CN107426750A - A kind of mobile telephone registration failure automatic diagnosis method and system - Google Patents
A kind of mobile telephone registration failure automatic diagnosis method and system Download PDFInfo
- Publication number
- CN107426750A CN107426750A CN201710585025.6A CN201710585025A CN107426750A CN 107426750 A CN107426750 A CN 107426750A CN 201710585025 A CN201710585025 A CN 201710585025A CN 107426750 A CN107426750 A CN 107426750A
- Authority
- CN
- China
- Prior art keywords
- registration
- daily record
- model
- mobile phone
- mobile telephone
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000003745 diagnosis Methods 0.000 title claims abstract description 20
- 238000001514 detection method Methods 0.000 claims description 39
- 238000012549 training Methods 0.000 claims description 14
- 230000000153 supplemental effect Effects 0.000 claims description 12
- 230000002093 peripheral effect Effects 0.000 claims description 6
- 238000006748 scratching Methods 0.000 claims description 4
- 230000002393 scratching effect Effects 0.000 claims description 4
- 238000005516 engineering process Methods 0.000 abstract description 5
- 238000012360 testing method Methods 0.000 description 10
- 238000012545 processing Methods 0.000 description 8
- 238000007418 data mining Methods 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 4
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000013480 data collection Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000012417 linear regression Methods 0.000 description 2
- 238000005065 mining Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000007637 random forest analysis Methods 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 238000013475 authorization Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000001351 cycling effect Effects 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000004321 preservation Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/02—Arrangements for optimising operational condition
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/08—Testing, supervising or monitoring using real traffic
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W60/00—Affiliation to network, e.g. registration; Terminating affiliation with the network, e.g. de-registration
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Telephone Function (AREA)
Abstract
The present invention relates to intelligent measuring technology field, more particularly to a kind of mobile telephone registration failure automatic diagnosis method and system.Registration daily record collection module including registration failure diagnostic module and in mobile phone.Registration daily record collection module can reach the registration log information of the SIM card registered network of mobile phone of the crawl of automation under without environment.Most of all, the crawl of log information is captured at the very start from registration, and stops crawl after register flow path terminates at once, the log information outside the flow of SIM registered networks is largely eliminated.Model analysis is carried out based on the daily record data that the registration daily record collection module mobile phone arrives, simplifies the difficulty of modeling, and substantially increase the accuracy and specific aim for establishing model.
Description
Technical field
The present invention relates to intelligent measuring technology field, more particularly to a kind of mobile telephone registration failure automatic diagnosis method and it is
System.
Background technology
The SIM card of mobile phone often by the substantial amounts of test of artificial tracking, and is based on when registered network goes wrong
The data of test, solved using the professional knowledge of oneself, the cycle for solving problem is longer.Need：
Log (log information) when a large amount of mobile telephone registrations of one, are normal, such as different environment, different SIM cards are different
Registration instrument, different operators, different places, base station etc., carry out the substantial amounts of normal log information of mobile telephone registration.
Log (log information) when a large amount of mobile telephone registrations of two, fail, such as different environment, different SIM cards are different
Registration instrument, different operators, different places, base station etc., situation is more complicated better.
And it is for the mobile phone of log information, most common method：Mobile phone is placed in a certain environment, captured since start
Log, mobile telephone registration success or failure is checked by mobile phone interface.
The B of Authorization Notice No. CN 102223662, the patent of invention in authorized announcement date on October 2nd, 2013 disclose one kind
Testing mobile phone falls the method and system of card.Mobile phone is connected by serial ports with being provided with the computer of card test software：Mobile phone is opened
After machine, by computer fall card test software to mobile phone send control mobile phone cycling switch machine the first AT instruct, computer to
Mobile phone sends the 2nd AT instructions, to obtain network registry information during mobile phone power-on and carry out statistics preservation, then controls mobile phone
Automatic shutdown, and detect whether handset switch machine number reaches predetermined number, when being, computer stops controlling mobile phone to open automatically
Shutdown, and automatically to causing to fall the information of card to carry out statistics guarantor in SIM cards of mobile phones registered network during above-mentioned handset switch machine
Deposit, and automatically generate card test report, automatic some important informations for recording mobile phone power-on registered network process, to engineer
Analysis falls card reason and provides important evidence.But the technical scheme of the patent is only to provide analysis mobile phone for engineer to fall
The data basis of card reason, and really fall the analysis of card reason or can only sentence by the experience and professional knowledge of engineer
It is disconnected, the accuracy rate of different engineer and the time spent in it is all different.
The A of application publication number CN 105227729, the application for a patent for invention in data of publication of application on January 6th, 2016 is in order to solve
It can only be used when mobile phone radio frequency is tested in the prior art and manually restart mobile phone, influenceed testing efficiency and cause human resources
Waste, it is impossible to meet the needs of automatic test.Disclose it is a kind of automatically control radio frequency testing mobile telephone registration networking method and
System.It is connected with mobile phone is established into ADB with intelligent terminal, is connected to mobile phone transmission of the adjustment under shell patterns and flown by ADB
The control instruction of row mode, control mobile phone open and close offline mode, its automatic login network access are realized, so as to realize that mobile phone exists
When carrying out radio frequency testing, can reset automatically login network access.
The implementation of the technical scheme of above two all can not really realize the automatic diagnosis of mobile telephone registration failure.Also,
Different environment, different SIM cards, different registration instruments, different operators, the difference in different places, base station etc.
Will have influence on mobile telephone registration result, and in above-mentioned technical proposal for the data of engineer's analysis test depend on by mobile phone with
Equipment outside mobile phone is attached, and is not easy to the movement of mobile phone, can not obtain perfect test data.
The content of the invention
In order to solve the above-mentioned technical problem the present invention, proposes a kind of mobile telephone registration failure automatic diagnosis method, its feature exists
In, including：
Step S-1, obtain registration daily record during registration failure.
Step S-2, the registration daily record is imported into diagnostic model；
Step S-3, the diagnostic model are analyzed and export failure cause.
Preferably, the method for building up of the diagnostic model includes：
Step S-A, registers log collection, and the registration daily record refers to the log information of mobile telephone registration network；
Step S-B, log integrity is registered, registration daily record is collected into the step S-A and carries out pretreatment acquisition effectively
Daily record data；
Step S-C, diagnostic model are established, and diagnostic model is established based on the effective daily record data obtained in the step S-B.
Preferably, the step S-A includes：
Step S-A-1, when one in the signal intensity of mobile phone, signal quality, peripheral base station quantity changes, make mobile phone
Registered network；
Step S-A-2, when detecting that registration starts key message, start log information crawl；
Step S-A-3, when detecting that registration terminates key message, stop daily record information scratching；
Step S-A-4, preserve registration daily record.
Preferably, the step S-C includes：
Step S-C-1, model training are initial to obtain based on model training is carried out in most of effectively daily record data
Model；
Step S-C-2, Modifying model, based on the initial model effectively described in the remainder amendment of daily record data to obtain
Obtain diagnostic model.
Preferably, the registration daily record include when succeeding in registration succeed in registration daily record and registration failure when registration lose
Lose daily record；In step S-C-1, the significant figure of the registration failure daily record of the valid data and A% of the daily record of succeeding in registration based on A%
According to progress model training.So that aim at ratio shared in all effective daily records of succeeding in registration the day of succeeding in registration for model training
Example is identical with for the ratio shared by the aiming at registration failure day in the effective daily record of all registration failures of model training.A value
It should be greater than 50 and be less than 100, such as 60,70 or 80 etc..
Preferably, the effectively daily record data includes supplemental characteristic, result data；The step S-C-2 includes：
Step S-C-2-1, the initial model is imported into using the supplemental characteristic of the valid data as input；
Step S-C-2-2, the initial model output model result；
Step S-C-2-3, whether the model result is consistent with the result data, and institute is adjusted if inconsistent
The preference pattern and model parameter of initial model are stated, the initial model is not adjusted if consistent.
The present invention also provides a kind of mobile telephone registration failure auto-check system, it is characterised in that：Including in mobile phone
Registration daily record collection module, registration failure diagnostic module；The registration failure diagnostic module can lead input registration daily record
Enter into diagnostic model and export diagnostic result；The input registration daily record refers to the date of registration of mobile telephone registration network failure
Will, the diagnostic model are established based on the registration daily record that the registration daily record collection module is collected.
Preferably, the registration daily record collection module is used for the registration daily record for obtaining the log information of mobile telephone registration network
Placement unit；The log-on message placement unit, start crawl log information when detecting that registration starts key message, and
Stop crawl log information when detecting that registration terminates key message.
Preferably, the registration daily record collection module include the signal strength detection unit for detection signal intensity,
Signal quality detection unit for detection signal quality, the base station number detection unit for detecting peripheral base station quantity, institute
Mobile telephone registration can be triggered by stating signal strength detection unit, the signal quality detection unit and the base station number detection unit
Network.
Preferably, the mobile phone includes at least two Mobile phone cards, institute on mobile phone described in the registration daily record collection module
There is the registration daily record of Mobile phone card.
Brief description of the drawings
Fig. 1 is the flow chart that diagnostic model of the present invention is established.
Fig. 2 is present invention registration log acquisition flow chart one.
Fig. 3 is present invention registration log acquisition flowchart 2.
Embodiment
Specific examples below is only explanation of the invention, and it is not limitation of the present invention, art technology
Personnel can make the modification of no creative contribution to the present embodiment as needed after this specification is read, but as long as
All protected in scope of the presently claimed invention by Patent Law.
Embodiment one
A kind of mobile telephone registration failure auto-check system, including registration failure diagnostic module and the registration in mobile phone
Log collection module.
Registration daily record collection module is mounted in the module in mobile phone, and it includes the letter for being used to detect the signal intensity of mobile phone
Number intensity detection unit, the signal quality detection unit of signal quality for detecting mobile phone, for detecting mobile phone current location
The base station number detection unit of the base station number of surrounding and the registration log collection thread established in interior of mobile phone.
Signal strength detection unit, it can detect the current signal intensity of mobile phone, and can be set in advance according to reference
The threshold range for division signals strength grade put, output can reflect the signal of the current signal strength grade of mobile phone.
Signal quality detection unit, it can be commented the signal quality of mobile phone according to the current network state of mobile phone
Estimate, and export the signal for the current signal quality grade that can reflect mobile phone.
Base station number detection unit, it can detect the base station number around mobile phone current location, and export reflection and be somebody's turn to do
The information of base station number.
Registration log collection thread should possess following function：
A) registration log collection thread can detect crucial Log information when mobile phone starts registration (such as in high-pass platform
Keyword：CM_SERVICE_REQ), it may determine that the SIM card of mobile phone starts to register net according to the crucial Log information thread
Network.The thread only just proceeds by log information collection when detecting keyword CM_SERVICE_REQ.
B) whether successfully registration log collection thread can detect mobile telephone registration crucial Log information (such as high-pass platform
In keyword：CM_SERVICE_CFG), it may determine that the SIM card registered network of mobile phone is according to the crucial Log information thread
No end, and know registering result.The thread stops log information receipts once keyword CM_SERVICE_CFG is detected
Collection.
C) registration log collection thread can pass through signal strength detection unit, signal quality detection unit, base station number
The intensity and the base station number of quality and surrounding of the quantity monitoring current phone reception signal of detection unit monitoring network environment
Information, when finding the signal intensity of the mobile phone signal either base station around the signal quality of mobile phone signal or mobile phone current location
(i.e. when in signal strength detection unit, signal quality detection unit and base station number detection unit when quantity varies widely
The output signal of at least one unit when changing) triggering mobile phone SIM card registered network function so that mobile phone weighs automatically
The new register flow path for starting network, and the daily record in registration process is captured again.
Such as Fig. 2, the registration log collection based on above-mentioned registration daily record collection module includes：
Step S-A-1, when one in the signal intensity of mobile phone, signal quality, peripheral base station quantity changes, make mobile phone
Registered network；
Step S-A-2, when detecting that registration starts key message, start log information crawl；
Step S-A-3, when detecting that registration terminates key message, stop daily record information scratching；
Step S-A-4, preserve registration daily record.
Registration daily record collection module can reach the SIM card registered network of mobile phone of the crawl of automation under without environment
Registration log information.Most of all, the crawl of log information is captured at the very start from registration, and in registration flow
Journey stops crawl at once after terminating, and largely eliminates the log information outside the flow of SIM registered networks.Based on this
The daily record data that registration daily record collection module mobile phone arrives carries out model analysis, simplifies the difficulty of modeling, and substantially increase
Establish the accuracy and specific aim of model.
Registration failure diagnostic module can be independently of mobile phone be used for the outer of fault diagnosis is carried out to mobile phone registration failure
Portion's equipment, its USB interface that can lead to, wifi, serial ports etc. is wired or is wirelessly connected with mobile phone.Can also be same
A program module in mobile phone, or be a core dedicated for accident analysis being similarly installed in mobile phone
Piece.Once failing during the SIM card registered network of mobile phone, then the registration daily record of this mobile telephone registration process is imported
Automated analysis is carried out into registration failure diagnostic module, the reason for orienting mistake.
The diagnostic model of registration failure diagnostic module is that the registration daily record collected based on foregoing registration daily record collection module is built
Vertical.The flow of foundation includes：
Step S-A, registers log collection, and registration daily record refers to the log information of mobile telephone registration network.It is a large amount of to utilize registration
Log collection module largely captures the registration daily record of mobile telephone registration network.Here registration daily record needs to include SIM cards of mobile phones
Registered network successfully registers registration daily record (the registration mistake of daily record (daily record of succeeding in registration) and the failure of SIM cards of mobile phones registered network
Lose daily record).The sample size of registration daily record is bigger, and the model of foundation is more accurate.For example, it is successful to capture collection registered network
Register 1000 parts of daily record sample, 1000 parts of the registration daily record sample of registered network failure.
Step S-B, log integrity is registered, registration daily record is collected into step S-A and carries out the effective daily record of pretreatment acquisition
Data.Railway Project be present in original registration daily record data：Inconsistency, repeatability, Noise, latitude are high.Wherein, due to this
Crawl of the registration daily record collection module of invention to log information is captured at the very start from registration, and in register flow path
Stop crawl after end at once, largely eliminate the log information outside the flow of SIM registered networks, noise content
Than relatively low.But, however it remains inconsistency, it is repeated the problems such as.Pretreatment in step S-B is clear by data comprising needing
Wash, the one or more in data integration, data conversion and hough transformation this several method, original registration daily record is carried out
Processing, to choose suitable attribute as data mining attribute.In the present embodiment, first, to original registration daily record number
According to participle, common segmentation methods have：Participle based on dictionary, the participle based on statistics, the participle based on semanteme, Ke Yigen
Need voluntarily to select according to what model was established.Secondly, duplicate removal processing is carried out to segmenting later data, removes unique attribute, removes
Repeatability, take out negligible field.Again, data regularization processing is carried out to data, it is smaller than initial data more to obtain, but
The mining data collection of data integrity is not destroyed, and the data set can obtain and initial data identical data mining effect.
Step S-C, diagnostic model are established, and diagnostic model is established based on the effective daily record data obtained in step S-B.Specifically
Including step：
Step S-C-1, model training, based on carrying out model training in most of effectively daily record data to obtain introductory die
Type.The 800 parts of daily records of succeeding in registration that will be registered in step S-A in the 1000 parts of daily records of succeeding in registration being collected into of daily record collection module
Effective daily record data and step S-A in 800 parts in the 1000 parts of registration failure daily records being collected into of registration daily record collection module
Effective daily record data of registration failure daily record carries out model training to obtain initial model.
Step S-C-2, Modifying model, based on the remainder amendment initial model of effective daily record data to obtain diagnosis mould
Type.Effective daily record data includes supplemental characteristic and result data, and wherein result data refers to represent hand corresponding to the valid data
The whether successful data of machine register flow path.In step S-C-2, for registering what daily record collection module mobile phone arrived in step S-A
In 1000 parts of daily records of succeeding in registration log collection is registered in the valid data and step S-A of remaining 200 parts of daily records of succeeding in registration
Module cellphone to 1000 parts of registration failure daily records in remaining 200 parts of registration failure daily records valid data, these are effective
Each in data carries out following processing：
Step S-C-2-1, the supplemental characteristic of valid data is imported into initial model as input.
Step S-C-2-2, initial model output model result, initial model are carried out at analysis according to the supplemental characteristic of input
Reason, it is contemplated that whether the mobile telephone registration based on the supplemental characteristic succeeds, and is exported estimated result as model result.
Whether step S-C-2-3, model result and the result data for comparing initial model output are consistent：If inconsistent
Adjust the preference pattern and/or model parameter of initial model；Initial model is not adjusted if consistent.Wherein, introductory die is adjusted
Type includes the preference pattern (for example, linear regression, random forest etc.) for reselecting initial model, or in adjustment model
The size (result exported according to model, appropriate increase or the value for reducing model parameter) of model parameter, finally gives determination
Diagnostic model.
Such as Fig. 1, the mobile telephone registration failure automatic diagnosis method based on the present embodiment includes：
Step S-1, obtain registration daily record during registration failure., the step can be by the independent mould in mobile phone
Block, which is realized, can also be integrated in the function in registration daily record collection module.Registration failure diagnostic module is independently of the use of mobile phone
In to mobile phone registration failure carry out fault diagnosis external equipment when, its USB interface that can lead to, wifi, serial ports etc. it is wired or
The registration daily record captured when person wirelessly fails mobile telephone registration is transferred to registration failure diagnostic module as the defeated of module
Enter.When registration failure diagnostic module is a program module being similarly installed in mobile phone, the registration daily record of crawl can be direct
The registration failure diagnostic module in same mobile phone is input to as input.
Step S-2, registration daily record is imported into diagnostic model.The registration daily record of input is imported into by registration failure diagnostic module
In diagnostic model.
Step S-3, diagnostic model are analyzed and export failure cause.Registration failure diagnostic module is carried out based on the diagnostic model
Calculate, analyze to obtain result, the reason for feeding back registration failure.
Embodiment two
A kind of mobile telephone registration failure auto-check system, including registration failure diagnostic module and the registration in mobile phone
Log collection module.
Registration daily record collection module is mounted in the module in mobile phone, and it includes the letter for being used to detect the signal intensity of mobile phone
Number intensity detection unit, the signal quality detection unit of signal quality for detecting mobile phone, for detecting mobile phone current location
The base station number detection unit of the base station number of surrounding and the registration log collection thread established in interior of mobile phone.For
Carry out being mounted with two SIM cards (SIM card 1 and SIM card 2) in the mobile phone of registration log collection, and this two SIM card is different
The SIM card of operator's either different model of same operator.
Signal strength detection unit, it can detect the current signal intensity of mobile phone, and can be set in advance according to reference
The threshold range for division signals strength grade put, output can reflect the signal of the current signal strength grade of mobile phone.
Signal quality detection unit, it can be commented the signal quality of mobile phone according to the current network state of mobile phone
Estimate, and export the signal for the current signal quality grade that can reflect mobile phone.
Base station number detection unit, it can detect the base station number around mobile phone current location, and export reflection and be somebody's turn to do
The information of base station number.
Registration log collection thread should possess following function：
A) registration log collection thread can detect crucial Log information when mobile phone starts registration (such as in high-pass platform
Keyword：CM_SERVICE_REQ), it may determine that the SIM card of mobile phone starts to register net according to the crucial Log information thread
Network.The thread only just proceeds by log information collection when detecting keyword CM_SERVICE_REQ.
B) whether successfully registration log collection thread can detect mobile telephone registration crucial Log information (such as high-pass platform
In keyword：CM_SERVICE_CFG), it may determine that the SIM card registered network of mobile phone is according to the crucial Log information thread
No end, and know registering result.The thread stops log information receipts once keyword CM_SERVICE_CFG is detected
Collection.
C) registration log collection thread can pass through signal strength detection unit, signal quality detection unit, base station number
The intensity and the base station number of quality and surrounding of the quantity monitoring current phone reception signal of detection unit monitoring network environment
Information, when finding the signal intensity of the mobile phone signal either base station around the signal quality of mobile phone signal or mobile phone current location
(i.e. when in signal strength detection unit, signal quality detection unit and base station number detection unit when quantity varies widely
The output signal of at least one unit when changing) triggering mobile phone SIM card registered network function so that mobile phone weighs automatically
The new register flow path for starting network, and the daily record in registration process is captured again.
D) register log collection thread can automatically SIM card 1 and the progress network registry of SIM card 2 in switch mobile phone, when
After SIM card 1 completes network registry flow, the network registry flow of automatic start SIM card 2.Automation can thus be reached
Crawl different operators SIM card register flow path log information.
Such as Fig. 3, the registration daily record crawl flow based on above-mentioned registration daily record collection module：
Step S-A-1, when one in the signal intensity of mobile phone, signal quality, peripheral base station quantity changes, make expert
The registered network of SIM card 1 of machine；
Step S-A-2, when detecting that registration starts key message, start log information crawl；
Step S-A-3, when detecting that registration terminates key message, stop daily record information scratching；
Step S-A-4, preserve registration daily record；
Step S-A-5, start the registered network of SIM cards of mobile phones 2, and repeat step S-A-2 to step S-A-4.
The registration daily record letter of the different SIM card registered network of mobile phone of the crawl of automation under without environment can be reached
Breath.Most of all, the crawl of log information is captured at the very start from registration, and after register flow path terminates at once
Stop crawl, largely eliminate the log information outside the flow of SIM card registered network.Received based on the registration daily record
The daily record data that collection module cellphone arrives carries out model analysis, simplifies the difficulty of modeling, and substantially increase and establish model
Accuracy and specific aim.
Registration failure diagnostic module can be independently of mobile phone be used for the outer of fault diagnosis is carried out to mobile phone registration failure
Portion's equipment, its USB interface that can lead to, wifi, serial ports etc. is wired or is wirelessly connected with mobile phone.Can also be same
A program module in mobile phone, or be a core dedicated for accident analysis being similarly installed in mobile phone
Piece.Once failing during the SIM card registered network of mobile phone, then the registration daily record of this mobile telephone registration process is imported
Automated analysis is carried out into registration failure diagnostic module, the reason for orienting mistake.
The diagnostic model of registration failure diagnostic module is that the registration daily record collected based on foregoing registration daily record collection module is built
Vertical.The flow of foundation includes：
Step S-A, registers log collection, and registration daily record refers to the log information of mobile telephone registration network.It is a large amount of to utilize registration
Log collection module largely captures the registration daily record of mobile telephone registration network.Here registration daily record needs to include SIM cards of mobile phones
Registered network successfully registers registration daily record (the registration mistake of daily record (daily record of succeeding in registration) and the failure of SIM cards of mobile phones registered network
Lose daily record).The sample size of registration daily record is bigger, and the model of foundation is more accurate.For example, it is successful to capture collection registered network
Register 1000 parts of daily record sample, 1000 parts of the registration daily record sample of registered network failure.
Step S-B, log integrity is registered, registration daily record is collected into step S-A and carries out the effective daily record of pretreatment acquisition
Data.Railway Project be present in original registration daily record data：Inconsistency, repeatability, Noise, latitude are high.Wherein, due to this
Crawl of the registration daily record collection module of invention to log information is captured at the very start from registration, and in register flow path
Stop crawl after end at once, largely eliminate the log information outside the flow of SIM registered networks, noise content
Than relatively low.But, however it remains inconsistency, it is repeated the problems such as.Pretreatment in step S-B is clear by data comprising needing
Wash, the one or more in data integration, data conversion and hough transformation this several method, original registration daily record is carried out
Processing, to choose suitable attribute as data mining attribute.In the present embodiment, first, to original registration daily record number
According to participle, common segmentation methods have：Participle based on dictionary, the participle based on statistics, the participle based on semanteme, Ke Yigen
Need voluntarily to select according to what model was established.Secondly, duplicate removal processing is carried out to segmenting later data, removes unique attribute, removes
Repeatability, take out negligible field.Again, data regularization processing is carried out to data, it is smaller than initial data more to obtain, but
The mining data collection of data integrity is not destroyed, and the data set can obtain and initial data identical data mining effect.
Step S-C, diagnostic model are established, and diagnostic model is established based on the effective daily record data obtained in step S-B.Specifically
Including step：
Step S-C-1, model training, based on carrying out model training in most of effectively daily record data to obtain introductory die
Type.The 800 parts of daily records of succeeding in registration that will be registered in step S-A in the 1000 parts of daily records of succeeding in registration being collected into of daily record collection module
Effective daily record data and step S-A in 800 parts in the 1000 parts of registration failure daily records being collected into of registration daily record collection module
Effective daily record data of registration failure daily record carries out model training to obtain initial model.
Step S-C-2, Modifying model, based on the remainder amendment initial model of effective daily record data to obtain diagnosis mould
Type.Effective daily record data includes supplemental characteristic and result data, and wherein result data refers to represent hand corresponding to the valid data
The whether successful data of machine register flow path.In step S-C-2, for registering what daily record collection module mobile phone arrived in step S-A
In 1000 parts of daily records of succeeding in registration log collection is registered in the valid data and step S-A of remaining 200 parts of daily records of succeeding in registration
Module cellphone to 1000 parts of registration failure daily records in remaining 200 parts of registration failure daily records valid data, these are effective
Each in data carries out following processing：
Step S-C-2-1, the supplemental characteristic of valid data is imported into initial model as input.
Step S-C-2-2, initial model output model result, initial model are carried out at analysis according to the supplemental characteristic of input
Reason, it is contemplated that whether the mobile telephone registration based on the supplemental characteristic succeeds, and is exported estimated result as model result.
Whether step S-C-2-3, model result and the result data for comparing initial model output are consistent：If inconsistent
Adjust the preference pattern and/or model parameter of initial model；Initial model is not adjusted if consistent.Wherein, introductory die is adjusted
Type includes the preference pattern (for example, linear regression, random forest etc.) for reselecting initial model, or in adjustment model
The size (result exported according to model, appropriate increase or the value for reducing model parameter) of model parameter, finally gives determination
Diagnostic model.
Mobile telephone registration failure automatic diagnosis method based on the present embodiment includes：
Step S-1, obtain registration daily record during registration failure., the step can be by the independent mould in mobile phone
Block, which is realized, can also be integrated in the function in registration daily record collection module.Registration failure diagnostic module is independently of the use of mobile phone
In to mobile phone registration failure carry out fault diagnosis external equipment when, its USB interface that can lead to, wifi, serial ports etc. it is wired or
The registration daily record captured when person wirelessly fails mobile telephone registration is transferred to registration failure diagnostic module as the defeated of module
Enter.When registration failure diagnostic module is a program module being similarly installed in mobile phone, the registration daily record of crawl can be direct
The registration failure diagnostic module in same mobile phone is input to as input.
Step S-2, registration daily record is imported into diagnostic model.The registration daily record of input is imported into by registration failure diagnostic module
In diagnostic model.
Step S-3, diagnostic model are analyzed and export failure cause.Registration failure diagnostic module is carried out based on the diagnostic model
Calculate, analyze to obtain result, the reason for feeding back registration failure.
Specific embodiment described herein is only to spirit explanation for example of the invention.Technology belonging to the present invention is led
The technical staff in domain can be made various modifications or supplement to described specific embodiment or be replaced using similar mode
Generation, but without departing from the spiritual of the present invention or surmount scope defined in appended claims.
Claims (10)
- The automatic diagnosis method 1. a kind of mobile telephone registration fails, it is characterised in that including：Step S-1, obtain registration daily record during registration failure；Step S-2, the registration daily record is imported into diagnostic model；Step S-3, the diagnostic model are analyzed and export failure cause.
- A kind of 2. mobile telephone registration failure automatic diagnosis method according to claim 1, it is characterised in that the diagnostic model Method for building up include：Step S-A, registers log collection, and the registration daily record refers to the log information of mobile telephone registration network；Step S-B, log integrity is registered, registration daily record is collected into the step S-A and carries out the effective daily record of pretreatment acquisition Data；Step S-C, diagnostic model are established, and diagnostic model is established based on the effective daily record data obtained in the step S-B.
- A kind of 3. mobile telephone registration failure automatic diagnosis method according to claim 2, it is characterised in that the step S-A Including：Step S-A-1, when one in the signal intensity of mobile phone, signal quality, peripheral base station quantity changes, make mobile telephone registration Network；Step S-A-2, when detecting that registration starts key message, start log information crawl；Step S-A-3, when detecting that registration terminates key message, stop daily record information scratching；Step S-A-4, preserve registration daily record.
- A kind of 4. mobile telephone registration failure automatic diagnosis method according to claim 3, it is characterised in that the step S-C Including：Step S-C-1, model training, based on carrying out model training in most of effectively daily record data to obtain introductory die Type；Step S-C-2, Modifying model, based on the initial model effectively described in the remainder amendment of daily record data to be examined Disconnected model.
- A kind of 5. mobile telephone registration failure automatic diagnosis method according to claim 4, it is characterised in that：The registration daily record During including succeeding in registration succeed in registration daily record and registration failure when registration failure daily record；In step S-C-1, based on A%'s Succeed in registration daily record valid data and A% registration failure daily record valid data carry out model training.
- A kind of 6. mobile telephone registration failure automatic diagnosis method according to claim 5, it is characterised in that effective daily record Data include supplemental characteristic, result data；The step S-C-2 includes：Step S-C-2-1, the initial model is imported into using the supplemental characteristic of the valid data as input；Step S-C-2-2, the initial model output model result；Step S-C-2-3, whether the model result is consistent with the result data, is adjusted if inconsistent described first The preference pattern and model parameter of beginning model, the initial model is not adjusted if consistent.
- The auto-check system 7. a kind of mobile telephone registration fails, it is characterised in that：Including the registration log collection in mobile phone Module, registration failure diagnostic module；Input registration daily record can be imported into diagnostic model by the registration failure diagnostic module And export diagnostic result；The input registration daily record refers to the registration daily record of mobile telephone registration network failure, the diagnostic model The registration daily record collected based on the registration daily record collection module is established.
- A kind of 8. mobile telephone registration failure auto-check system according to claim 7, it is characterised in that：The registration daily record Collection module is used for the registration daily record placement unit for obtaining the log information of mobile telephone registration network；The log-on message crawl is single Member, start crawl log information when detecting that registration starts key message, and when detecting that registration terminates key message Stop crawl log information.
- A kind of 9. mobile telephone registration failure auto-check system according to claim 8, it is characterised in that：The registration daily record Collection module includes the signal strength detection unit for detection signal intensity, the signal quality for detection signal quality detects Unit, the base station number detection unit for detecting peripheral base station quantity, the signal strength detection unit, the signal quality Detection unit and the base station number detection unit can trigger mobile telephone registration network.
- 10. mobile telephone registration Information Collection System according to claim 9, it is characterised in that：The mobile phone includes at least two Open Mobile phone card, the registration daily record for registering all Mobile phone cards on mobile phone described in daily record collection module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710585025.6A CN107426750A (en) | 2017-07-18 | 2017-07-18 | A kind of mobile telephone registration failure automatic diagnosis method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710585025.6A CN107426750A (en) | 2017-07-18 | 2017-07-18 | A kind of mobile telephone registration failure automatic diagnosis method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107426750A true CN107426750A (en) | 2017-12-01 |
Family
ID=60430794
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710585025.6A Pending CN107426750A (en) | 2017-07-18 | 2017-07-18 | A kind of mobile telephone registration failure automatic diagnosis method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107426750A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113467423A (en) * | 2021-07-01 | 2021-10-01 | 中山大学 | PEMFC fault diagnosis method and system based on cloud platform |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080263398A1 (en) * | 2007-04-19 | 2008-10-23 | Ntt Docomo, Inc. | Mobile terminal apparatus and diagnosis method for mobile terminal apparatus |
| CN102045181A (en) * | 2009-10-10 | 2011-05-04 | 中国移动通信集团公司 | Method and device for handling terminal offline fault |
| CN102158748A (en) * | 2010-05-14 | 2011-08-17 | 青岛海信传媒网络技术有限公司 | Methods and systems for collecting and uploading logs |
| CN102402517A (en) * | 2010-09-09 | 2012-04-04 | 北京启明星辰信息技术股份有限公司 | Method and system for establishing database normal login model and detecting abnormal login behavior |
| CN103051956A (en) * | 2012-12-24 | 2013-04-17 | 乐视致新电子科技（天津）有限公司 | Set-top box for realizing log report and fault diagnosis and method thereof |
| CN105653444A (en) * | 2015-12-23 | 2016-06-08 | 北京大学 | Internet log data-based software defect failure recognition method and system |
| CN105975604A (en) * | 2016-05-12 | 2016-09-28 | 清华大学 | Distribution iterative data processing program abnormity detection and diagnosis method |
| CN106570513A (en) * | 2015-10-13 | 2017-04-19 | 华为技术有限公司 | Fault diagnosis method and apparatus for big data network system |
-
2017
- 2017-07-18 CN CN201710585025.6A patent/CN107426750A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080263398A1 (en) * | 2007-04-19 | 2008-10-23 | Ntt Docomo, Inc. | Mobile terminal apparatus and diagnosis method for mobile terminal apparatus |
| CN102045181A (en) * | 2009-10-10 | 2011-05-04 | 中国移动通信集团公司 | Method and device for handling terminal offline fault |
| CN102158748A (en) * | 2010-05-14 | 2011-08-17 | 青岛海信传媒网络技术有限公司 | Methods and systems for collecting and uploading logs |
| CN102402517A (en) * | 2010-09-09 | 2012-04-04 | 北京启明星辰信息技术股份有限公司 | Method and system for establishing database normal login model and detecting abnormal login behavior |
| CN103051956A (en) * | 2012-12-24 | 2013-04-17 | 乐视致新电子科技（天津）有限公司 | Set-top box for realizing log report and fault diagnosis and method thereof |
| CN106570513A (en) * | 2015-10-13 | 2017-04-19 | 华为技术有限公司 | Fault diagnosis method and apparatus for big data network system |
| CN105653444A (en) * | 2015-12-23 | 2016-06-08 | 北京大学 | Internet log data-based software defect failure recognition method and system |
| CN105975604A (en) * | 2016-05-12 | 2016-09-28 | 清华大学 | Distribution iterative data processing program abnormity detection and diagnosis method |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113467423A (en) * | 2021-07-01 | 2021-10-01 | 中山大学 | PEMFC fault diagnosis method and system based on cloud platform |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104143078B (en) | Living body faces recognition methods, device and equipment | |
| TWI706318B (en) | Solder paste printing quality detecting method, data processing device and computer storage medium | |
| CN105788355B (en) | A kind of parking stall monitoring System and method for based on Beacon technology | |
| CN109253888A (en) | Detection method and system for vehicle vehicle condition | |
| CN105426862B (en) | Analysis method and its system based on RFID, location technology and video technique | |
| CN108241580A (en) | The test method and terminal of client-side program | |
| CN114444865A (en) | Municipal Epidemic Prevention Management Device and Method | |
| CN104175313B (en) | Circuit board testing monitoring system and method | |
| CN109874148A (en) | Antenna feed anomaly detection method, device, system and computer equipment | |
| CN108123935A (en) | Circuit board detection method, device, storage medium and terminal | |
| CN112104613B (en) | Honey net testing system based on data flow packet analysis and testing method thereof | |
| CN110493085A (en) | Statistical method, system, electronic equipment and the medium of IPv6 active users | |
| CN106789391A (en) | A kind of automated testing method and device of router DHCP functions | |
| CN108414008A (en) | A kind of city environmental quality monitoring system | |
| CN105046614A (en) | Mobile medical system based on mobile terminal | |
| CN107426750A (en) | A kind of mobile telephone registration failure automatic diagnosis method and system | |
| CN106571971A (en) | Empty shell website detection method, device and system | |
| JP4913243B2 (en) | Method for generating a signal sequence reproducible by a tester inspecting a mobile radio | |
| CN111858331B (en) | Front-end control coverage detection method, device and equipment | |
| CN105430678B (en) | The method and device of PTN device interface testing | |
| CN112927720A (en) | Audio anomaly detection method and device | |
| CN106896710A (en) | Scene configurable time synchronization test system and implementation method | |
| CN106934985A (en) | A kind of system of examination hall invigilator and the application method of the system | |
| CN112784627B (en) | Method, device, electronic device and medium for identity recognition | |
| CN104794812A (en) | Card swiping system and managing method thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171201 |