CN116963134A

CN116963134A - Fault diagnosis method and device, terminal equipment and network platform

Info

Publication number: CN116963134A
Application number: CN202211279262.7A
Authority: CN
Inventors: 王晴; 赵睿; 万鸿俊; 王曦泽
Original assignee: China Mobile Communications Group Co Ltd; China Mobile Communications Ltd Research Institute
Current assignee: China Mobile Communications Group Co Ltd; China Mobile Communications Ltd Research Institute
Priority date: 2022-10-19
Filing date: 2022-10-19
Publication date: 2023-10-27

Abstract

The invention provides a fault diagnosis method, a fault diagnosis device, terminal equipment and a network platform. The method comprises the following steps: under the condition of network connection failure, collecting operation interaction data of terminal equipment; performing fault analysis according to the operation interaction data to obtain fault analysis data; and after the network is recovered to be normal, reporting fault analysis data to the network platform. By adopting the method, the terminal equipment and the network platform respectively perform data acquisition and fault detection, so that the network fault cause is obtained by end-to-side collaborative analysis, the network fault can be effectively solved, and the labor cost and the operation and maintenance cost are reduced.

Description

Fault diagnosis method and device, terminal equipment and network platform

Technical Field

The invention relates to the technical field of the Internet of things, in particular to a fault diagnosis method, a fault diagnosis device, terminal equipment and a network platform.

Background

The current automatic and digital operation and maintenance means are mainly realized through a remote network operation and maintenance platform, the terminal equipment reports the data such as the operation and the performance of the cellular network to the remote network operation and maintenance platform in real time, and the remote network operation and maintenance platform performs unified analysis, fault diagnosis and positioning delimitation.

In an industry network, equipment disconnection is an important and urgent fault which frequently occurs, and is related to production safety, and according to user feedback fault work order information counted by multiple manufacturers, abnormal connection events account for a large proportion. The reason for the disconnection may be a reason for the SIM card, such as arrearage; may be a field environmental cause, such as a power outage; may be the cause of the device itself, such as a crash; network link failures may also occur, resulting in interruption of communications. The existing method mainly relies on real-time data analysis, no matter which causes network connection faults, a network platform cannot acquire data from terminal equipment in real time, the fault causes are analyzed, most of abnormal fault disconnection conditions are caused, operation and maintenance personnel are required to go to the site to check the fault causes one by one according to logs, and then equipment manufacturers or operators are contacted according to the fault causes to solve the problems, so that a great deal of labor cost and time cost are consumed.

Disclosure of Invention

The technical scheme of the invention aims to provide a fault diagnosis method, a fault diagnosis device, terminal equipment and a network platform, which are used for solving the problems that when the equipment in the prior art fails to perform abnormal disconnection, the fault is required to be manually checked, the disconnection problem of the equipment cannot be timely and effectively solved, and a large amount of labor cost and time cost are consumed.

The embodiment of the invention provides a fault diagnosis method, which is executed by terminal equipment and comprises the following steps:

under the condition of network connection failure, collecting operation interaction data of the terminal equipment;

performing fault analysis according to the operation interaction data to obtain fault analysis data;

and after the network is recovered to be normal, reporting the fault analysis data to a network platform.

Optionally, the fault diagnosis method, wherein the method further includes:

under the normal condition of network connection, performing predictive analysis according to the operation interaction data of the terminal equipment to obtain a predictive result of whether network faults can be generated within a first preset duration;

and reporting alarm information to a network platform when the prediction result indicates that network faults can occur.

Optionally, in the fault diagnosis method, the running interaction data includes AT instructions and/or module interaction logs.

Optionally, the fault diagnosis method, wherein performing fault analysis according to the operation interaction data to obtain fault analysis data includes:

performing fault analysis of one or more of the following according to the operation interaction data to obtain fault analysis data:

Analyzing the problem of the SIM card;

analyzing module problems;

analyzing network coverage problems;

analyzing network connection problems;

analyzing authentication problems;

invalid message problem analysis.

Optionally, in the fault diagnosis method, after performing fault analysis according to the operation interaction data to obtain fault analysis data, the method further includes:

obtaining the fault type of the network connection fault according to the fault analysis data;

and controlling the gateway indicator lamp of the terminal equipment to carry out flickering prompt in a mode corresponding to the fault type.

The embodiment of the invention also provides a fault diagnosis method, which is executed by the network platform and comprises the following steps:

acquiring fault analysis data reported to a network platform by terminal equipment; the fault analysis data are obtained by carrying out fault analysis on the terminal equipment according to the collected operation interaction data when network connection faults are generated;

according to the fault analysis data, an abnormal sequence template of the terminal equipment is established;

when the network connection fault of the terminal equipment is monitored, detecting the abnormal state of the real-time network data to obtain abnormal data;

And matching the abnormal data with the abnormal sequence template, and judging the fault type of the network connection fault.

Optionally, the fault diagnosis method, wherein the real-time network data includes SIM card data of the terminal device obtained by a SIM card management platform, and/or network state data collected by the network platform within a second preset duration before a network connection fault occurs.

Optionally, the fault diagnosis method, wherein the detecting of the abnormal state of the real-time network data includes one or more of the following:

detecting abnormal states of the SIM card data;

detecting abnormal state of network quality index data in the network state data;

detecting abnormal state of the flow index data in the network state data;

and detecting the abnormal state of the equipment index data in the network state data.

Optionally, the fault diagnosis method, wherein the detecting the abnormal state of the real-time network data to obtain the abnormal data includes:

preprocessing the real-time network data to obtain processing data of the real-time network data;

extracting the characteristics of the processed data to obtain a characteristic matrix of the processed data;

Obtaining a plurality of detection sample data of the feature matrix by a binary tree construction method;

and carrying out abnormality judgment on each detection sample data, and judging whether the corresponding detection sample data is abnormal data or not.

Optionally, in the fault diagnosis method, performing an anomaly judgment on each detection sample data, and judging whether the corresponding detection sample data is anomaly data includes:

calculating the path length of the detection sample data in the constructed binary tree;

calculating an abnormal score of the detection sample data according to the path length;

and when the abnormal score is larger than a preset judgment value, determining the corresponding detection sample data as abnormal data.

Optionally, the fault diagnosis method, wherein performing feature extraction on the processing data to obtain a feature matrix of the processing data includes:

calculating the processing data by adopting different mean value calculation methods to obtain a predicted value corresponding to each mean value calculation method;

and generating a characteristic matrix of the processing data according to the difference value between each processing data and different predicted values.

Optionally, the fault diagnosis method, wherein the matching the abnormal data with the abnormal sequence template, and judging the fault type of the network fault, includes:

Constructing a distance mapping matrix according to the abnormal data and the abnormal sequence template;

calculating a cost minimum path among a plurality of characteristic parameters of the distance mapping matrix according to a dynamic rule algorithm;

calculating the accumulated cost of the corresponding characteristic parameters according to the cost minimum path;

determining whether the abnormal data corresponding to the characteristic parameters are matched with the corresponding abnormal sequence templates according to the accumulated cost;

and determining the fault type of the network fault corresponding to the abnormal data according to the matched abnormal sequence template under the condition that the abnormal data corresponding to the characteristic parameters are matched with the corresponding abnormal sequence template.

Optionally, the fault diagnosis method, wherein determining, according to the accumulated cost, whether the abnormal data corresponding to the feature parameter matches with the corresponding abnormal sequence template includes:

judging whether the accumulated cost is smaller than a preset value or not;

and under the condition that the accumulated cost is smaller than the preset value, determining that the abnormal data corresponding to the characteristic parameters are matched with the corresponding abnormal sequence templates.

Optionally, the fault diagnosis method, wherein the method further includes:

Generating model update information according to the fault analysis data and/or abnormal data obtained by detecting abnormal states of the real-time network data;

sending the model update information to the terminal equipment; the model updating information comprises first model updating information and/or second model updating information; the first model updating information is used for indicating the terminal equipment to update a first fault analysis model for carrying out fault analysis, and the second model updating information is used for indicating the terminal equipment to update a first fault prediction model for carrying out fault prediction.

The embodiment of the invention also provides a terminal device, which comprises a processor and a transceiver, wherein:

the processor is used for collecting operation interaction data of the terminal equipment under the condition of network connection failure; and

the transceiver is used for reporting the fault analysis data to a network platform after the network is recovered to be normal.

The embodiment of the invention also provides a network platform, which comprises a transceiver and a processor, wherein:

the transceiver is used for acquiring fault analysis data reported to the network platform by the terminal equipment; the fault analysis data are obtained by carrying out fault analysis on the terminal equipment according to the collected operation interaction data when network connection faults are generated;

The processor is used for establishing an abnormal sequence template of the terminal equipment according to the fault analysis data; when the network connection fault of the terminal equipment is monitored, detecting the abnormal state of the real-time network data to obtain abnormal data; and matching the abnormal data with the abnormal sequence template, and judging the fault type of the network connection fault.

The embodiment of the invention also provides a fault diagnosis device, which is applied to the terminal equipment and comprises:

the data acquisition module is used for acquiring operation interaction data of the terminal equipment under the condition of network connection failure;

the analysis module is used for carrying out fault analysis according to the operation interaction data to obtain fault analysis data;

and the reporting module is used for reporting the fault analysis data to the network platform after the network is recovered to be normal.

The embodiment of the invention also provides a fault diagnosis device, which is applied to the network platform and comprises:

the information acquisition module is used for acquiring fault analysis data reported to the network platform by the terminal equipment; the fault analysis data are obtained by carrying out fault analysis on the terminal equipment according to the collected operation interaction data when network connection faults are generated;

The template establishing module is used for establishing an abnormal sequence template of the terminal equipment according to the fault analysis data;

the detection module is used for detecting the abnormal state of the real-time network data when the network connection fault of the terminal equipment is monitored, so as to obtain abnormal data;

and the matching module is used for matching the abnormal data with the abnormal sequence template and judging the fault type of the network connection fault.

The embodiment of the invention also provides a terminal device, which comprises: a processor, a memory, and a program stored on the memory and executable on the processor, which when executed by the processor implements the fault diagnosis method as claimed in any one of the above.

The embodiment of the invention also provides a network platform, which comprises: a processor, a memory, and a program stored on the memory and executable on the processor, which when executed by the processor implements the fault diagnosis method as claimed in any one of the above.

The embodiment of the invention also provides a readable storage medium, wherein the readable storage medium stores a program, and the program realizes the steps in the fault diagnosis method according to any one of the above when being executed by a processor.

At least one of the above technical solutions of the invention has the following beneficial effects:

by adopting the fault diagnosis method provided by the embodiment of the invention, when the terminal equipment generates network connection faults, the terminal equipment collects operation interaction data and performs fault analysis to obtain fault analysis data, after the network is recovered to be normal, the fault analysis data is reported to the network platform, so that the network platform can establish an abnormal sequence template by using the fault analysis data, after the network fault detection of the terminal is performed, the detected abnormal data can be matched with the abnormal sequence template to determine the cause of the network fault, and the effect of the network fault is determined by the cooperative analysis of the terminal equipment and the network platform, thereby reducing the human analysis cost and the operation maintenance cost.

Drawings

FIG. 1 is a flow chart of a fault diagnosis method according to an embodiment of the invention;

FIG. 2 is a flow chart of a fault diagnosis method according to another embodiment of the invention;

FIG. 3 is a schematic overall flow chart of a fault diagnosis method according to an embodiment of the present invention;

fig. 4 is a schematic structural diagram of a terminal device according to an embodiment of the present invention;

fig. 5 is a schematic structural diagram of a network platform according to an embodiment of the present invention;

FIG. 6 is a schematic diagram of a fault diagnosis apparatus according to an embodiment of the present invention;

fig. 7 is a schematic structural diagram of a fault diagnosis apparatus according to another embodiment of the present invention.

Detailed Description

In order to make the technical problems, technical solutions and advantages to be solved more apparent, the following detailed description will be given with reference to the accompanying drawings and specific embodiments.

In order to solve the problems that in the prior art, when equipment fails and is abnormally dropped, a fault needs to be manually checked, so that the equipment failure problem cannot be timely and effectively solved, and a great amount of labor cost and time cost are consumed, the embodiment of the invention provides a fault diagnosis method, which is characterized in that data acquisition and fault detection are respectively carried out by terminal equipment and a network platform, so that the network failure cause is obtained by end-edge collaborative analysis, the network failure can be effectively solved, and the labor cost and the operation and maintenance cost are reduced.

One embodiment of the present invention provides a fault diagnosis method, which is executed by a terminal device, as shown in fig. 1, and includes:

s110, under the condition of network connection failure, collecting operation interaction data of terminal equipment;

s120, performing fault analysis according to the operation interaction data to obtain fault analysis data;

And S130, after the network is recovered to be normal, reporting the fault analysis data to a network platform.

By adopting the fault diagnosis method provided by the embodiment of the invention, when the terminal equipment generates network connection faults, the terminal equipment collects operation interaction data and performs fault analysis to obtain fault analysis data, after the network is recovered to be normal, the fault analysis data is reported to the network platform, so that the network platform can establish an abnormal sequence template by using the fault analysis data, after the network fault detection of the terminal is performed, the detected abnormal data can be matched with the abnormal sequence template to determine the cause of the network fault, the effect of the network fault is determined by the cooperative analysis of the terminal equipment and the network platform, the guarantee is provided for effectively solving the network fault, and the labor cost and the operation and maintenance cost are reduced.

In one embodiment, optionally, the fault analysis data includes a correspondence between a fault diagnosis result of performing fault analysis according to the operation interaction data and the operation interaction data.

In an embodiment of the fault diagnosis method of the present invention, the fault diagnosis method further includes:

and when the terminal equipment detects the first network access failure or detects that the networking state generates disconnection, determining the network connection failure.

By adopting the embodiment, under the condition that the first network access fails or the network connection state is detected to generate the disconnection, the terminal equipment collects operation interaction data and performs fault analysis; and after the network of the terminal equipment is restored to be normal, namely the network is successfully connected, reporting fault indication information to a network platform.

In one embodiment, the running interaction data includes AT instructions and/or module interaction logs.

By adopting the embodiment, under the condition of network connection faults, the terminal equipment performs fault analysis by acquiring AT instructions and/or module interaction logs, such as SIM card fault analysis, module fault analysis, network fault analysis, authentication problem analysis and invalid message problem analysis, and judges abnormal points causing faults to obtain final fault analysis data.

Optionally, in step S120, performing fault analysis according to the operation interaction data to obtain fault analysis data, including:

and performing fault analysis of one or more of the following according to the operation interaction data to obtain fault analysis data:

analyzing faults of the SIM card;

analyzing a module failure;

analyzing network coverage problems;

analyzing network connection problems;

analyzing authentication problems;

Invalid message problem analysis.

By adopting the embodiment, the AT command and/or the module interaction log are utilized to respectively analyze one or more faults, so that the network faults can be comprehensively analyzed, and an accurate fault analysis result can be obtained.

Optionally, performing fault analysis according to the operation interaction data, and after obtaining fault analysis data, the method further includes:

Specifically, after fault analysis is performed according to the operation interaction data to obtain fault analysis data, different fault types of network connection faults indicated by the fault analysis data are prompted through different colors and/or flashing modes of the gateway indicator lamps, so that the fault type generating the network faults can be explicitly prompted.

For example, in step S120, fault analysis is performed according to the operation interaction data, so as to obtain fault analysis data, which may include one or more of the following analyses:

SIM card failure analysis: and detecting the related error codes of the SIM card through the AT instruction, and if the error codes of the SIM failure, the SIM busy, the SIM wrong and the like are detected, detecting the failure of the SIM card. In one embodiment, when a SIM card failure is detected, a gateway indicator lamp may be continuously turned on to indicate red light;

And (3) module fault analysis: detecting interaction abnormality through an AT instruction, and when abnormality occurs in a heart beat, the terminal equipment can attempt to repair through restarting, so that the AT interaction abnormality is detected, and judging module abnormality or module failure; the abnormal problem of the module can be prompted by flashing red light through the gateway indicator lamp;

network coverage failure analysis: the reference signal received power (Reference Signal Receiving Power, RSRP) value is obtained by the AT command, and if the RSRP is less than a preset value, such as less than 105dbm, then a coverage anomaly may be determined. Wherein, the network coverage abnormality can indicate a similar continuous bright yellow prompt through the gateway;

authentication problem analysis: and detecting the related error code of UE authentication by an AT instruction, and if an Illegal user Ilegal UE is detected, and the user identity cannot be identified by a network UE identity cannot be derived by the network, and the error code such as Implicitly deregistered is implicitly logged off, confirming that the authentication problem is detected. The terminal authentication problem can be prompted by flashing yellow light through the gateway indicator lamp;

network connection problem analysis: acquiring relevant error codes of problems such as network Congestion, insufficient resources, faults and the like through AT instructions, if Congestion Congestion is detected, confirming that network connection problems are detected when special slices and DNN have insufficient resources Insufficient resources for specific slice and DNN and PDU session reaches maximum limit Maximum number of PDU sessions reached and other error codes; the network connection problem can be continuously indicated by the blue light through the gateway indicator lamp;

Invalid message problem analysis: the AT instruction is used for acquiring error codes related to problems such as protocol incompatibility, unrecognizable semantics, incorrect and the like, for example, when the error codes such as semantic error message Semantically incorrect message, conditional IE error Conditional IE error, inconsistent message type and protocol state Message type not compatible with protocol state and the like are detected, the invalid message problem is confirmed to be detected. Alternatively, the invalid message problem may be prompted by a gateway indicator flashing blue light.

By adopting the implementation process, the AT command can be utilized to sequentially analyze the faults of the user identity recognition module (Subscriber Identity Module, SIM) card and the module faults, further, the AT command can be utilized to collect error codes, analysis and judgment of network fault causes can be carried out according to the collected error codes, and prompt and alarm display can be carried out in time by adopting a corresponding flashing mode and/or color through the gateway indicator lamp according to different determined types of network faults.

In the embodiment of the present invention, optionally, in step S120, fault analysis is performed according to the operation interaction data, so as to obtain fault analysis data, including:

performing fault analysis on the operation interaction data by adopting a first fault analysis model to obtain fault analysis data;

Wherein the method further comprises:

acquiring first model updating information sent by the network platform;

and updating the first fault analysis model according to the first model updating information.

According to the embodiment, the network platform is used for matching the detected abnormal data with the abnormal sequence template, the accuracy of the fault analysis model adopted by the terminal equipment and the network platform for fault analysis is judged according to the verification result, the fault analysis model is updated in real time, and the updated fault analysis model is sent to the terminal equipment so as to ensure the accuracy of fault detection by the fault analysis model adopted by the terminal equipment.

It should be noted that, when the terminal device detects a first network access failure or detects that a network connection state generates a disconnection, the terminal device may determine a network failure by adopting the above manner.

In an embodiment of the fault diagnosis method of the present invention, optionally, the fault diagnosis method further includes:

and reporting alarm information to a network platform when the prediction result indicates that network faults can occur. By adopting the embodiment, the network fault is predicted in advance under the condition that the network connection is normal, and the warning information is timely reported to the network platform, so that the network platform can timely perform network processing to prevent the occurrence of the network fault.

In one embodiment, optionally, performing prediction analysis according to operation interaction data of the terminal device to obtain a prediction result of whether a network failure will occur within a first preset duration, where the prediction result includes:

performing predictive analysis on the operation interaction data by adopting a first fault prediction model to obtain a prediction result of whether network faults can be generated within a first preset duration;

the method further comprises the steps of:

obtaining second model updating information sent by the network platform;

and updating the first fault prediction model according to the second model updating information.

By adopting the implementation mode, the network platform can construct a fault prediction model for respectively carrying out fault prediction on the terminal equipment and the network platform, and the fault prediction model is updated at regular time so as to ensure the accuracy of the fault prediction model on fault prediction.

Optionally, the method further comprises:

and saving operation interaction data and/or the fault analysis data.

When the data storage quantity is large, the operation interaction data and/or the fault analysis data 30 s-1 min before the network fault can be stored, and the stored data is reported to the network platform after the network is recovered to be normal.

Another implementation manner of the embodiment of the present invention further provides a fault diagnosis method, which is executed by a network platform, as shown in fig. 2, and includes:

s210, acquiring fault analysis data reported to a network platform by terminal equipment; the fault analysis data are obtained by carrying out fault analysis on the terminal equipment according to the collected operation interaction data when network connection faults are generated;

s220, establishing an abnormal sequence template of the terminal equipment according to the fault analysis data;

s230, when the network connection fault of the terminal equipment is monitored, detecting an abnormal state of real-time network data to obtain abnormal data;

and S240, matching the abnormal data with the abnormal sequence template, and judging the fault type of the network connection fault.

According to the fault diagnosis method, the network platform can establish the abnormal sequence template according to the fault analysis data reported by the terminal equipment, the detected abnormal data can be matched with the abnormal sequence template after the network fault of the terminal is detected, the network fault cause is determined, the effect that the terminal equipment and the network platform cooperatively analyze and determine the network fault is realized, the network fault can be effectively solved, and the labor cost and the operation and maintenance cost are reduced.

Optionally, in one embodiment, the fault analysis data includes a correspondence between a fault diagnosis result and operation interaction data, where the correspondence is used to explicitly indicate a correspondence between a fault type of the terminal device and the operation interaction data, and according to the correspondence, the network platform may directly establish an abnormal sequence template of the terminal device, where the established abnormal sequence template includes multiple fault types and operation interaction data corresponding to each fault type respectively. In this way, the network platform performs anomaly detection on real-time network data through anomaly detection calculation by utilizing the established anomaly sequence template, and determines the corresponding fault type by matching the detected anomaly data with operation interaction data respectively corresponding to each fault type in the anomaly sequence template so as to perform fault detection by combining fault analysis data of terminal equipment at the network platform side, determine the fault type, reduce misjudgment and ensure detection accuracy. Specifically, the network platform analyzes real-time network data, detects abnormal data, matches the abnormal data obtained by the detection result with an abnormal sequence template, obtains the fault type of the network fault according to the matching result, and outputs the reason of the network abnormality.

Optionally, in step S220, an abnormal sequence template of the terminal device is established according to the fault analysis data, where the established abnormal sequence template includes multiple fault types and operation interaction data corresponding to each fault type respectively.

Optionally, the real-time network data includes SIM card data of the terminal device obtained by a SIM card management platform, and/or network state data of a second preset duration before the network failure collected by the network platform. When the network fault of the terminal equipment is monitored, the network platform detects abnormal states by utilizing real-time network data comprising SIM card data and/or network state data of the terminal equipment, and abnormal data is obtained.

When the network failure occurs in the terminal equipment due to network disconnection, the network platform cannot receive information from the terminal equipment, and the abnormal data are judged by analyzing the conditions of the SIM card, the network running state, the performance, the network quality, the flow, the terminal equipment and the like according to the network state data in the second preset time period before the network failure occurs.

Specifically, abnormal state detection is performed on real-time network data, including one or more of the following:

Detecting abnormal states of the SIM card data; optionally, the SIM card data is used for detecting a card balance, a card status, a card on-off status, a card separation status, and the like; the SIM card data can be acquired through the SIM card management platform;

detecting abnormal state of network quality index data in the network state data; optionally, the network quality index data includes RSRP, RSRQ, SINR, RSSI, and the like;

detecting abnormal state of the flow index data in the network state data; optionally, the traffic index data includes uplink traffic, downlink traffic, video traffic, network traffic, and the like;

detecting an abnormal state of the equipment index data in the network state data; optionally, the device index data includes CPU usage, flash usage, memory usage, temperature, and the like.

Optionally, besides the SIM card data, network quality index data, flow index data and equipment index data are acquired by the network management platform according to the historical data acquisition.

When the networking disconnection condition occurs in the terminal equipment, the network platform obtains SIM card data through connecting the SIM card management platform to judge whether abnormal information exists in the SIM card, if so, an alarm is given, if the SIM card is normal, abnormal detection is carried out according to historical data before network disconnection, whether abnormal points exist in the network quality, the flow and the equipment are judged, and besides the abnormal condition judgment of disconnection fault, the method can also be used for daily fault monitoring.

In one embodiment of the present invention, an abnormal state detection algorithm of key performance indicators (Key Performance Indicator, KPIs) of an independent forest may be used to detect abnormal states of real-time network data, so as to obtain abnormal data.

Optionally, in step S230, when the network failure of the terminal device is monitored, abnormal state detection is performed on real-time network data to obtain abnormal data, including:

In one embodiment, preprocessing the real-time network data to obtain processing data of the real-time network data includes:

sequencing the real-time network data according to time information of the real-time network data and with a third preset duration as a period to obtain a time sequence of the real-time network data, wherein the time sequence data (such as traffic data and/or RSRP data) in the real-time network data;

Optionally, setting a third preset duration as N, ordering the plurality of real-time network data with the N duration as a period according to the time information of the plurality of real-time network data, and obtaining a time sequence of the plurality of real-time network data, where { x } ₁ ,x ₂ ,...x _n }；

And carrying out standardized processing on each time sequence data in the time sequence to obtain the processing data of the real-time network data.

Alternatively, the time series data in the time series { x1, x2,..xn } may be normalized in the following manner:

wherein x is _i Is one time sequence data in the time sequence, mu _x Is the mean value of time sequence data; delta _x Is the variance.

In another embodiment, preprocessing the real-time network data to obtain processing data of the real-time network data includes:

and performing one-hot onehot coding processing on discrete data (such as card state data and/or registration state data) in the real-time network data to obtain processing data of the real-time network data.

Optionally, for time sequence data in the real-time network data, extracting features of the processing data to obtain a feature matrix of the processing data, including:

In one embodiment, the calculating the processing data by using different mean value calculating methods to obtain a predicted value corresponding to each mean value calculating method includes:

analyzing the processing data of the time sequence { x1, x2,. Xn } by using algorithms such as a Difference algorithm, a moving average algorithm (moving average), a weighted moving average algorithm (weighted moving average), an exponential moving average (ewma), a holter-temperature (Holt-windows) algorithm, a differential autoregressive moving average model (Auto Regression Integreate Moving Average, ARIMA) and the like respectively to obtain a predicted value of the processing data of the time sequence corresponding to each mean algorithm, for example, p;

further, according to each processing dataDifference |x between each of the prediction values _i -p _i A feature matrix of the processing data of the sequence { x1, x2,..xn } is generated.

Optionally, for discrete data in the real-time network data, extracting features of the processing data to obtain a feature matrix of the processing data, including:

evidence weight (weight of Evidence, woe) coding and information value (information value, IV) calculation are respectively carried out on the processing data, so that a characteristic matrix of the processing data is obtained.

By adopting the embodiment, after the feature matrix of the real-time network data is obtained, the feature matrix can be analyzed in the following manner to judge whether the data corresponding to the corresponding features of the feature matrix is abnormal data or not:

and carrying out abnormality judgment on each detection sample data, and judging whether the corresponding detection sample data is abnormal data or not. For example, by a binary tree construction method, obtaining a plurality of detection sample data of the feature matrix includes:

extracting features in the feature matrix, and constructing a plurality of binary trees; wherein the number of binary numbers constructed depends on the number of features.

When constructing the binary tree, one feature in the feature matrix can be randomly selected as an initial node, a value is randomly selected between the maximum value and the minimum value of the feature, data smaller than the value in the detection sample data corresponding to the feature is marked to the left branch, and data larger than or equal to the value is marked to the right branch. Then, in the left and right branch data, the above steps are repeated until the following condition is satisfied:

The data is not subdivided, i.e.: only one piece of data is contained, or all the data are the same; and/or

The binary tree reaches a defined maximum depth.

Based on the above manner, by constructing a plurality of detection sample data obtained by a binary tree method, performing anomaly judgment on each detection sample data, and judging whether the corresponding detection sample data is anomaly data, including:

Alternatively, the anomaly score for each test sample data may be calculated in the following manner:

calculating the path length h (xi) of the detection sample data xi corresponding to the characteristic in the binary tree itrage: h (xi) =e+c (t.size);

where e is the number of edges that x passes from the root node to the leaf node of the itrate, C (t.size) can be considered a correction value representing the average path length in a binary tree constructed from t.size pieces of sample data. Wherein C (n) =2h (n-1) -2 (n-1)/n; h (n-1) =ln (n-1) +0.5772156649; wherein H (n-1) is a harmonic function in outlier detection.

Calculating an anomaly Score (xi) of the detection sample data xi by integrating the results of the plurality of trees:

when the anomaly score satisfies the following condition, the corresponding data xi is judged to be an anomaly value:

Score(xi)>θ；

wherein θ is a preset determination value.

In the embodiment of the present invention, optionally, matching the abnormal data with the abnormal sequence template, and judging the fault type of the network fault includes:

Determining whether the abnormal data corresponding to the characteristic parameters is matched with the corresponding abnormal sequence templates according to the accumulated cost comprises the following steps:

Judging whether the accumulated cost is smaller than a preset value or not;

For example, an anomaly sequence template is set to be represented by Q { Q1, Q2,..qn }, anomaly data corresponding to the feature parameters is represented by C { C1, C2,..cm }, and a many-to-many distance mapping matrix ED (Q, C) is constructed with a size of n×m.

And calculating the minimum-cost path which satisfies monotonicity and continuity of the characteristic parameters from (0, 0) to (n, m) in the distance mapping matrix ED (Q, C) by using a dynamic programming algorithm. Wherein q _i One abnormal sequence template; c _i Is one of the abnormal data;

calculating accumulated cost l (i, j) according to the path:

l(i,j)＝d(q _i ,c _i )+min{l(i-1,j),l(i,j-1),l(i-1,j-1)}；

and under the condition that the accumulated cost is smaller than a preset value alpha, namely l < alpha, determining that the abnormal data is matched with the corresponding abnormal sequence template, and determining the fault type of the network fault corresponding to the abnormal data according to the matched abnormal sequence template.

Wherein, since the abnormal sequence template can record the abnormal data detected by the terminal equipment and the corresponding fault types, the fault types of network faults generated by the abnormal data can be determined by matching the abnormal data detected by the platform equipment with the abnormal sequence template.

By adopting the fault diagnosis method of the embodiment, the terminal equipment can perform fault analysis to obtain a fault analysis result after generating network faults, and report fault analysis data to the network platform after the network is recovered to be normal, so that the network platform can establish an abnormal sequence template according to the actual dropped abnormal data and the fault result detected by the terminal equipment, and match the detected abnormal data with the abnormal sequence template after the network platform side performs abnormal detection through an abnormal detection algorithm, thereby reducing misjudgment and ensuring detection accuracy.

In an embodiment of the present invention, optionally, the method further includes:

after receiving the fault analysis data sent by the terminal equipment, the network platform builds a first fault prediction model and a second fault prediction model, sends the first fault prediction model to the terminal equipment, enables the terminal equipment to conduct prediction analysis on operation interaction data by using the first fault prediction model to obtain a prediction result of whether network faults occur within a first preset duration, and enables the network platform to conduct fault prediction in advance by using the real-time network data under the condition that network connection of the terminal equipment is normal by using the second fault prediction model, so that occurrence of network faults is avoided.

Further, the method further comprises:

Optionally, the model update information may further include third model update information that is updated by the network platform for fault analysis and/or fourth model update information that is updated by the network platform for fault prediction, for respectively updating a model of the network platform for fault analysis and a model for fault prediction.

By adopting the embodiment, the network platform can update the first fault analysis model for carrying out fault analysis on the terminal equipment and/or the first fault prediction model for carrying out fault prediction in real time so as to ensure the accuracy of fault analysis and fault prediction of the terminal.

By adopting the fault diagnosis method provided by the embodiment of the invention, the terminal equipment and the network platform are used for respectively carrying out data acquisition and fault detection, so that the network fault cause is obtained by end-edge collaborative analysis, the network fault can be effectively solved, and the labor cost and the operation and maintenance cost are reduced.

Fig. 3 is a schematic overall flow chart of a fault diagnosis method according to an embodiment of the invention. In case of failure of the terminal device to communicate with the network platform, i.e. in case of network failure of the terminal device, the terminal device performs the following implementation steps:

s3001, collecting operation interaction data;

s3002, performing fault analysis on the operation interaction data by adopting a first fault analysis model;

s3003, performing flickering prompt in a mode corresponding to the fault type according to the fault type obtained by fault analysis;

s3004, after the network is recovered to be normal, reporting fault analysis data to the network platform.

While the terminal device performs the above implementation steps, the network platform performs the following implementation steps:

s3011, SIM card data obtained by the SIM card management platform is used for detecting abnormal states of the SIM card data;

s3012, detecting network quality, flow index, equipment index and the like before the disconnection according to the network state data before the disconnection;

S3013, matching the detection result with an abnormal sequence template, and judging the fault type of the network fault;

s3014, outputting a detection result of the network platform;

s3015, updating the abnormality detection models of the terminal equipment and the network platform, sending first model updating information to the terminal equipment, and updating a first fault analysis model of the terminal equipment;

s3016, constructing an anomaly prediction database, updating anomaly prediction models of the terminal equipment and the network platform, sending second model updating information to the terminal equipment, and updating a first failure preset model of the terminal equipment.

Under the condition that the terminal equipment and the network platform normally communicate, the terminal equipment executes the following implementation steps:

s3021, performing predictive analysis by adopting a first fault prediction model to obtain a prediction result of whether network faults can occur within a first preset duration;

s3022, reporting alarm information to a network platform under the condition that network faults are predicted to occur;

under the condition that the terminal equipment and the network platform normally communicate, the network platform executes the following implementation steps:

s3031, performing predictive analysis by adopting a second fault prediction model to obtain a prediction result of whether network faults can be generated within a fourth preset duration;

S3032, under the condition that the network fault is predicted to occur, the terminal equipment is controlled according to the fault reason so as to prevent the terminal equipment from being disconnected.

It should be noted that, the implementation process may be executed under the condition that the terminal device is connected with the network, and when the device fails to access the network for the first time, the terminal device may mainly collect data, judge the fault type, and indicate through the indicator light. And uploading the acquired data and the diagnosis result to a network platform for storage after the terminal equipment is successfully networked, and verifying the fault analysis by the network platform.

According to the implementation process, when the terminal equipment generates network faults, the terminal equipment collects data and performs preliminary diagnosis on the data, the collected data is reported to the network platform, and an algorithm database can be expanded according to the data reported by the terminal equipment; the network platform can carry out auxiliary analysis according to the data reported by the terminal equipment, and updates a fault analysis model and a fault prediction model of the terminal equipment and the network platform according to the operation and maintenance information.

By adopting the fault diagnosis method provided by the embodiment of the invention, the problem of abnormal networking disconnection of equipment which is most urgent to be processed in the industrial network at the current stage is respectively analyzed from the terminal side and the platform side, so that on-site operators can find and check faults in time, the labor cost and time cost of on-site operation and maintenance are reduced, and the fault repair time is shortened.

The embodiment of the present invention further provides a terminal device, as shown in fig. 4, the terminal device 400 includes a processor 410 and a transceiver 420, where:

the processor 410 is configured to collect operation interaction data of the terminal device in case of a network connection failure; and

the transceiver 420 is configured to report the fault analysis data to a network platform after the network is restored to normal.

Optionally, the terminal device, wherein the processor 410 is further configured to:

the transceiver 420 is further configured to report an alert to the network platform when the prediction indicates that a network failure may occur.

Optionally, the terminal device, wherein the running interaction data includes AT instructions and/or module interaction logs.

Optionally, the terminal device, wherein the processor 410 performs fault analysis according to the operation interaction data to obtain fault analysis data, including:

analyzing the problem of the SIM card;

analyzing module problems;

analyzing network coverage problems;

analyzing network connection problems;

analyzing authentication problems;

invalid message problem analysis.

wherein the processor 410 is further configured to:

acquiring first model updating information sent by the network platform;

Optionally, the terminal device, wherein the processor 410 performs prediction analysis according to operation interaction data of the terminal device to obtain a prediction result of whether a network failure will occur within a first preset duration, including:

The processor 410 is further configured to:

obtaining second model updating information sent by the network platform;

Optionally, the terminal device performs fault analysis according to the operation interaction data, and after obtaining fault analysis data, the processor 410 is further configured to:

The embodiment of the present invention further provides a network platform, as shown in fig. 5, the network platform 500 includes a transceiver 510 and a processor 520, wherein:

the transceiver 510 is configured to obtain fault analysis data reported by a terminal device to a network platform; the fault analysis data are obtained by carrying out fault analysis on the terminal equipment according to the collected operation interaction data when network connection faults are generated;

the processor 520 is configured to establish an abnormal sequence template of the terminal device according to the fault analysis data; when the network connection fault of the terminal equipment is monitored, detecting the abnormal state of the real-time network data to obtain abnormal data; and matching the abnormal data with the abnormal sequence template, and judging the fault type of the network connection fault.

Optionally, the network platform includes the real-time network data including SIM card data of the terminal device obtained by a SIM card management platform and/or network status data collected by the network platform within a second preset duration before a network connection failure occurs.

Optionally, the network platform, wherein the processor 520 performs abnormal state detection on the real-time network data, including one or more of the following:

detecting abnormal states of the SIM card data;

detecting abnormal state of the flow index data in the network state data;

Optionally, the network platform, wherein the processor 520 performs abnormal state detection on the real-time network data to obtain abnormal data, including:

Optionally, the network platform, wherein the processor 520 performs an anomaly determination on each detection sample data, and determining whether the corresponding detection sample data is an anomaly data includes:

Optionally, the network platform, wherein the processor 520 pre-processes the real-time network data to obtain processing data of the real-time network data, includes:

sequencing the real-time network data according to time information of the real-time network data and with a third preset duration as a period, so as to obtain a time sequence of the real-time network data;

and performing onehot coding processing on discrete data in the real-time network data to obtain processing data of the real-time network data.

Optionally, the network platform, wherein the processor 520 performs feature extraction on the processing data to obtain a feature matrix of the processing data, and includes:

and woe coding and IV value calculation are respectively carried out on the processing data, so that a characteristic matrix of the processing data is obtained.

Optionally, the network platform, wherein the processor 520 matches the anomaly data with the anomaly sequence template, and determining the fault type of the network fault includes:

Optionally, the network platform, wherein the determining, by the processor 520, whether the abnormal data corresponding to the feature parameter matches the corresponding abnormal sequence template according to the accumulated cost includes:

judging whether the accumulated cost is smaller than a preset value or not;

Optionally, the network platform, wherein the processor 520 is further configured to:

the transceiver 510 is further configured to send the model update information to the terminal device; the model updating information comprises first model updating information and/or second model updating information; the first model updating information is used for indicating the terminal equipment to update a first fault analysis model for carrying out fault analysis, and the second model updating information is used for indicating the terminal equipment to update a first fault prediction model for carrying out fault prediction.

The embodiment of the invention also provides a fault diagnosis device, which is applied to the terminal equipment, as shown in fig. 6, and comprises:

the data acquisition module 610 is configured to acquire operation interaction data of the terminal device in case of a network connection failure;

the analysis module 620 is configured to perform fault analysis according to the operation interaction data, so as to obtain fault analysis data;

and the reporting module 630 is configured to report the fault analysis data to a network platform after the network is restored to be normal.

Optionally, the fault diagnosis apparatus, wherein the analysis module 620 is further configured to:

the reporting module 630 is further configured to report alarm information to the network platform when the prediction result indicates that a network failure will occur.

Optionally, the fault diagnosis device, wherein the running interaction data includes AT instructions and/or module interaction logs.

Optionally, in the fault diagnosis apparatus, the analyzing module 620 performs fault analysis according to the operation interaction data to obtain fault analysis data, including:

analyzing the problem of the SIM card;

analyzing module problems;

analyzing network coverage problems;

analyzing network connection problems;

analyzing authentication problems;

invalid message problem analysis.

Wherein the analysis module 620 is further configured to:

acquiring first model updating information sent by the network platform;

Optionally, in the fault diagnosis apparatus, the analyzing module 620 performs prediction analysis according to operation interaction data of the terminal device to obtain a prediction result of whether a network fault will occur within a first preset duration, including:

the analysis module 620 is further configured to:

obtaining second model updating information sent by the network platform;

Optionally, in the fault diagnosis apparatus, after performing fault analysis according to the operation interaction data to obtain fault analysis data, the analysis module 620 is further configured to:

The embodiment of the invention also provides a fault diagnosis device, which is applied to a network platform, as shown in fig. 7, and comprises:

the information acquisition module 710 is configured to acquire fault analysis data reported to the network platform by the terminal device; the fault analysis data are obtained by carrying out fault analysis on the terminal equipment according to the collected operation interaction data when network connection faults are generated;

the template establishing module 720 is configured to establish an abnormal sequence template of the terminal device according to the fault analysis data;

the detection module 730 is configured to detect an abnormal state of real-time network data when a network connection failure of the terminal device is detected, so as to obtain abnormal data;

and the matching module 740 is configured to match the abnormal data with the abnormal sequence template, and determine a fault type of the network connection fault.

Optionally, the fault diagnosis device, wherein the real-time network data includes SIM card data of the terminal device obtained by a SIM card management platform, and/or network state data collected by the network platform within a second preset duration before a network connection fault occurs.

Optionally, the fault diagnosis apparatus, wherein the detection module 730 performs abnormal state detection on the real-time network data, including one or more of the following:

Detecting abnormal states of the SIM card data;

detecting abnormal state of the flow index data in the network state data;

Optionally, in the fault diagnosis apparatus, the detecting module 730 detects an abnormal state of the real-time network data to obtain abnormal data, including:

Optionally, in the fault diagnosis apparatus, the detecting module 730 performs an anomaly determination on each detection sample data, and determines whether the corresponding detection sample data is an anomaly data, including:

Optionally, the fault diagnosis apparatus, wherein the detecting module 730 performs preprocessing on the real-time network data to obtain processing data of the real-time network data, includes:

Optionally, in the fault diagnosis apparatus, the detecting module 730 performs feature extraction on the processing data to obtain a feature matrix of the processing data, including:

Optionally, the fault diagnosis apparatus, wherein the matching module 740 matches the abnormal data with the abnormal sequence template, and determines a fault type of the network fault, including:

Optionally, in the fault diagnosis apparatus, the matching module 740 determines, according to the accumulated cost, whether the abnormal data corresponding to the feature parameter matches with a corresponding abnormal sequence template, including:

judging whether the accumulated cost is smaller than a preset value or not;

Optionally, the fault diagnosis device, wherein the device further includes:

the model generating module 750 is configured to generate model update information according to the fault analysis data and/or abnormal data obtained by performing abnormal state detection on real-time network data;

a sending module 760, configured to send the model update information to the terminal device; the model updating information comprises first model updating information and/or second model updating information; the first model updating information is used for indicating the terminal equipment to update a first fault analysis model for carrying out fault analysis, and the second model updating information is used for indicating the terminal equipment to update a first fault prediction model for carrying out fault prediction.

In the embodiment of the present invention, the specific implementation process of executing the fault diagnosis method by the processor on the terminal device may refer to the description of the method section, and will not be described in detail herein.

An embodiment of the present invention provides a network platform, which is characterized by comprising: a processor, a memory, and a program stored on the memory and executable on the processor, which when executed by the processor implements the fault diagnosis method as claimed in any one of the above.

In the embodiment of the present invention, the specific implementation process of the fault diagnosis method performed by the processor on the network platform may refer to the description of the method section, which is not described in detail herein.

In addition, a specific embodiment of the present invention also provides a computer-readable storage medium having stored thereon a computer program, wherein the program when executed by a processor implements the steps in the fault diagnosis method as described in any one of the above.

Specifically, the computer readable storage medium is applied to the above terminal device or network platform, and when applied to the terminal device or network platform, the execution steps of the corresponding fault diagnosis method are described in detail above, and are not repeated herein.

In the several embodiments provided in the present application, it should be understood that the disclosed methods and apparatus may be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of the units is merely a logical function division, and there may be additional divisions when actually implemented, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.

In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may be physically included separately, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in hardware plus software functional units.

The integrated units implemented in the form of software functional units described above may be stored in a computer readable storage medium. The software functional unit is stored in a storage medium, and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform part of the steps of the transceiving method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.

While the foregoing is directed to the preferred embodiments of the present invention, it will be appreciated by those skilled in the art that various modifications and changes can be made without departing from the principles of the present invention, and such modifications and changes should also be considered as being within the scope of the present invention.

Claims

1. A fault diagnosis method, characterized by being executed by a terminal device, the method comprising:

2. The fault diagnosis method according to claim 1, characterized in that the method further comprises:

3. The fault diagnosis method according to claim 1 or 2, wherein the running interaction data comprises AT instructions and/or module interaction logs.

4. The fault diagnosis method according to claim 1, wherein performing fault analysis based on the operation interaction data to obtain fault analysis data comprises:

analyzing the problem of the SIM card;

analyzing module problems;

analyzing network coverage problems;

analyzing network connection problems;

analyzing authentication problems;

invalid message problem analysis.

5. The fault diagnosis method according to claim 1, wherein after performing fault analysis based on the operation interaction data to obtain fault analysis data, the method further comprises:

6. A method of fault diagnosis, performed by a network platform, the method comprising:

7. The method according to claim 6, wherein the real-time network data includes SIM card data of the terminal device obtained by a SIM card management platform and/or network status data collected by the network platform within a second preset time period before a network connection failure occurs.

8. The fault diagnosis method according to claim 7, wherein the abnormal state detection is performed on the real-time network data, including one or more of the following:

detecting abnormal states of the SIM card data;

detecting abnormal state of the flow index data in the network state data;

9. The fault diagnosis method according to claim 6, wherein the performing abnormal state detection on the real-time network data to obtain abnormal data comprises:

10. The fault diagnosis method according to claim 9, wherein performing abnormality judgment for each detection sample data to judge whether the corresponding detection sample data is abnormal data, comprises:

11. The fault diagnosis method according to claim 9, wherein performing feature extraction on the processing data to obtain a feature matrix of the processing data, comprises:

12. The fault diagnosis method according to claim 6, wherein matching the anomaly data with the anomaly sequence template, judging the fault type of the network fault, comprises:

13. The fault diagnosis method according to claim 12, wherein determining whether the abnormal data corresponding to the characteristic parameter matches the corresponding abnormal sequence template according to the accumulated cost comprises:

judging whether the accumulated cost is smaller than a preset value or not;

14. The fault diagnosis method according to claim 6, characterized in that the method further comprises:

15. A terminal device comprising a processor and a transceiver, characterized in that:

16. A network platform comprising a transceiver and a processor, characterized by:

17. A fault diagnosis apparatus, characterized by being applied to a terminal device, comprising:

18. A fault diagnosis apparatus for use with a network platform, the apparatus comprising:

19. A terminal device, comprising: a processor, a memory, and a program stored on the memory and executable on the processor, which when executed by the processor, implements the fault diagnosis method as claimed in any one of claims 1 to 5.

20. A network platform, comprising: a processor, a memory, and a program stored on the memory and executable on the processor, which when executed by the processor, implements the fault diagnosis method as claimed in any one of claims 6 to 14.

21. A readable storage medium, characterized in that the readable storage medium has stored thereon a program which, when executed by a processor, realizes the steps in the fault diagnosis method according to any one of claims 1 to 5 or the steps in the fault diagnosis method according to any one of claims 6 to 14.