core network root cause alarm location method according to claim 2, characterized in that: The constructing, based on the item set of the same type of network elements, a first two-dimensional matrix for describing the relationship between the number of occurrences of alarms under the same type of network elements comprises: Traversing the item set of the same type of network elements, counting the number of times the first alarm type occurs before the second alarm type, and the number of times the first alarm type occurs and the second alarm type does not occur; wherein the first alarm type is any alarm type, and the second alarm type is one of other alarm types except the any alarm type; Based on the number of times the first alarm type occurs before the second alarm type and the number of times the first alarm type occurs but the second alarm type does not occur, a first two-dimensional matrix is constructed to describe the relationship between the number of occurrences of alarms in network elements of the same type.
5. The core network root cause alarm location method according to claim 4, characterized in that: The evaluating the relationship between alarms based on the first two-dimensional matrix and constructing a network element fault Bayesian network for describing the relationship between alarms under the same type of network elements includes: Calculate, based on the number of times the first alarm type occurs before the second alarm type and the total number of occurrences between alarms, a first evaluation value of whether the first alarm type is a factor causing the second alarm type; Calculate, based on the number of times the second alarm type occurs before the first alarm type and the total number of occurrences between alarms, a second evaluation value of the second alarm type as a triggering factor of the first alarm type; Calculate the correlation evaluation value of the first alarm type and the second alarm type according to the number of times the first alarm type occurs before the second alarm type, the number of times the second alarm type occurs before the first alarm type, and the total number of occurrences between the alarms; Determine, according to the first evaluation value and the correlation evaluation value, a relationship between the first alarm type and the second alarm type; the relationship includes a parent-child relationship and an association relationship; Based on the relationship between the first alarm type and the second alarm type and the constructed core network device topology relationship, a network element fault Bayesian network for describing the relationship between alarms under the same type of network elements is constructed.
6. The core network root cause alarm location method according to claim 1, characterized in that: The analyzing the current alarm information according to the pre-built network element fault Bayesian network and network element first-order predicate logic to determine the network element level root cause alarm location result includes: Taking network element as granularity, obtain the first initial root cause alarm set within the preset range of the fault time point; Traversing all alarm types in the first initial root cause alarm set, when determining according to the network element failure Bayesian network that there is an alarm triggering the current traversal alarm in the first initial root cause alarm set, removing the current traversal alarm from the first initial root cause alarm set to obtain a first final root cause alarm set; The first final root cause alarm set is grouped to obtain multiple groups of root cause alarms; wherein the root cause alarms in each group are associated with each other, and the root cause alarms between the groups are not associated with each other; For the root cause alarm in each group, the network element first-order predicate logic is used to analyze and determine the network element level root cause alarm location result.
7. The core network root cause alarm location method according to claim 2, characterized in that: The network fault Bayesian network is constructed in the following way: The historical alarm information of multiple network elements associated with the same domain network elements, the same type of network elements and the same protocol is divided into the same group, and the historical alarm information of multiple groups of associated network elements is obtained; For each group of associated network elements, a scalable sliding time window algorithm is used to divide the historical alarm information of the network elements into multiple items, and the earliest alarms of different types in each item are retained to obtain an item set of each group of associated network elements; According to the item set of each group of associated network elements, construct a second two-dimensional matrix for describing the alarm occurrence relationship under the same group of associated network elements; Based on the second two-dimensional matrix, the parent-child relationship between alarms is evaluated, and a network fault Bayesian network for describing the relationship between alarms in the same group of associated network elements is constructed.
8. The core network root cause alarm location method according to claim 1, characterized in that: The analyzing the current alarm information according to the pre-built network fault Bayesian network and the network first-order predicate logic to determine the network-level root cause alarm location result includes: Using domain relationships and connection relationships as granularity, obtain the second initial root cause alarm set within the preset range of the fault time point; Traversing all alarm types in the second initial root cause alarm set, and when determining according to the network fault Bayesian network that there is an alarm triggering the currently traversed alarm in the second initial root cause alarm set, removing the currently traversed alarm from the second initial root cause alarm set to obtain a second final root cause alarm set; The second final root cause alarm set is grouped to obtain multiple groups of root cause alarms; wherein the root cause alarms in each group are associated with each other, and the root cause alarms between the groups are not associated with each other; For the root cause alarm in each group, the network first-order predicate logic is used to analyze and determine the network-level root cause alarm location result.
9. A core network root cause alarm location device, characterized in that: include: Information acquisition module, used to obtain current alarm information; A first positioning module is used to analyze the current alarm information according to a pre-built network element fault Bayesian network and network element first-order predicate logic to determine a network element level root cause alarm positioning result; A second positioning module is used to analyze the current alarm information according to a pre-built network fault Bayesian network and a network first-order predicate logic to determine a network-level root cause alarm positioning result; The output module is used to output a final root cause alarm locating result based on the network element level root cause alarm locating result and the network level root cause alarm locating result.
10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, it implements the core network root cause alarm locating method as described in any one of claims 1 to 8.
11. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, it implements the core network root cause alarm locating method as described in any one of claims 1 to 8.
12. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, it implements the core network root cause alarm locating method as described in any one of claims 1 to 8.
Citation Information
Patent Citations
Power grid fault intelligent diagnosis method and device, computer equipment and storage medium
CN110646710A
Method for positioning 5G network virtualization cross-layer problem through AI algorithm
CN113542039A
Data correlation analysis method and system
CN115037592A
Method and device for generating alarm root cause positioning model
CN115883324A
Alarm quality evaluation device, alarm quality evaluation method, and alarm quality evaluation program
JP2019028619A
Cited By
Power distribution communication network fault positioning method, device, equipment and medium
CN120811877A
Constellation core network test fault positioning method and device
CN122661102A
