Fault repair method, device, electronic device, and storage medium

Abstract

An embodiment of the present application relates to the field of communications and discloses a method, device, electronic device and storage medium for repairing a fault, the method including the steps of acquiring fault characteristics of a fault, determining whether there is a fault cause corresponding to the fault characteristics, determining whether there is a repair rule corresponding to the fault cause if a corresponding fault cause exists, repairing the fault according to the repair rule if a corresponding repair rule exists, and repairing the fault according to the repair command after reporting the fault cause and obtaining a repair command if a corresponding repair rule does not exist, and updating the repair rule base.

Description

[Technical Field]

[0001] This application is based on and claims priority from a Chinese patent application bearing application number "202111177755.5" and filed on October 9, 2021, the entire contents of which are hereby incorporated by reference into this application.

[0002] The present application relates to the field of communications, and in particular to a method, device, electronic device and storage medium for repairing a fault. [Background technology]

[0003] With the rapid development of information technology, base station equipment has evolved from 2G to 5G, becoming increasingly complex. A base station equipment is composed of many operating modules, and the connections between different modules are tight. A failure in one module will always cause a chain reaction, and even cause the entire system to malfunction. Base station equipment is always built on high-altitude buildings such as steel towers, and at the same time, it operates in harsh environments such as extreme heat and cold, remote mountainous areas, and high-altitude regions. Fault detection of base station equipment generally requires the assistance of specialized instruments, which rapidly increases the difficulty and cost of fault detection and repair of base station equipment. Summary of the Invention [Problem to be solved by the invention]

[0004] Therefore, how to quickly detect, locate and repair faults in base station equipment without being limited by external instrumentation is an urgent problem that needs to be solved. [Means for solving the problem]

[0005] An embodiment of the present application provides a method for repairing a failure, including the steps of: acquiring failure characteristics of the failure; determining whether there is a failure cause corresponding to the failure characteristics; if there is a corresponding failure cause, determining whether there is a repair rule corresponding to the failure cause; if there is a corresponding repair rule, repairing the failure according to the repair rule; if there is no corresponding repair rule, reporting the failure cause, obtaining a repair instruction, and then repairing the failure according to the repair instruction and updating the repair rule base.

[0006] An embodiment of the present application further provides a failure repair device, including: a feature module for obtaining failure features of the failure; a diagnosis module for obtaining a failure cause of the failure based on the failure features; and a repair module for searching whether a repair rule corresponding to the failure cause exists in a repair rule base, and if so, repairing the failure according to the repair rule; or if not, reporting the failure cause, obtaining a repair instruction, and then repairing the failure according to the repair instruction, and updating the repaired rule base.

[0007] An embodiment of the present application further provides an electronic device, including at least one processor and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and when the instructions are executed by the at least one processor, the at least one processor can perform the above-mentioned fault repair method.

[0008] An embodiment of the present application further provides a computer-readable storage medium having a computer program stored therein, the computer program implementing the above-described fault recovery method when executed by a processor. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a schematic diagram of an internal structure of a base station device according to an embodiment of the present application. [Figure 2]1 is an internal block diagram of an intelligent fault diagnosis system according to an embodiment of the present application; [Figure 3] 1 is a flowchart of a fault recovery method according to an embodiment of the present application; [Figure 4] 2 is a flowchart of an ACPR blind identification method according to an embodiment of the present application; [Figure 5] FIG. 2 is a structural schematic diagram of a rule base according to an embodiment of the present application; [Figure 6] 1 is an interaction flowchart of a fault repair method according to an embodiment of the present application; [Figure 7] 1 is a structural schematic diagram of a fault recovery device according to an embodiment of the present application; [Figure 8] 1 is a structural schematic diagram of an electronic device according to an embodiment of the present application; DETAILED DESCRIPTION OF THE INVENTION

[0010] In order to clarify the objectives, technical solutions, and advantages of the embodiments of the present application, the following detailed description of each embodiment will be given with reference to the accompanying drawings. However, those skilled in the art will understand that many technical details are proposed in each embodiment of the present application to help readers better understand the present application. However, even without these technical details and various modifications and variations based on the following embodiments, the technical solutions claimed in the present application can be realized.

[0011] The objective of the present application is to provide a fault repair method, device, electronic equipment and storage medium to solve the above problems, and to realize online intelligent diagnosis and repair of faults in base station equipment.

[0012] One embodiment of the present application relates to a fault repair method, including the steps of obtaining fault characteristics of a fault, determining whether there is a fault cause corresponding to the fault characteristics, and if a corresponding fault cause exists, determining whether there is a repair rule corresponding to the fault cause, and if a corresponding repair rule exists, repairing the fault according to the repair rule, and if a corresponding repair rule does not exist, reporting the fault cause, obtaining a repair command, and then repairing the fault according to the repair command and updating the repair rule base, thereby realizing online intelligent diagnosis and repair of faults in base station equipment.

[0013] The fault repair method of the embodiment of the present application is applied to a base station, where the fault repair method of the embodiment of the present application can be realized by pre-installing a diagnostic system in the base station, and the fault repair method of the embodiment of the present application can be applied to situations such as laboratory debugging of base station equipment, production line production, and field production.

[0014] In the embodiment of the present application, the fault characteristics are acquired in real time to find the corresponding fault cause, and the corresponding repair rule is inquired according to the fault cause, and the fault is repaired or the repair rule is updated according to the inquiring result, thereby realizing online intelligent diagnosis and repair of the fault of the base station equipment, and greatly improving the intelligent level of operation and maintenance of the base station equipment and the user experience.

[0015] The following provides a detailed description of the implementation of the data scheduling method of this embodiment. The following content is merely implementation details provided for ease of understanding and is not essential for implementing this method.

[0016] As shown in FIG. 1, in this application, to facilitate online diagnosis, the base station equipment is simplified into components such as a baseband module, an intermediate frequency module, a DA / AD module, a radio frequency link, a power amplifier, and an antenna array. The diagnostic system inputs two components: the operating status of each module in the base station equipment and the feedback signal of the channel power amplifier port or antenna port. The diagnostic system outputs fault repair instructions and a human-machine interface. The diagnostic system includes modules such as feature extraction, rule-based, diagnostic reasoning, fault repair, interpreter, and human-machine interaction. Here, the inputs of the system modules are the fault features and key point data of each module in the RRU (Remote Radio Unit) or AAU (Active Antenna Unit), and the system outputs the root cause of the fault and repair instructions.

[0017] The internal realization of the intelligent fault diagnosis system is shown in Figure 2, and the diagnosis system may be composed of the feature extraction, rule base, diagnostic reasoning, fault repair, human-machine interaction, and interpreter modules in Figure 2.

[0018] The diagnostic system operates as follows: a feature extraction module is used to generate a fault feature set, which is then combined with a rule base to perform diagnostic inference to identify and report the root cause of the fault; simultaneously, an online repair rule base is used to issue a fault repair command; finally, a re-detection is performed to determine whether the fault has been repaired and a new diagnosis is performed; if it is discovered during the rule-based diagnostic inference process that no rule corresponding to the current fault exists, a machine learning diagnostic inference is performed to generalize and infer the root cause of the fault; the fault feature set and the diagnostic result are reported to the engineer for judgment; the engineer generates a new diagnostic result based on the information collected by the diagnostic system, stores it in the diagnostic system, and forms a repair command; finally, a re-detection is performed to determine whether the fault has been repaired and a new diagnosis is performed; once the fault is resolved, the diagnostic system adds the fault feature set and the diagnostic result to the machine learning training sample to generate a new fault rule; finally, a self-detection is performed using a fault sample whose root cause is known; if the new rule passes the self-detection after it has been added, it is added to the rule base; here, an interpretation of the fault's root cause and process is transmitted to the user interface via the human-machine interface and displayed.

[0019] The specific functions realized by each module are as follows:

[0020] (1) Feature extraction: The fault feature flags of each module of the base station equipment are extracted and classified into function-class fault features and performance-class fault features. The function-class fault features are calculated using the key node flags of each module, while the performance-class fault features need to be calculated using blind identification of repeated data of each module. The diagnostic system does not need to collect data from each channel of the base station equipment, but only needs to collect the fault features reported by each module of each channel.

[0021] (2) Rule base: Based on the operating principles of the base station equipment and the division of modules, a rule base is created that defines the mutual influence relationships between the fault feature flags of each module. To further improve the efficiency of diagnostic inference, the rule base is created by separating the rules of each module, creating hierarchical rules within the modules, and linking the rules between top-level modules. The newly added online repair rule base enables online repair of the root cause of the fault.

[0022] (3) Diagnostic inference: Includes rule-based diagnostic inference and machine learning diagnostic inference. Rule-based inference uses a forward or backward inference algorithm to infer the root cause of a failure based on the fault characteristics of the diagnostic system, the module-level rule base, and the rule base within the module. Machine learning inference uses a low-complexity learning algorithm such as a decision tree to perform rule training on complex fault samples without clear rules, and then uses a sample library with known fault root causes to complete rule self-detection and add new rules to the rule base.

[0023] (4) Fault repair: Based on the root cause of the fault and the online repair rule base, determine whether the root cause of the current fault can be repaired online, and if so, repair the fault online.

[0024] (5) Human-machine interaction and interpreter: Interpret the fault feature set, the root cause of the fault, and the reasoning process as needed, and display them graphically on the user interface through the human-machine interface, allowing for convenient interaction with engineers and improving the user experience.

[0025] The specific process of the fault repair method of the present application is shown in FIG. 3, and may include the following steps 301 to 304.

[0026] In step 301, the fault characteristics of the fault are obtained.

[0027] Specifically, the base station extracts fault feature flags of each module of the base station device, where the fault features are classified into function classes and performance classes, the function class fault features are calculated using key node flags of each module of the base station device, and the performance class fault features are calculated using blind identification of repeated data of each module, where the function fault features are used to characterize the hardware state of each equipment function module, and the performance fault features are used to characterize the signal processing performance of each of the equipment function modules, and the function class fault features may be reported to the diagnostic system in real time by each equipment function module, where the performance fault features may be obtained by a detection module of the base station based on the signal processing performance parameters of each of the equipment function modules.

[0028] In one example, the base station first extracts fault features and then generates a fault feature set, i.e., the step of obtaining fault features in the diagnostic system may include uploading basic features via each module, the diagnostic system further extracting the basic features, and finally obtaining fault features available to the diagnostic system.

[0029] In one example, the base station acquires a transmission / reception signal of an equipment function module, the base station identifies configuration parameters of the transmission / reception signal, the base station calculates a performance index of the equipment function module based on the configuration parameters, and the base station obtains the performance failure characteristic based on the performance index.

[0030] In one example, the base station extracts performance failure features by calculating the ACPR (Adjacent Channel Power Ratio) index. The ACPR blind identification method is shown in Figure 4, and the specific process is as follows:

[0031] First, the ACPR calculation module must blindly identify the constituent parameters of the signal to be analyzed (such as the number of carriers, frequency points, and bandwidth). That is, the blind identification method is as follows: collect the signal to be analyzed, add a window to the signal data to be analyzed, then calculate the power spectrum using PSD (Power Spectral Density), and set a power threshold to determine information such as the number of carriers, bandwidth size, and carrier frequency points of the signal to be analyzed.

[0032] Here, in order to improve the accuracy of blind identification of multi-carrier unequal power signals, the signals to be analyzed are the forward signal and feedback signal after time delay adjustment, the forward signal is a signal without nonlinear distortion before the transmit channel DPD (digital pre-distortion) module in the base station device and is used for blind identification of the signal configuration parameters, the feedback signal is a signal including the power amplifier nonlinear distortion coupled from the output of the power amplifier and is used to calculate the ACPR index and determine the degree of nonlinearity of a single channel, and after time delay adjustment of the forward signal and feedback signal, the configuration parameters of the signal identified by the forward signal may be provided to the feedback signal and used for ACPR calculation.

[0033] In addition, the calculation of ACPR blind identification can be realized through an iterative process such as the DPD module of the base station device, eliminating the need for steps such as separately collecting samples and adjusting time delays, and allowing the ACPR performance indicator to be reported without affecting the normal operation process of the base station device.

[0034] In one example, the performance index of the base station device may be an EVM index, an amplitude difference between channels, or the like, and the above index may be used as an index for determining the performance level of the base station device.

[0035] In this embodiment, by using a method for reporting function-class fault features in real time and blindly identifying performance-class fault features, on the one hand, it is possible to break away from the limitations of external instruments and realize online real-time extraction of fault features, and on the other hand, it is possible to convert a large amount of equipment fault data into fault features in advance, and use the fault features for transmission between the diagnosis system and the equipment to be diagnosed, thereby reducing the amount of input data to the diagnosis system. Breaking away from external limitations on fault feature extraction and reducing the amount of data transmission are important prerequisites for realizing online intelligent diagnosis of base station equipment faults.

[0036] In step 302, the base station determines whether there is a fault cause corresponding to the fault feature.

[0037] Specifically, the base station determines the equipment function module where the fault is located, and obtains the fault cause corresponding to the fault feature according to the diagnostic rule corresponding to the equipment function module in the pre-set diagnostic rule base, where the diagnostic rule is the correspondence relationship between the fault feature and the fault cause.

[0038] For example, in an embodiment of the present application, the cause of a fault can be determined by rule-based diagnostic inference, requiring the creation of a rule base. The rule base is created with the idea that the rules for each module are separated and the hierarchy of rules within each module is clearly defined, with the relationships between modules established by rules at the highest module level. The specific module division and interrelationships are created based on the operating principles of the base station device. During the fault diagnostic inference process, the module-level rule base is first used to identify the module in which the root cause of the fault exists, and then diagnostic inference is performed within the faulty module, thereby avoiding the involvement of unrelated module rules in the fault inference and improving diagnostic efficiency. A structural diagram of the rule base is shown in Figure 5.

[0039] In addition, the rule base of base station devices of the same platform can be shared, and if there are some differences between base station devices of different platforms, it is only necessary to change the modules and the corresponding rules within the modules based on the operating principles.

[0040] In one example, if there is a rule corresponding to the fault feature in the rule base, the base station determines that there is a fault cause corresponding to the fault feature and executes step 303; if there is no rule corresponding to the fault feature in the rule base, the base station determines that there is no fault cause corresponding to the fault feature, and the base station starts diagnostic inference by machine learning to infer the generalized root cause of the fault, and reports the fault feature set and the diagnostic result to the engineer for judgment; the engineer issues a new diagnostic result based on the information collected by the diagnostic system, stores it in the diagnostic system, and forms a repair instruction. After the engineer successfully repairs the fault, the diagnostic system uses the new fault feature set and diagnostic parameters Add Finally, self-detection is performed on the new rules and diagnostic parameters using a fault sample whose root cause is known. If self-detection is successful, the new rules are added to the rule base and the machine learning diagnostic parameters are updated. If self-detection fails, the machine learning parameters are optimized and retrained. Here, the root cause and process of the fault are communicated to the user interface and displayed via the man-machine interface. Here, self-detection after updating the rule base can ensure the accuracy of the rule repair.

[0041] In one example, the machine learning algorithm is selected as a decision tree algorithm, which has low complexity and high interpretability (in actual applications, the algorithm is not limited to a decision tree algorithm, and other machine learning algorithms, such as a vector machine algorithm, which have low complexity and high interpretability, may be selected).

[0042] A decision tree is an intelligent method based on machine learning theory, and it consists of decision nodes, branches, and leaf nodes. It essentially learns from if / else problems layer by layer and reaches a conclusion. The branches of a decision tree correspond to fault rules, and the attributes and their values ​​from the root node to the leaf nodes correspond to rule conditions, and the leaf nodes are rule conclusions. Decision trees have low complexity and are capable of self-learning and certain generalized reasoning capabilities.

[0043] For example, the decision tree C4.5 algorithm uses the information gain rate as the selection criterion, and the specific steps for building a decision tree in C4.5 are as follows:

[0044] (1) For the acquired fault feature set, calculate the information gain and information gain rate of each feature, select the feature with the largest information gain rate as the current fault feature node, and obtain the root node of the decision tree.

[0045] (2) Based on each possible value of the node attribute, corresponding to the subset, recursively operate on the sample subsets and perform the process of step (1) until the data in each subset has the same value for the classification attribute, thereby generating a decision tree.

[0046] (3) Fault rules are extracted based on the constructed decision tree, and diagnosis and self-detection are performed for the new fault feature set. The diagnostic system performs self-detection based on an existing rule base or a pre-stored fault sample set whose root causes are known.

[0047] (4) Under the assumption of ensuring correctness, appropriate pruning is performed on the generated decision tree to simplify the decision tree structure.

[0048] JPEG0007812917000001.jpg84170

[0049] JPEG0007812917000002.jpg63170

[0050] In step 303, if there is a corresponding failure cause, the base station determines whether there is a restoration rule corresponding to the failure cause.

[0051] In one example, before the base station determines whether there is a repair rule corresponding to the cause of the failure, the base station performs rule-based diagnostic inference to determine the root cause of the failure, where the inference algorithm used in the rule-based diagnostic inference is a conventional forward inference algorithm, and the base station device matches the failure feature set to the rule base and retains the successfully matched rule until all rules in the rule base to be matched are matched, and the finally retained rule can diagnose the root cause of the failure.

[0052] Specifically, the base station performs rule-based diagnostic reasoning to identify the root cause of the failure, and then determines whether the root cause of the failure can be repaired online based on the online repair rule base, i.e., determines whether there is a repair rule corresponding to the failure cause.

[0053] In step 304, if a corresponding repair rule exists, the base station repairs the fault according to the repair rule, i.e., executes step 304-1; if no corresponding repair rule exists, the base station reports the fault cause, obtains a repair command, and then repairs the fault according to the repair command and updates the repair rule base, i.e., executes step 304-2.

[0054] In one example, if there is no repair rule corresponding to the cause of the failure, the root cause of the failure is reported to the man-machine interface and instructed to an engineer via the user interface. The engineer repairs based on the root cause of the failure, and after the repair is successful, updates the repair rule for the root cause of the failure to the online repair rule base.

[0055] The embodiment of the present application uses a human-machine interaction interface to display the overall fault status of the RRU or AAU on the interface, and highlights the faults or potential faults of each module of the base station equipment, thereby greatly improving the user experience.

[0056] The embodiments of the present application mainly combine fault diagnosis and artificial intelligence technologies to propose an intelligent fault diagnosis method for base station equipment, which can break away from the limitations of the external environment and instruments, and achieve functions such as online fault diagnosis, prediction and repair of base station equipment, and has functions of online self-learning and self-detection, which can greatly improve the intelligent level of base station equipment operation and maintenance, and greatly expand the application scope of intelligent fault diagnosis of base station equipment. The embodiments of the present application can achieve online intelligent fault diagnosis and repair of base station equipment, and at the same time have functions such as online self-learning and self-repair, so they have low requirements on the network environment, etc.

[0057] In order to make the process of the fault repair method in the embodiment of the present application clearer, this embodiment further provides a fault repair method, as shown in FIG. 6, the specific steps are as follows:

[0058] In step 601, the base station obtains the fault characteristics of the fault.

[0059] In step 602, the base station determines whether there is a fault cause corresponding to the fault feature.

[0060] In one example, if there is no rule corresponding to the fault feature in the rule base, it is considered that there is no fault cause corresponding to the fault feature, and the corresponding operations of steps 604 to 608 are performed; if there is a rule corresponding to the fault feature in the rule base, it is considered that there is a fault cause corresponding to the fault feature, and step 603 and the following steps are performed.

[0061] In step 604, machine learning diagnostic inference is performed to provide generalized root causes of failures and report the fault feature set and root causes of failures.

[0062] Specifically, if there is no rule corresponding to the fault feature in the rule base, the base station will consider that there is no fault cause corresponding to the fault feature, and the base station will start diagnostic inference using machine learning to infer the generalized root cause of the fault, and report the fault feature set and diagnostic results to the engineer for judgment.

[0063] In step 605, the engineer diagnoses and repairs the fault and sends a repair instruction to the base station, and after the base station receives the repair instruction, the base station repairs the fault according to the repair instruction and updates the repair rule base.

[0064] Specifically, the base station can add new repair rules through machine learning. After obtaining repair instructions, the base station uses the repair instructions as new examples for machine learning training to generate new rules and diagnostic parameters, and updates the repair rule base. Here, machine learning inference refers to using a low-complexity learning algorithm such as a decision tree to perform rule training on complex fault samples without clear rules, and using a sample library with known root causes of the faults to complete rule self-detection and add new rules to the rule base.

[0065] In one example, an engineer generates new diagnostic results based on the information collected by the diagnostic system, stores them in the diagnostic system, and forms repair instructions. After the engineer successfully repairs the fault, the diagnostic system adds the new fault feature set and diagnostic results to machine learning training samples, and generates new fault rules and diagnostic parameters through machine learning training.

[0066] Step 606 involves self-detecting the correctness of the new rule base and diagnostic parameters.

[0067] In one example, after generating new fault rules and diagnostic parameters through machine learning training, the base station performs self-detection on the new rules and diagnostic parameters using fault samples whose root causes are known, and if the self-detection is successful, performs step 607, that is, adds the new rules to the rule base and updates the machine learning diagnostic parameters, and if the self-detection is not successful, performs step 608, that is, optimizes the machine learning parameters and performs retraining. Here, machine learning inference is to use a low-complexity learning algorithm such as a decision tree to perform rule training on complex fault samples without clear rules, and utilize a sample library whose root causes are known to complete rule self-detection and add the new rules to the rule base.

[0068] In step 603, rule-based diagnostic inference is performed to determine the root cause of the failure. Here, the inference algorithm used in the rule-based diagnostic inference is a conventional forward inference algorithm, and the base station device matches the failure feature set to the rule base and retains the successfully matched rules until all rules in the rule base to be matched are matched. The finally retained rule can diagnose the root cause of the failure.

[0069] In step 609, it is determined whether there is a repair rule corresponding to the cause of the failure.

[0070] In one example, the base station determines whether there is a repair rule corresponding to the root cause of the failure based on the root cause of the failure determined in step 603. If it determines that there is no repair rule corresponding to the failure cause, it performs the corresponding operations of above steps 605 to 608. Specifically, if there is no corresponding repair rule, it reports the root cause of the failure to the man-machine interface and instructs the engineer through the user interface. The engineer repairs based on the root cause of the failure. After the repair is successful, it updates the repair rule of the root cause of the failure in the online repair rule base. If it determines that there is a repair rule corresponding to the failure cause, it performs step 610 and the following steps.

[0071] In step 610, the fault is repaired according to the repair rules.

[0072] Specifically, the base station determines that the root cause of the fault can be repaired in combination with the online repair rule base, and then issues a fault repair command to repair the fault.

[0073] In step 611, it is determined whether the failure has been successfully repaired.

[0074] Here, if the fault has been repaired, step 612 is performed, and if the fault has not been repaired successfully, step 601 is performed again.

[0075] In step 612, the diagnostic or repair results are interpreted and explained, where the interpretation is communicated to the user interface by the man-machine interface and displayed.

[0076] In one example, after step 607 above, ie, after updating the rule base and diagnostic parameters, the diagnostic results also need to be interpreted and explained.

[0077] Furthermore, the present application is not limited to intelligent online diagnosis of faults in base station equipment, but includes, for example:

[0078] (1) Radio frequency algorithm fault diagnosis field, such as the fault diagnosis field of radio frequency algorithms in CFR modules, DPD modules, etc. By utilizing this patent application, it is possible to realize online intelligent diagnosis of radio frequency algorithm faults by simply adding appropriate function and performance feature flags to modules such as CFR and DPD, and creating a corresponding rule base according to the operating principle.

[0079] (2) Baseband processing fault diagnosis field. For example, intelligent fault diagnosis of each module in a BBU (Building Baseband Unit) system. By utilizing this patent application, it is possible to realize online intelligent diagnosis of baseband processing faults simply by adding appropriate function and performance feature flags to the corresponding modules and creating a corresponding rule base according to the operating principle.

[0080] (3) The field of hardware equipment fault diagnosis, such as modules such as transceiver units and radio frequency modules. By utilizing this patent application, the fault processing of hardware equipment can be easily performed by simply adding appropriate function and performance feature flags to the corresponding modules and creating a corresponding rule base according to the operating principle. online It is possible to realize intelligent diagnosis of

[0081] The embodiments of this application mainly combine fault diagnosis and artificial intelligence technologies to propose an intelligent fault diagnosis method for base station equipment, which is free from the limitations of the external environment and instruments, and can realize functions such as online fault diagnosis, prediction and repair of base station equipment, and has the functions of online self-learning and self-detection, which can greatly improve the intelligent level of base station equipment operation and maintenance and greatly expand the application scope of intelligent fault diagnosis of base station equipment.

[0082] The steps of the various methods described above are divided for clarity of description only, and in implementation, they may be merged into one step or some steps may be divided into multiple steps, as long as they contain the same logical relationship, and are within the scope of protection of this patent. Adding non-critical modifications to the algorithms or processes or introducing non-critical designs without changing the core design of the algorithms or processes are also within the scope of protection of this patent.

[0083] The embodiment of the present application further provides a fault repair device, which includes a feature module 701, a diagnosis module 702, and a repair module 703, as shown in FIG.

[0084] Specifically, the feature module 701 is used to obtain the fault features of the fault, the diagnosis module 702 is used to obtain the fault cause of the fault based on the fault features, and the repair module 703 is used to search whether there is a repair rule corresponding to the fault cause in the repair rule base, and if there is, repair the fault according to the repair rule, or if there is no fault cause, report the fault cause, obtain a repair command, and then repair the fault according to the repair command, and update the repair rule base.

[0085] In one example, the fault features are divided into function classes and performance classes, the function fault features are used to characterize the hardware status of each equipment function module, and the performance fault features are used to characterize the signal processing performance of each of the equipment function modules, where the function class fault features are reported in real time, and the performance fault features are obtained based on the signal processing performance parameters of each of the equipment function modules.

[0086] In one example, the feature module 701 acquires the performance failure feature by the base station acquiring a transmission / reception signal of an equipment function module, the base station identifying configuration parameters of the transmission / reception signal, the base station calculating a performance indicator of the equipment function module based on the configuration parameters, and the base station obtaining the performance failure feature based on the performance indicator.

[0087] Specifically, the diagnostic module 702 determines the equipment function module where the fault is located, and obtains the fault cause corresponding to the fault feature according to the diagnostic rule corresponding to the equipment function module in the preset diagnostic rule base, where the diagnostic rule is the correspondence relationship between the fault feature and the fault cause.

[0088] In addition, when there is no corresponding failure cause, the failure cause is obtained by self-learning inference based on the correspondence between existing failure features and failure causes, and the self-learning may select a decision tree algorithm or a support vector machine algorithm, or may select other machine learning algorithms with low complexity and high interpretability.

[0089] In one example, after obtaining the failure cause of the failure, rule-based diagnostic inference is performed to identify the root cause of the failure; after identifying the root cause of the failure, it is determined whether the root cause of the failure can be repaired online based on the online repair rule base; if a repair rule exists, a repair instruction for the failure is issued; finally, it is detected whether the failure has been successfully repaired and a new detection diagnosis is performed; if a repair rule does not exist, the root cause of the failure is reported to the human-machine interface and instructed by the user interface to an engineer; the engineer repairs based on the root cause of the failure and uploads the repair instruction; the diagnostic system generates a new repair rule through self-learning according to the repair instruction, detects the correctness of the new repair rule based on the failure, the failure cause, and the new repair rule; if it detects that the new repair rule is correct, it updates the repair rule for the root cause of the failure to the online repair rule base; if it detects that the new repair rule is incorrect, it adjusts the parameters of the self-learning algorithm, regenerates a new repair rule, and detects the correctness of the regenerated new repair rule.

[0090] The fault repair device of this embodiment can realize functions such as online intelligent diagnosis and repair of base station equipment faults, online self-learning and self-detection of complex faults, etc., breaking away from the limitations of external computers and instruments, etc., and is widely applied to scenes such as laboratories, production manufacturing, and fields, greatly improving the intelligent level and user experience of base station equipment operation and maintenance.

[0091] Obviously, this embodiment is an apparatus embodiment corresponding to the above-mentioned embodiment of the fault repair method, and this embodiment can be implemented in combination with the above-mentioned embodiment of the fault repair method. The relevant technical details and technical effects mentioned in the above-mentioned embodiment of the fault repair method are also valid in this embodiment, and to reduce duplication, the redundant description will be omitted here. Correspondingly, the relevant technical details mentioned in this embodiment can also be applied to the above-mentioned embodiment.

[0092] It should be noted that each module in this embodiment is a logical module, and in actual application, one logical unit may be one physical unit, may be part of one physical unit, or may be realized by a combination of multiple physical units. In addition, in order to highlight the innovative aspects of this application, units that are not closely related to solving the technical problem mentioned in this application are not introduced in this embodiment, but this does not mean that there are other units in this embodiment.

[0093] Another embodiment of the present application relates to an electronic device, as shown in FIG. 8, including at least one processor 801 and a memory 802 communicatively connected to the at least one processor 801, wherein the memory 802 stores instructions executable by the at least one processor 801, and when the instructions are executed by the at least one processor 801, the at least one processor can perform the above-mentioned fault repair method.

[0094] Here, the memory and the processor are connected by a bus, which may include any number of interconnected buses and bridges, connecting various circuits of one or more processors and memories. The bus may also connect various other circuits, such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore will not be further described herein. The bus interface provides an interface between the bus and a transceiver. The transceiver may be a single element or multiple elements, such as multiple receivers and transmitters, providing a unit for communicating with various other devices over a transmission medium. Data processed by the processor is transmitted over a wireless medium via an antenna, which further receives data and transmits it to the processor.

[0095] The processor is responsible for managing the bus and normal processing and may also provide a variety of functions including timing, peripheral interfaces, voltage regulation, power management and other control functions. Memory can be used to store data used by the processor when performing operations.

[0096] The above product can execute the method according to the embodiment of the present application, and has functional modules and beneficial effects corresponding to the execution method. For technical details not described in detail in this embodiment, please refer to the method according to the embodiment of the present application.

[0097] An embodiment of the present application further provides a computer-readable storage medium having a computer program stored thereon, the computer program being adapted to implement the above method embodiment when executed by a processor.

[0098] As can be understood by those skilled in the art, all or some of the steps of implementing the methods of the above embodiments can be completed by a program issuing instructions to related hardware, and the program is stored in a storage medium and includes multiple instructions for causing a device (such as a single-chip microcomputer, chip, etc.) or a processor to execute all or some of the steps of the methods described in each embodiment of the present application. The storage medium mentioned above includes various media that can store program code, such as USB memory, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0099] The above embodiments are provided for those skilled in the art to realize and use the present application, and those skilled in the art may make various modifications or changes to the above embodiments without departing from the inventive concept of the present application. Therefore, the scope of protection of the present application is not limited to the above embodiments, but should correspond to the widest scope of novel features set forth in the claims.

Claims

1. 1. A method for repairing a fault, the method being performed by a base station, comprising: obtaining a fault signature of the base station fault; determining whether there is a fault cause corresponding to the fault feature; If a corresponding failure cause exists, determining whether there is a repair rule corresponding to the failure cause in a repair rule base; repairing the fault according to the corresponding repair rule if one exists; if there is no corresponding repair rule, reporting the cause of the failure to a man-machine interface, obtaining a repair command through the man-machine interface, repairing the failure according to the repair command, and updating the repair rule base; The step of determining whether there is a fault cause corresponding to the fault feature includes: determining the equipment functional module in which the fault is located; obtaining the fault cause corresponding to the fault feature according to a diagnostic rule corresponding to the device function module in a preset diagnostic rule base; The diagnostic rule is a correspondence relationship between the fault characteristics and the fault causes. How to repair the malfunction.

2. If there is no corresponding failure cause, the method further includes obtaining the failure cause by self-learning inference based on a correspondence between existing failure features and the failure cause. The method for repairing a fault according to claim 1 .

3. The self-learning is realized by a decision tree algorithm or a support vector machine algorithm. The method for repairing a fault according to claim 2.

4. The failure features include functional failure features and / or performance failure features; The functional failure characteristics are used to characterize the hardware status of each device functional module; the performance failure signatures are used to characterize the signal processing performance of each of the equipment function modules; The functional fault characteristics are obtained by reporting each of the equipment functional modules; The performance fault signature is obtained based on the signal processing performance parameters of each of the equipment function modules. The method for repairing a fault according to claim 1 .

5. The step of obtaining the performance fault signature based on the signal processing performance parameters of each of the equipment function modules includes: acquiring a transmission / reception signal of the device function module; identifying configuration parameters of the transmitted and received signals; calculating a performance index for the equipment function module based on the configuration parameters; and obtaining the performance fault signature based on the performance index. The method for repairing a fault according to claim 4.

6. The step of updating the repair rule base comprises: generating new repair rules through self-learning according to the repair instructions; detecting the correctness of the new repair rule according to the fault, the fault cause, and the new repair rule; If the new repair rule is correct, updating the repair rule base with the new repair rule; If the new repair rule is incorrect, adjusting parameters of a self-learning algorithm, regenerating a new repair rule, and detecting the correctness of the regenerated new repair rule. The method for repairing a fault according to claim 1 .

7. A failure repair device, a feature module for obtaining fault features of the base station fault; a diagnostic module for obtaining a fault cause of the fault based on the fault characteristics; Searching whether there is a repair rule corresponding to the cause of the failure in a repair rule base; If the repair rule exists, repairing the fault according to the repair rule; a repair module for reporting the cause of the failure to a human-machine interface if the repair rule does not exist, and for obtaining a repair command through the human-machine interface, and then repairing the failure according to the repair command and updating the repair rule base; The step of determining whether or not there is a fault cause corresponding to the fault feature includes: determining the equipment functional module in which the fault is located; obtaining the fault cause corresponding to the fault feature according to a diagnostic rule corresponding to the device function module in a preset diagnostic rule base; The diagnostic rule is a correspondence relationship between the fault characteristics and the fault causes. Malfunction repair device.

8. An electronic device, at least one processor; a memory communicatively coupled to the at least one processor; The memory stores instructions executable by the at least one processor, and when the instructions are executed by the at least one processor, the at least one processor can perform the failure repair method described in any one of claims 1 to 6. Electronic device.

9. A computer-readable storage medium having a computer program stored therein, the computer program realizing the method for repairing a fault according to any one of claims 1 to 6 when executed by a processor.

Images