Network anomaly repair method and device and storage medium

Abstract

The invention records a network anomaly repair method and device and a storage medium, and the method comprises the steps: firstly carrying out the matching of a fault feature with a fault feature set in a preset fault rule library, and carrying out the continuous traversal of a repair strategy library under the condition that the fault feature matching fails, thereby achieving the repair of a network anomaly. And carrying out network anomaly repair on the network by utilizing the traversed repair strategy, thereby solving the network anomaly which is not covered by the preset fault rule base in time.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of network diagnosis, in particular to a network exception repair method, device and storage medium. BACKGROUND

[0002] With the rapid development of the Internet of Things, the Internet of Things has been widely applied in various fields. For example, the Internet of Vehicles formed by the deep integration of the Internet of Things and intelligent transportation technology has become a core component of modern intelligent networked vehicles. The stability, real-time performance and reliability of network connection are directly related to user experience, and even affect driving safety in the case of application in the Internet of Vehicles. In the actual running environment, the network environment is complex and changeable, and network connection exceptions such as sudden bandwidth reduction, delay jitter and frequent disconnection are prone to occur due to reasons such as signal shielding, base station switching, network congestion or hardware failure. If these exceptions cannot be automatically repaired in time and effectively, it will cause the interruption or failure of critical data transmission, such as real-time road conditions, remote control instructions and high-precision map updates in the Internet of Vehicles, which seriously restricts the reliability and development of the Internet of Vehicles.

[0003] The common network exception processing scheme in the field of the Internet of Vehicles mostly adopts a repair strategy response mechanism based on pre-defined rules. The core of this scheme is that the Internet of Vehicles terminal continuously monitors network performance, and then compares the monitoring value with the threshold value set in the database in advance. Once a certain indicator exceeds the threshold range, it is determined that a specific type of network exception has occurred. Finally, the network repair action is executed according to the unique repair strategy pre-configured for each type of exception. However, in this scheme, the correspondence between the rules in the pre-set fault rule library and the repair strategies completely depends on manual experience setting, and it is difficult to cover the complex and changeable network scenarios in the real world, resulting in the inability to timely solve network exceptions in some cases where network exceptions are not set in the pre-defined repair strategy. SUMMARY

[0004] The main purpose of the present application is to provide a network exception repair method, device and storage medium, which aims to solve the problem that the pre-set fault rule library in the prior art is difficult to cover the complex and changeable network scenarios in the real world, resulting in the inability to timely solve some network exceptions.

[0005] To achieve the above purpose, the present application provides a network exception repair method, which comprises: In the case of network exception, the fault characteristics of the network are obtained; The fault characteristics are matched with the fault characteristics set in the pre-set fault rule library; In the case of failure of the fault characteristic matching, the repair strategy library is traversed, and the repair strategy library includes at least one repair strategy for network exception. repair the network according to the current traversed repair strategy.

[0006] Optionally, before the traversing of the repair strategy library in the case of the failure of the matching of the fault feature, the method further comprises: obtaining a set of repair strategies for solving the network anomaly; obtaining a historical repair success rate of each repair strategy in the set of repair strategies; sorting each repair strategy in the set of repair strategies according to the historical repair success rate to obtain the repair strategy library.

[0007] Optionally, the traversing of the repair strategy library in the case of the failure of the matching of the fault feature, the repair strategy library storing repair strategies for network anomaly comprises: obtaining multi-dimensional key performance data presented at the moment when the network state is the network anomaly; constructing an anomaly feature vector according to the multi-dimensional key performance data; calculating the similarity between the anomaly feature vector and a rule feature vector corresponding to each repair strategy in the repair strategy library; traversing the repair strategies in the repair strategy library in descending order of the similarity.

[0008] Optionally, after the network is repaired according to the current traversed repair strategy, the method further comprises: matching the network anomaly with a repair strategy for solving the network anomaly to obtain a fault rule of the network anomaly; updating the fault rule to a preset fault rule library.

[0009] Optionally, the number of times that the repair strategy solves the network anomaly is recorded; in the case that the number of times reaches a preset number of times, updating the fault rule to a preset fault rule library; in the case that the number of times does not reach the preset number of times, improving the traversal priority of the repair strategy in the repair strategy library.

[0010] Optionally, after the network is repaired according to the current traversed repair strategy, the method further comprises: detecting the network state after the network anomaly repair; if the network state after the network anomaly repair is not the network anomaly, determining that the network anomaly repair is completed.

[0011] Optionally, after the detection of the network state after the network anomaly repair, the method further comprises If the network state after the network exception repair is still the network exception, the step of traversing the repair strategy library is re-executed until the network state after the network exception repair is not the network exception.

[0012] Optionally, after the matching of the fault feature with the fault features set in the preset fault rule library, the method further comprises: In the case that the network exception matching is successful, the network exception is repaired by using the repair strategy corresponding to the matched fault feature.

[0013] In addition, to achieve the above-mentioned purpose, the application further provides a network exception repair device, which comprises: a network state monitoring module, configured to acquire a fault feature of a network in the case that the network state is a network exception; a fault matching module, configured to match the fault feature with fault features set in a preset fault rule library; a fault traversal module, configured to traverse a repair strategy library in the case that the fault feature matching fails, wherein the repair strategy library comprises at least one repair strategy for network exception; a fault repair module, configured to repair the network exception of the network according to the currently traversed repair strategy.

[0014] In addition, to achieve the above-mentioned purpose, the application further provides a network exception repair device, which comprises a memory, a processor, and a network exception repair program stored in the memory and executable on the processor, wherein the network exception repair program is configured to implement the steps of the network exception repair method as described above.

[0015] In addition, to achieve the above-mentioned purpose, the application further provides a storage medium, wherein the storage medium stores a network exception repair program, and the network exception repair program implements the steps of the network exception repair method as described above when executed by a processor.

[0016] The application discloses a network exception repair method, device, equipment and storage medium. The method comprises the following steps: acquiring a fault feature of a network in the case that the network state is a network exception; matching the fault feature with fault features set in a preset fault rule library; traversing a repair strategy library in the case that the fault feature matching fails; and repairing the network exception of the network according to the currently traversed repair strategy. The application firstly matches the fault feature with the fault features set in the preset fault rule library, and in the case that the fault feature matching fails, the repair strategy library is continuously traversed, and the network exception of the network is repaired by using the traversed repair strategy, so that the network exception that is not covered by the preset fault rule library can be solved in time. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the network anomaly repair device structure in the hardware operating environment involved in the embodiments of the present invention; Figure 2 This is a flowchart illustrating the first embodiment of the network anomaly repair method of the present invention; Figure 3 This is a reference diagram for setting up the preset fault rule base in the network anomaly repair method of the present invention; Figure 4 This is a schematic diagram of the first process of the second embodiment of the network anomaly repair method of the present invention; Figure 5 This is a schematic diagram of the second process of the second embodiment of the network anomaly repair method of the present invention; Figure 6 This is a flowchart illustrating the third embodiment of the network anomaly repair method of the present invention; Figure 7 This is a flowchart illustrating the fourth embodiment of the network anomaly repair method of the present invention; Figure 8 This is a structural block diagram of the first embodiment of the network anomaly repair device of the present invention.

[0018] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0019] It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the invention.

[0020] Reference Figure 1 , Figure 1 This is a schematic diagram of the network anomaly repair device structure in the hardware operating environment involved in the embodiments of the present invention.

[0021] like Figure 1As shown, the network anomaly repair device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to establish communication between these components. The user interface 1003 may include a display screen or an input unit such as a keyboard; optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wireless-Fidelity (Wi-Fi) interface). The memory 1005 may be high-speed random access memory (RAM) or stable non-volatile memory (NVM), such as a disk drive. Optionally, the memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0022] Those skilled in the art will understand that Figure 1 The structure shown does not constitute a limitation on the network anomaly repair device and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0023] like Figure 1 As shown, the memory 1005, which serves as a storage medium, may include an operating system, a network communication module, a user interface module, and a network fault repair program.

[0024] exist Figure 1 In the network anomaly repair device shown, the network interface 1004 is mainly used for data communication with the network server; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the network anomaly repair device of the present invention can be set in the network anomaly repair device, and the network anomaly repair device calls the network anomaly repair program stored in the memory 1005 through the processor 1001 and executes the network anomaly repair method provided in the embodiment of the present invention.

[0025] This invention provides a method for repairing network anomalies, referring to... Figure 2 , Figure 2 This is a flowchart illustrating the first embodiment of the network anomaly repair method of the present invention.

[0026] In this embodiment, the network anomaly repair method includes the following steps: Step S10: When the network status is abnormal, obtain the fault characteristics of the network; It should be noted that the executing entity of the method in this embodiment can be a computing service device with abnormal state repair, network communication, and program execution functions, such as a vehicle controller, mobile phone, tablet computer, personal computer, etc., or other electronic devices that achieve the same or similar functions. The following description uses the aforementioned network abnormality repair device to illustrate this embodiment and the subsequent embodiments.

[0027] It should be understood that existing vehicle-to-everything (V2X) terminals cannot effectively handle unknown or complex network anomaly scenarios that are not predefined in their preset fault rule base. When an unknown anomaly is detected in the network, and no complete match can be found in the preset fault rule base, the system will be unable to resolve the network fault, thus affecting the user experience.

[0028] Understandably, real-time network status monitoring is necessary before attempting to repair network anomalies to confirm their occurrence. Conversely, if the network is functioning normally, network anomaly repair is unnecessary. During monitoring, key network performance data can be monitored, such as signal strength, signal-to-noise ratio (SNR), network throughput, and packet loss rate. Anomalies in any of these metrics (e.g., low signal strength, high SNR, or excessively long ping times) indicate a network anomaly. Conversely, if all key performance data are normal, the network is considered normal. Network fault characteristics, specifically those indicative of an anomaly, are reflected in key performance data, such as signal strength below -100dBm, ping failures, or latency exceeding 1 second.

[0029] In practical implementation, during network status detection, the application layer can be used to assess current network characteristics such as ping latency and packet loss rate, successful network registration, and signal strength below -110dBm to confirm whether a network failure has occurred. When the network status is abnormal, fault characteristics such as specific signal strength values ​​and ping latency values ​​can be collected.

[0030] Step S20: Match the fault features with the fault features set in the preset fault rule base.

[0031] It should be understood that the preset fault rule base is a fault rule base set by those skilled in the art based on experience. Each fault feature in this preset fault rule base has a corresponding repair strategy, which can be used to repair network anomalies corresponding to the fault features. (Refer to...) Figure 3The preset fault rule base can be set in the form of: status-policy-rule. The status is the output result of the network status detection module, which has different network parameter indicators and combinations. The status can be the state presented by the fault characteristics. The policy is the predetermined network repair strategy, which can be combined and traversed. The rule is the correspondence between status and policy.

[0032] It should be noted that since the preset fault rule base contains network anomaly repair strategies, the fault characteristics can be matched with the fault characteristics in the preset fault rule base to determine the network anomaly repair strategy that can resolve the fault characteristic.

[0033] In practice, fault characteristics in network anomalies can be matched with fault characteristics in a preset fault rule base. If a match is successful, the corresponding repair strategy can be selected to fix the network anomaly. For example, if the current fault characteristic is low signal strength, and the fault characteristic database contains this fault characteristic, then the repair strategy of switching network interfaces corresponding to low signal strength can be directly selected to fix the network anomaly. Of course, if a match fails, the subsequent steps of traversing the repair strategy database are executed.

[0034] Step S30: If the fault feature matching fails, the repair strategy library is traversed, which contains repair strategies for network anomalies.

[0035] It should be understood that, given the limited number of fault features already included in the preset fault rule base, there will still be fault features whose features cannot be matched successfully. For these network faults, no corresponding repair strategy can be obtained, and the network anomaly cannot be repaired.

[0036] It should be noted that the repair strategy library is a pre-set library containing a large number of repair strategies. If the network anomaly is not included in the preset fault rule base and cannot be repaired, a repair strategy from the repair strategy library can be selected to fix the network anomaly. The repair strategy library includes multiple repair strategies, all designed to repair network anomalies; it simply contains multiple repair strategies. Repair strategies can include single strategies, such as switching network standards, adjusting TCP parameters, reconnecting to a specific frequency band, etc., as well as composite repair strategies, which include combinations of different repair strategies and their sequential arrangements. For example, one composite repair strategy might be A+B+C, and another might be A+C+B. The repair strategies in the repair strategy library are a set of pre-set low-level operations based on a deep understanding of network protocols, exhibiting orthogonality and complementarity. For example, repair strategies might include: restarting TCP connections, disabling and re-enabling wireless radio frequency, switching network access, and forcibly registering to a specific frequency band.

[0037] In practice, if a fault characteristic fails to match a fault characteristic in the preset fault rule base, a repair strategy is selected from the repair strategy library and then used to repair the network anomaly. This traversal involves selecting each repair strategy in the library one by one. Since a single repair strategy selected from the library may not resolve the network anomaly, this traversal method is used to repair the network anomaly one by one using all available repair strategies.

[0038] Step S40: Perform network anomaly repair on the network according to the currently traversed repair strategy.

[0039] It should be understood that after traversing the repair strategy library, you can first use the repair strategy to attempt to repair network anomalies. For example, if you encounter a network restart option, you can control the currently used network to restart.

[0040] This embodiment describes a network anomaly repair method. The method involves: acquiring network fault characteristics when the network is in an anomaly state; matching the fault characteristics with fault characteristics set in a preset fault rule base; if a fault characteristic match fails, traversing a repair strategy library; and repairing the network anomaly according to the currently traversed repair strategy. This embodiment first matches fault characteristics with fault characteristics set in the preset fault rule base. If a fault characteristic match fails, it continuously traverses the repair strategy library and uses the traversed repair strategies to repair the network anomaly, thereby promptly resolving network anomalies not covered by the preset fault rule base.

[0041] refer to Figure 4 , Figure 4 This is a schematic diagram of the first process of the second embodiment of the network anomaly repair method of the present invention.

[0042] Based on the first embodiment described above, in this embodiment, the steps preceding step S30 include: Step S31: Obtain a set of repair strategies to resolve network anomalies.

[0043] It should be understood that before iterating through repair strategies using the repair strategy library, this library needs to be built. This library contains repair strategies for network anomalies. To ensure that network anomalies can be effectively resolved, a set of repair strategies can be obtained first. This set of repair strategies includes all current repair strategies for network anomalies, including both single repair strategies and composite repair strategies resulting from combinations of multiple strategies. In practice, known repair strategies can be collected first, and then this collected set of repair strategies can be used.

[0044] Step S32: Obtain the historical repair success rate of each repair strategy in the repair strategy set.

[0045] Understandably, the repair strategy library includes all known repair strategies. However, considering the real-time and uncertain nature of network dynamics, each network repair strategy may yield different results in resolving network anomalies. This can lead to a longer timeframe for resolving network anomalies when randomly iterating through the library. Therefore, when setting up the repair strategy library, it is also necessary to consider the different resolving capabilities of various repair strategies and prioritize them accordingly.

[0046] It should be noted that the historical repair success rate is the percentage of times a repair strategy has successfully restored network status during its history of addressing network anomalies. When a repair strategy is used multiple times to address a network anomaly, some repair processes may fail to resolve the anomaly, while others may succeed. The ratio of the number of times the repair strategy successfully resolved the anomaly to the total number of times it was used is the historical repair success rate of that strategy.

[0047] In practice, the number of repair attempts for each repair strategy and the number of successful repairs of network anomalies corresponding to that strategy can be obtained over a period of time or within a certain area. The repair success rate of the strategy can then be calculated based on the number of repair attempts and the number of successful repairs. For example, when using device reboot for network repair, the process may involve network recovery or the network remaining abnormal. In the case of multiple repair attempts, the repair success rate corresponding to the device reboot repair strategy can be determined based on the number of repair attempts and the number of successful repairs.

[0048] Step S33: Sort the repair strategies in the repair strategy set according to the historical repair success rate to obtain the repair strategy library.

[0049] It should be understood that, given the historical success rate of each repair strategy within the repair strategy set, the repair strategies can be sorted according to their historical success rate. Repair strategies with higher historical success rates are more likely to resolve network anomalies, and in this case, the repair strategy can be placed at the beginning of the repair strategy library so that it can be accessed more frequently when traversing the repair strategy library.

[0050] In practice, the historical repair success rates of each repair strategy in the repair strategy set can be compared, the historical repair success rates can be sorted, and then the corresponding repair strategies can be prioritized according to the results of the historical repair success rate sorting to obtain the repair strategy library.

[0051] This embodiment sorts the repair strategies in the repair strategy set according to their historical repair success rates. When traversing the repair strategy library, it can prioritize repair strategies with higher historical repair success rates to repair network anomalies, thereby effectively reducing the time required to repair network anomalies and quickly resolving network anomalies.

[0052] refer to Figure 5 , Figure 5 This is a schematic diagram of the second process of the second embodiment of the network anomaly repair method of the present invention.

[0053] Based on the second embodiment described above, in this embodiment, step S30 includes: Step S301: Obtain multi-dimensional key performance data presented at the moment when the network status is abnormal.

[0054] It should be understood that during the traversal of the repair strategy library, repair strategies are mainly selected by going through the library one by one. However, this traversal method also suffers from the problem of slow network anomaly repair. Therefore, in this embodiment, it is also possible to first obtain the parameters of the current network anomaly, and then use these parameters to select repair strategies in the repair strategy library in a sequential manner, which can improve the efficiency of network anomaly repair.

[0055] It should be noted that the multi-dimensional key performance data refers to the performance data presented by the network at the moment of network anomaly. This multi-dimensional key performance data includes performance data of multiple different standards, such as network registration and ping latency in the network connectivity dimension, and signal strength and packet loss rate in the network transmission dimension.

[0056] In practice, when a network anomaly is confirmed, multi-dimensional key performance data can be directly collected at the current moment when the network is in an abnormal state, thereby obtaining this multi-dimensional data. Alternatively, data such as ping latency, packet loss rate, signal strength, and signal-to-noise ratio during network signal transmission can be collected at the moment of network anomaly to obtain multi-dimensional key performance data.

[0057] Step S302: Construct an anomaly feature vector based on the multi-dimensional key performance data.

[0058] Understandably, after determining the multi-dimensional key performance data under network anomaly conditions, it is necessary to apply the multi-dimensional key performance data to the traversal process. Considering that there is a lot of multi-dimensional key performance data, it is possible to preprocess the multi-dimensional key performance data so that the repair strategy can be quickly traversed using the multi-dimensional key performance data in the subsequent traversal of the repair strategy library.

[0059] It's important to note that the anomaly feature vector is a feature vector composed of multi-dimensional key performance data in the event of network anomalies. This anomaly feature vector includes preprocessed multi-dimensional key performance data. In practice, the weight of each key performance data point can be determined based on its impact on the network anomaly. Then, the anomaly feature vector corresponding to each key performance data point and its corresponding weight is determined. For example, if the network anomaly is a low signal strength, then the corresponding signal strength data and packet loss rate will have a larger weight, while the weight of network connectivity-related key performance data will be smaller.

[0060] Step S303: Calculate the similarity between the abnormal feature vector and the rule feature vectors corresponding to each repair strategy in the repair strategy library.

[0061] Understandably, the repair strategy library includes various repair strategies, and different strategies can handle one or more network anomalies. For example, the strategy of switching network interfaces can usually resolve network registration failures, high ping latency, and other network anomalies. Therefore, different repair strategies in different repair strategy libraries will also have corresponding rule feature vectors. These rule feature vectors usually have a certain degree of similarity to the anomaly feature vectors corresponding to multi-dimensional key performance data. For example, if the network anomaly is a high signal-to-noise ratio, the weight of interference in the anomaly signal feature will be large. In contrast, the network interference level in the rule feature vector corresponding to the repair strategy of disabling and re-enabling radio frequency in the repair strategy library is high. In this case, there is a high similarity between this anomaly feature vector and the rule feature vector.

[0062] Once each repair strategy in the repair strategy library is determined, the corresponding rule feature vector for each repair strategy can be determined. Different repair strategies have different rule feature vectors. For example, the repair strategy that reduces the signal-to-noise ratio has a larger weight in the rule feature vector of the former than the repair strategy that increases the signal strength.

[0063] In practice, the rule feature vector corresponding to each repair strategy in the repair strategy library can be compared with the abnormal feature vector constructed from the multi-dimensional key performance data at the current network anomaly moment, thereby determining the similarity between the rule feature vector and the abnormal feature vector of each repair strategy.

[0064] Step S304: Traverse the repair strategies in the repair strategy library in descending order of similarity.

[0065] It should be understood that traversing the repair strategy library one by one is inefficient. However, when the similarity between the regular feature vector and the abnormal feature vector of each repair strategy is determined, the repair strategies with higher similarity can be prioritized during the traversal process, based on the similarity between the abnormal feature vector and the regular feature vector. These are the repair strategies that are more likely to solve network anomalies.

[0066] In practice, the repair strategies in the repair strategy library can be traversed in descending order of similarity. The repair strategy with the highest similarity is traversed first to repair the network anomaly, and then the repair strategies with gradually decreasing similarity are traversed to repair the network anomaly, until the repair strategy resolves the network anomaly.

[0067] In this embodiment, by acquiring multi-dimensional key performance data at the moment of network anomaly, and then constructing an anomaly feature vector, the anomaly feature vector can be matched with the rule feature vectors of each repair strategy in the repair strategy library. This allows for a rapid traversal of the repair strategy with the highest probability of resolving network anomalies, thereby effectively improving the efficiency of resolving network anomalies.

[0068] A third embodiment of the network anomaly repair method of this application is proposed based on the first or second embodiment described above. (Refer to...) Figure 6 , Figure 6 This is a flowchart illustrating the third embodiment of the network anomaly repair method of the present invention.

[0069] In this embodiment, the method further includes the following after step S40: Step S41: Match the network anomaly with the repair strategy for resolving the network anomaly to obtain the fault rule of the network anomaly.

[0070] Understandably, pre-defined fault rule bases are typically built upon the prior knowledge of developers in the field. Once deployed, these rules are fixed, and even if updates are possible, they require specialized technical personnel to rebuild the rule base or add new rules. The network system cannot dynamically update the pre-defined fault rule base based on actual processing results. This means that when facing complex and ever-changing real-world vehicular network environments, the processing strategies may not be optimal, or may even be ineffective, preventing the achievement of a virtuous cycle of dynamic updates to the pre-defined fault rules.

[0071] It should be noted that the fault rules are mapping rules set in the preset fault rule base between the state of network anomalies, i.e., fault characteristics, and the repair strategies. Once the state of network anomalies, i.e. fault characteristics, is determined, the repair strategy corresponding to the fault characteristics can be mapped out by the fault rule once the fault characteristics are successfully matched.

[0072] It should be understood that if the network state is no longer in a state of network abnormality when the repair strategy obtained by traversing the repair strategy library is used, it means that the unknown network abnormality currently encountered by the network can be resolved by the traversal repair strategy. At this time, the unknown network abnormality and the repair strategy that resolves the network abnormality can also form a fault rule. Once a network abnormality with the same fault characteristics appears in the future, it can be repaired by the repair strategy.

[0073] In practice, if it is confirmed that the network anomaly has been resolved by the repair strategy that has been traversed, the network anomaly can be matched with the repair strategy, and then a new fault rule can be established between the network anomaly and the repair strategy.

[0074] Step S42: Update the fault rules to the preset fault rule library.

[0075] It should be understood that when a new fault rule is established between an unknown network anomaly and its corresponding repair strategy, this fault rule can be directly updated to the preset fault rule base. Then, in the next occurrence of the same network anomaly, the corresponding repair strategy can be directly obtained by matching fault characteristics with the preset fault rule base, achieving dynamic updates to the preset fault rule base. As the network is used for a longer period, the corresponding preset fault rule base becomes increasingly comprehensive. For example, in the case of vehicle-to-everything (V2X) communication, the more vehicles travel and the more diverse the scenarios encountered, the more unknown network anomalies are encountered, resulting in a more complete preset fault rule base and more accurate and rapid decision-making. The entire V2X system can continuously adapt to new network environments and technological changes, possessing a long-term vitality and robustness unmatched by traditional static systems.

[0076] Specifically, step S42 includes: Step S421: Record the number of times the repair strategy resolves the network anomaly.

[0077] Understandably, when using the repair strategy obtained through traversal to resolve network anomalies, considering that the repair process may have a certain degree of randomness, in order to avoid the formed fault rules failing to resolve the same network anomalies later, it is also necessary to exclude the randomness of the repair strategy in resolving the network anomaly.

[0078] In practice, the number of times the repair strategy resolves the same network anomaly during network operation can be recorded. If the repair strategy resolves the network anomaly a large number of times or accounts for a large proportion of the total number of times it resolves the network anomaly, it can be concluded that the problem of the repair strategy resolving the network anomaly is not accidental.

[0079] Step S422: If the number of attempts reaches a preset number, update the fault rule to the preset fault rule library.

[0080] It should be noted that the preset number of attempts is a pre-defined number of times used to determine whether a repair strategy can successfully resolve the current network anomaly. This preset number of attempts can be set multiple times, such as 3 or 5 times. If the repair strategy can resolve the network anomaly the preset number of times, it can be considered that the repair strategy has universal applicability in resolving the current network anomaly, and consequently, the corresponding fault rules also have universal applicability.

[0081] In practice, the number of times the repair strategy resolves network anomalies can be compared with the preset number of times. If the number of times is not less than the preset number of times, it can be determined that the fault rule after the repair strategy and the network anomaly are matched can be used to resolve the same network anomaly in the future. Therefore, the fault rule can be updated to the preset fault rule library.

[0082] Step S423: If the number of attempts does not reach the preset number, increase the traversal priority of the repair strategy in the repair strategy library.

[0083] It is understandable that if the repair strategy encountered can resolve the unknown network anomaly, although there may be some randomness involved, it also indicates that the repair strategy can resolve the network anomaly under certain circumstances.

[0084] Therefore, even if the number of times the repair strategy resolves network anomalies does not reach the preset number, the repair strategy can still be used to repair network anomalies in the event of the same network anomaly, even without updating the fault rules to the preset fault rule base, which can greatly improve the efficiency of network anomaly repair.

[0085] In practice, the repair strategy can be recorded based on the network anomaly. If the repair strategy fails to resolve the same network anomaly a preset number of times, the repair strategy library needs to be traversed. During the traversal of the repair strategy library, the repair strategy can be traversed first, and the network anomaly can be repaired using this strategy first, thereby improving the overall efficiency of the traversal process and the network anomaly repair process.

[0086] Furthermore, a fourth embodiment of the network anomaly repair method of this application is proposed based on any of the embodiments in the first to third embodiments described above. (Refer to...) Figure 7 , Figure 7 This is a flowchart illustrating the fourth embodiment of the network anomaly repair method of the present invention.

[0087] In this embodiment, after step S40, the method further includes: Step S50: Detect the network status after network anomaly repair.

[0088] Step S60: If the network status is no longer considered abnormal after network anomaly repair, then the network anomaly repair is considered complete.

[0089] Step S70: If the network status is still abnormal after network anomaly repair, then repeat the step of traversing the repair strategy library until the network status is no longer abnormal after network anomaly repair.

[0090] It should be understood that during the traversal, only one repair strategy is selected from the repair strategy library to fix the network anomaly. However, this repair strategy may not be able to resolve the network anomaly. Therefore, after using a repair strategy to fix the network anomaly, it is necessary to determine whether the network state is still abnormal. If the network state is still abnormal, it can be concluded that the currently traversed repair strategy cannot resolve the network anomaly. In this case, it is necessary to traverse the repair strategy library again to obtain another repair strategy and fix the network anomaly again. This process is repeated to utilize all the repair strategies in the repair strategy library to fix the network anomaly. If the network state is no longer abnormal after being repaired by a repair strategy, it can be determined that the existing network anomaly has been successfully fixed.

[0091] In practice, after repairing the network anomaly using the traversed repair strategies, the network status after repair can be re-checked. If the network status after repair is no longer considered abnormal, the network anomaly repair is considered complete, and network status monitoring continues. If the network status is still abnormal after repair, the existing network anomaly is considered to still exist. In this case, the repair strategy library needs to be traversed again, and the network repair should continue using the re-traversed repair strategies. After the network repair process is complete, the network status should be monitored again until the network status is no longer abnormal. At this point, the network anomaly can be considered repaired, and the network fault resolved.

[0092] In this embodiment, the network status after the repair strategy is executed is further detected to determine whether the repair strategy can restore the network status to a normal state. If the network status is still abnormal, the network repair can be further executed by traversing other repair strategies to ensure that the network status can be restored to a normal state.

[0093] Furthermore, this embodiment of the invention also proposes a storage medium storing a network anomaly repair program, which, when executed by a processor, implements the steps of the network anomaly repair method described above.

[0094] Reference Figure 8 , Figure 8 This is a structural block diagram of the first embodiment of the network anomaly repair device of the present invention.

[0095] like Figure 7 As shown, the network anomaly repair device proposed in this embodiment of the invention includes: The network status monitoring module 10 is used to acquire network fault characteristics when the network status is abnormal. Anomaly repair decision module 20 is used to match the fault characteristics with the fault characteristics set in the preset fault rule base; The anomaly repair decision module 20 is also used to traverse the repair strategy library when the fault feature matching fails, the repair strategy library including at least one repair strategy for network anomalies. The repair strategy execution module 30 is used to repair network anomalies according to the currently traversed repair strategy.

[0096] The network status monitoring module 10 is mainly used to collect and preprocess all key performance indicators of the terminal in real time, such as signal strength, network connection status, data transmission rate, packet loss rate, and latency. This module continuously monitors the network health status. The anomaly repair decision module 20 acts as the control center for the entire repair process. It can receive data output from the network status monitoring module 10. This module can have two core sub-units: a rule matching unit and a strategy traversal unit. The rule matching unit matches fault characteristics and can have a pre-set fault rule library, such as a base station reselection repair strategy when the signal strength is less than -110 dBm. The strategy traversal unit stores a repair strategy library containing various repair strategies. The repair strategy execution module 30 is the execution module in the network repair process. It receives instructions from the anomaly repair decision module 20 and calls the terminal's underlying network interface (such as the AT commands of a modem) to actually execute specific repair strategies to repair network anomalies.

[0097] In addition, in this embodiment, the network anomaly repair device also includes: a preset fault rule base update module 40, which can record the full-link data of each anomaly handling, and perform knowledge extraction and fault rule self-generation through feedback loop, and update the fault rules to the preset fault rule base to complete the dynamic update of the preset fault rule base.

[0098] In this embodiment, when the network status is abnormal, the network status monitoring module 10 acquires the fault characteristics of the network, and then uses the anomaly repair decision module 20 to match the fault characteristics with the fault characteristics set in the preset fault rule base. If the fault characteristic matching fails, the anomaly repair decision module 20 continuously traverses the repair strategy base, and the repair strategy execution module 30 uses the traversed repair strategies to repair the network anomaly, thereby timely resolving network anomalies not covered by the preset fault rule base.

[0099] The specific implementation of the network anomaly repair device of the present invention can be referred to the above-described method embodiments, and will not be repeated here.

[0100] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0101] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0102] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as read-only memory / random access memory, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0103] The above are merely preferred embodiments of the present invention and do not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for repairing network anomalies, characterized in that, The network anomaly repair method includes: When the network status is abnormal, obtain the fault characteristics of the network; The fault characteristics are matched with the fault characteristics set in the preset fault rule base; If the fault feature matching fails, the repair strategy library is traversed, which includes at least one repair strategy for network anomalies. The network anomaly is repaired according to the currently traversed repair strategy.

2. The network anomaly repair method as described in claim 1, characterized in that, Before traversing the repair strategy library in the event of a failure to match the fault features, the method further includes: Obtain a set of remediation strategies to resolve network anomalies; Obtain the historical repair success rate of each repair strategy in the repair strategy set; The repair strategy library is obtained by sorting the repair strategies in the repair strategy set according to the historical repair success rate.

3. The network anomaly repair method as described in claim 1, characterized in that, The step of traversing the repair strategy library when the fault feature matching fails includes: Obtain multi-dimensional key performance data at moments when the network status is abnormal; An anomaly feature vector is constructed based on the aforementioned multi-dimensional key performance data; Calculate the similarity between the abnormal feature vector and the regular feature vector corresponding to each repair strategy in the repair strategy library; The repair strategies in the repair strategy library are traversed in descending order of similarity.

4. The network anomaly repair method as described in claim 1, characterized in that, After performing network anomaly repair on the network according to the currently traversed repair strategy, the process further includes: The fault rules for the network anomaly are obtained by matching the network anomaly with the repair strategy for resolving the network anomaly. Update the fault rules to the preset fault rule library.

5. The network anomaly repair method as described in claim 4, characterized in that, The step of updating the fault rule to the preset fault rule library includes: Record the number of times the repair strategy resolves the network anomaly; If the number of attempts reaches a preset number, the fault rule will be updated to a preset fault rule library. If the number of iterations does not reach the preset number, the priority of the repair strategy in the repair strategy library is increased.

6. The network anomaly repair method as described in claim 1, characterized in that, After performing network anomaly repair on the network according to the currently traversed repair strategy, the process further includes: Detect the network status after network anomaly repair; If the network status is no longer considered abnormal after network anomaly repair, then the network anomaly repair is considered complete.

7. The network anomaly repair method as described in claim 6, characterized in that, After the network status is detected and repaired, the process also includes... If the network status remains abnormal after network anomaly repair, the step of traversing the repair strategy library is repeated until the network status is no longer abnormal after network anomaly repair.

8. The network anomaly repair method as described in claim 1, characterized in that, After matching the fault features with the fault features set in the preset fault rule base, the method further includes: If the network anomaly is successfully matched, the network anomaly is repaired using the repair strategy corresponding to the matched fault characteristics.

9. A network anomaly repair device, characterized in that, The device includes: a memory, a processor, and a network anomaly repair program stored in the memory and executable on the processor, the network anomaly repair program being configured to implement the steps of the network anomaly repair method as described in any one of claims 1 to 7.

10. A storage medium, characterized in that, The storage medium stores a network anomaly repair program, which, when executed by a processor, implements the steps of the network anomaly repair method as described in any one of claims 1 to 7.

Images