CN109286526A - A kind of wifi system running policy dynamic adjusting method and device - Google Patents
A kind of wifi system running policy dynamic adjusting method and device Download PDFInfo
- Publication number
- CN109286526A CN109286526A CN201811167503.2A CN201811167503A CN109286526A CN 109286526 A CN109286526 A CN 109286526A CN 201811167503 A CN201811167503 A CN 201811167503A CN 109286526 A CN109286526 A CN 109286526A
- Authority
- CN
- China
- Prior art keywords
- data
- flow rate
- mode
- rate mode
- wifi system
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0893—Assignment of logical groups to network elements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/06—Management of faults, events, alarms or notifications
- H04L41/0631—Management of faults, events, alarms or notifications using root cause analysis; using analysis of correlation between notifications, alarms or events based on decision criteria, e.g. hierarchy, tree or time analysis
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/142—Network analysis or design using statistical or mathematical methods
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/147—Network analysis or design for predicting network behaviour
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0876—Network utilisation, e.g. volume of load or congestion level
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/06—Testing, supervising or monitoring using simulated traffic
-
- 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
Abstract
The application provides a kind of wifi system running policy dynamic adjusting method and device, the method first collects the operation data of current wifi system, data aggregate is carried out to operation data again, to do batch statistics and analysis to operation data, generates the network health data including machine learning model and determine present flow rate mode；Operation data and machine learning model after comparing polymerization again, determine whether present flow rate mode exception occurs；If there is exception, new flow mode is matched in flow rate mode library using mode invocation module and adjusts the corresponding operation reserve of operation reserve of wifi system.By the prediction data and truthful data that compare machine learning model, dynamic adjustment operation reserve, to adapt to the different operating status variation of wifi system, the method of adjustment of traditional wifi network system is solved when there is new operating status variation, can not be from preset control strategy, the problem of selecting adaptable control strategy.
Description
Technical field
This application involves radio network technique field more particularly to a kind of wifi system running policy dynamic adjusting method and
Device.
Background technique
Extensive wifi network system, including the multiple wireless access points (Wireless being distributed in each region
Access Point, AP) and the networks such as multiple wireless access controllers (Wireless Access Controller, AC) set
It is standby.By multiple AP and AC, wirelessly access is provided for multiple wireless devices in region.In general, in WiFi network system
In, if the operating status of wifi network system changes, it is pre-configured to be in wifi network system that a control strategy will
It is activated, and goes to change the behavior of each network equipment in wifi network system, to adapt to the operating status after changing.
But by the method for adjustment of existing wifi network system it is found that the control strategy of wifi network system is preset at net
In the controller of network system, through the operating status to change in detection wifi network system, from preset control strategy
A compatible control strategy is selected, this requirement needs built-in enough control strategies in wifi network system.But wifi
Operating status in network system constantly changes in actual use, and as the long-time of wifi network system uses, meeting
Constantly there is new operating status to occur.At this moment, it by detecting the variation of operating status, then can not be selected from preset control strategy
Adaptable control strategy is selected, the normal operation of wifi network system is influenced.
Summary of the invention
This application provides a kind of wifi system running policy dynamic adjusting method and devices, to solve traditional wifi network
The method of adjustment of system can not select adaptable control from preset control strategy when there is new operating status variation
Make the problem of strategy.
On the one hand, the application provides a kind of wifi system running policy dynamic adjusting method, comprising:
Collect the operation data of current wifi system；
Data aggregate is carried out to the operation data, and the operation data after polymerization is saved to database；
Batch statistics and analysis done to the operation data saved in database, generate network health data and
Determine present flow rate mode；The network health data include the machine learning model established according to the operation data；
The operation data and the machine learning model after comparison polymerization；
According to comparing result, determine whether the present flow rate mode exception occurs；
If exception occurs in the present flow rate mode, new flow is matched in flow rate mode library using mode invocation module
Mode；And the operation reserve of the wifi system is adjusted as the operation reserve under the new flow mode.
Optionally, the Rest API that the method is defined by Swagger collects the operation data of current wifi system, institute
It states operation data and receives message byte number and total transmission message byte number including at least total on WAN LAN mouth.
Optionally, batch statistics and analysis is done to the operation data saved in database, generates network health
The step of data and determining present flow rate mode, comprising:
Scheme program starts the worker thread in worker thread pond；
The data on flows in the operation data by the worker thread, after extracting polymerization；
Corresponding present flow rate mode is determined according to extracted data on flows；
Average value, standard deviation and regression analysis are carried out to the data on flows, establish machine learning model.
Optionally, according to comparing result, determine whether the present flow rate mode abnormal step occurs, comprising:
Extract the calculating parameter in the operation data；
According to the calculating parameter and the machine learning model, model prediction data is determined；
If the operation data and the model prediction data, meets preset condition requirement, judge the present flow rate
Mode occurs abnormal；
If the operation data and the model prediction data, it is unsatisfactory for preset condition requirement, judges the current stream
Amount mode does not occur exception.
Optionally, if exception occurs in the present flow rate mode, the method also includes:
Worker thread sounds an alarm instruction to exception management module；
The alarm command is sent instrument board by the exception management module, so that the instrument board shows the alarm
Instruction；Alternatively,
Initial configuration notification rule of the exception management module according to wifi system, signal an alert.
Optionally, if the present flow rate mode occur it is abnormal, using mode invocation module in flow rate mode library
The step of with new flow mode, comprising:
Judge under the present flow rate mode exception, if need to change flow rate mode；
It is the corresponding operation reserve of new flow model selection if necessary to change flow rate mode；
Optionally, if necessary to change flow rate mode, the step of being new flow model selection corresponding operation reserve, packet
It includes:
Operation reserve under new flow mode described in periodic test；
Judge whether to selected new operation reserve；
If selected new operation reserve, the operation reserve of selection is executed.
Optionally, it if necessary to change flow rate mode, the step of being new flow model selection corresponding operation reserve, also wraps
It includes:
The policy tag edited is obtained by instrument board；
According to the policy tag, the corresponding operation reserve of the new flow mode is obtained in policy library；
Enable the corresponding operation reserve of the new flow mode obtained, and the corresponding operation of disabling present flow rate mode
Strategy.
Optionally, according to the policy tag, the corresponding operation reserve of the new flow mode is obtained in policy library
Step, comprising:
If operation reserve corresponding to the new flow mode has not been obtained from the policy library, according to the new flow
Mode creates new strategy；
The new strategy of creation is updated to the policy library.
On the other hand, the application also provides a kind of wifi system running policy dynamic adjusting device, comprising:
Data collection module, for collecting the operation data of current wifi system；
Data aggregate module, for carrying out data aggregate to the operation data, and by the operation number after polymerization
According to preservation to database；
Statistical analysis module generates net for doing batch statistics and analysis to the operation data saved in database
Network state of health data and determining present flow rate mode；The network health data include being built according to the operation data
Vertical machine learning model；
Contrast module, for comparing the operation data and the machine learning model after polymerizeing；
Abnormal judgment module, for determining whether the present flow rate mode exception occurs according to comparing result；
New flow Pattern Matching Module, if there is exception for the present flow rate mode, using mode invocation module
New flow mode is matched in flow rate mode library；
Operation reserve adjusts module, and the operation reserve for adjusting the wifi system is the fortune under the new flow mode
Row strategy.
From the above technical scheme, the application provides a kind of wifi system running policy dynamic adjusting method and device,
The method first collects the operation data of current wifi system, then carries out data aggregate to operation data, then do to operation data
Batch statistics and analysis generates network health data and determines present flow rate mode, wherein network health data
Including the machine learning model established according to operation data；Operation data and machine learning model after comparing polymerization again, and
According to comparing result, determine whether present flow rate mode exception occurs；If there is exception in present flow rate mode, using mode tune
New flow mode is matched in flow rate mode library with module；The operation reserve of wifi system is adjusted as the operation under new flow mode
Strategy.Technical solution provided by the present application, by comparing the prediction data and truthful data of machine learning model, dynamic adjustment fortune
Row strategy, to adapt to the different operating status variation of wifi system, the method for adjustment for solving traditional wifi network system is occurring
, can not be from preset control strategy when new operating status variation, the problem of selecting adaptable control strategy.
Detailed description of the invention
In order to illustrate more clearly of the technical solution of the application, letter will be made to attached drawing needed in the embodiment below
Singly introduce, it should be apparent that, for those of ordinary skills, without creative efforts, also
Other drawings may be obtained according to these drawings without any creative labor.
Fig. 1 is a kind of flow diagram of wifi system running policy dynamic adjusting method of the application；
Fig. 2 is the flow diagram that the application generates network health data；
Fig. 3 is that the application determines abnormal flow diagram；
Fig. 4 is the application to the flow diagram sounded an alarm extremely；
Fig. 5 is the flow diagram that the application selects operation reserve；
Fig. 6 is the flow diagram that the application executes operation reserve；
Fig. 7 is the flow diagram that the application determines operation reserve under new flow mode；
Fig. 8 is a kind of structural schematic diagram of wifi system running policy dynamic adjusting device of the application.
Specific embodiment
Embodiment will be illustrated in detail below, the example is illustrated in the accompanying drawings.In the following description when referring to the accompanying drawings,
Unless otherwise indicated, the same numbers in different drawings indicate the same or similar elements.Implement described in following embodiment
Mode does not represent all embodiments consistent with the application.It is only and be described in detail in claims, the application
The example of the consistent system and method for some aspects.
In technical solution provided by the present application, the wifi system is also known as wifi network, refer to including network controller,
Wireless access point device (AP equipment), wireless access control equipment (AC equipment) form, wifi network of certain scale.
Wherein, AP equipment and AC equipment all include multiple, are distributed in different working environments, AP equipment can directly with the end of user
End equipment connection, provides network service for terminal device.
It is a kind of flow diagram of wifi system running policy dynamic adjusting method of the application referring to Fig. 1.It can by Fig. 1
Know, wifi system running policy dynamic adjusting method provided by the present application, comprising the following steps:
S1: the operation data of current wifi system is collected.
WAN mouth in technical solution provided by the present application, after user connects the AP of WiFi system, on user data and AP
Data on flows can be collected, assemble.Can in wifi system it is built-in collect operation data device, which can be by right
The data message that each AP equipment (or AC equipment) reports in wifi system is grabbed, and operation data is got.
Further, it can be collected by the Rest API that Swagger is defined in technical solution provided by the present application current
The operation data of wifi system is directly docked collection procedure with wifi control system with will pass through Rest api interface.
The operation data includes at least total reception message byte number and total transmission message byte number on WAN LAN mouth.For example,
In the interface definition of Rest API, indicate to receive message byte number on WAN LAN mouth always with rxBytes；With txBytes
It indicates total on WAN LAN mouth and sends message byte number.
S2: data aggregate is carried out to the operation data, and the operation data after polymerization is saved to database.
In technical solution provided by the present application, polymerization and analysis in real time are carried out to data first, data is then saved and divides
Result is analysed into database.Wherein, polymerization is carried out to data to refer to and extract the data of collection, remove in operation data
Invalid data only retains the data that can characterize operating status, for example, summarizing hour data by txBytes, i.e., passes from AP
The defeated aggregated data to WAN port.
S3: batch statistics and analysis is done to the operation data saved in database, generates network health data
And determine present flow rate mode；The network health data include the machine learning mould established according to the operation data
Type；
In actual use, if the operating status of wifi system changes, such as network equipment quantity, the access of access
Equipment be engaged in different network behaviors, operating parameter in wifi system, such as operation load, flow are handled up, and will be changed.
When the operating parameter that wifi system provides in real time is it is impossible to meet the normal operation of access device, or access each equipment it
Between when mutually generating large effect, it may be determined that cannot be met the requirements for current network health status, that is, there is network health and ask
Topic.In order to guarantee that wifi system is capable of providing more stable network service, the operation reserve needs of wifi system are changed correspondingly.
It, can be by the control program (ISM) built in wifi system according to the fortune of collection in technical solution provided by the present application
The health status of row data and the flow rate mode determined analysis AP network.The state of health data is for evaluating current network
Health status, including the statistic analysis result according to the operation data, for example, average value, variance, standard deviation etc., and with number
According to regression analysis based on establish machine learning model, the machine learning model can be simple regression model, can also
To be to utilize machine learning algorithm, continuous data aggregate analysis and perfect complex model.The machine learning model is used for
Flow parameter value corresponding with present flow rate mode is predicted according to real-time running data.So as in the next steps to real-time
Operation data and the flow parameter value of prediction are compared.
Further, it as shown in Fig. 2, doing batch statistics and analysis to the operation data saved in database, generates
It is the step of network health data and determining present flow rate mode, further comprising the steps of:
S301: scheme program starts the worker thread in worker thread pond；
S302: the data on flows in the operation data by the worker thread, after extracting polymerization；
S303: corresponding present flow rate mode is determined according to extracted data on flows；
S304: average value, standard deviation and regression analysis are carried out to the data on flows, establish machine learning model.
In practical application, and wan interface total flow (RX byte, it is total to receive message byte number；With TX byte, message is always sent
Byte number), changing over time is very basic parameter for total flow pattern analysis.Obviously, interface total flow can be with
It is used together from other parameters with different potential combinations.Present flow rate mode can be preset at it is a variety of silent in wifi system
Recognize flow rate mode, can be and be manually set according to time of the wifi system in application.For example, flow rate mode may include
Monday flow rate mode, Tuesday flow rate mode, Wednesday flow rate mode, Thursday flow rate mode, Friday flow rate mode, holiday, every circumfluence
Amount mode etc..
Illustratively, if the runing time of current wifi system is Monday, accordingly determine that present flow rate mode is Monday
Flow rate mode, in a data-gathering process, the operation data being collected into is as follows: after polymerization processing
As it can be seen that total transmission message byte number of AP is opposite in wifi system in time in upper table from 8:00 to 22:00
It is larger in other times section, therefore, in order to which control is implemented in the operation more accurately to wifi system, 8:00 can be used to 22:
The data collected in 00 time carry out linear regression analysis, determine connection number x and total mathematics sent between message byte number Q
Model, as the machine learning model, regression result is as follows:
Q=18433.66+4057.76x；
Equally, the data collected according to 23:00 to 7:00 are it is found that the machine learning model from 23:00 to 7:00 is as follows:
Q=3.444+19.583x；
In this example, after determining that present flow rate mode is Monday flow rate mode and constructs above-mentioned machine learning model,
Can further it judge whether present flow rate mode exception occurs, it may be assumed that
S4: the operation data and the machine learning model after comparison polymerization.
In technical solution provided by the present application, after being collected into corresponding operation data, it can be predicted by operation data
Operation data value in the time of one end, and it is excessive if there are deviations between the operation data value got and the predicted value of data
Or it is not able to satisfy certain preset requirement, it is determined that present flow rate mode occurs abnormal, it may be assumed that
S5: according to comparing result, determine whether the present flow rate mode exception occurs.
Further, as shown in figure 3, according to comparing result, determine whether the present flow rate mode abnormal step occurs
Suddenly, further comprising the steps of:
S501: the calculating parameter in the operation data is extracted；
S502: according to the calculating parameter and the machine learning model, model prediction data is determined；
S503: if the operation data and the model prediction data, meeting preset condition requirement, judges described current
Flow rate mode occurs abnormal；
S504: if the operation data and the model prediction data, it is unsatisfactory for preset condition requirement, is worked as described in judgement
Preceding flow rate mode does not occur exception.
It, can first built-in a variety of different preset conditions requirements, institute in the control program of wifi system in the present embodiment
Preset condition is stated to require to can be the functional relation between a threshold value, a floating range or a kind of parameter.Clearly for
Different flow rate mode, wherein preset condition requires can also be different.
By taking the above-mentioned flow rate mode shown as an example, if 13:00 is collected into the operation that AP is transmitted in WAN port TX in the afternoon
Data, the calculating parameter extracted from the operation data are total to send message byte number are as follows: 600Kbytes；Connection number are as follows: 20.
Wherein, connection number 20 can be used as the input in established machine learning model, i.e., always sends the predicted value Q=of message byte number
1843.366+405.776 × 20=9958.886Kbytes.
Furthermore it is possible in the control program of wifi system, built-in following preset condition requirement:
if(AP.WAN_Q_TX/AP.WAN_TX_Q_ML_MODEL<10&&AP.AP.WAN_Q_TX<15MBPS&&
AP.WAN_Q_Busy_TX_Ave Counter>50){reject AP.NEW_STATION；}；
Wherein, AP.WAN_Q_TX is real-time TX handling capacity of the AP in WAN port；
AP.WAN_TX_Q_ML_MODEL is the AP WAN port handling capacity that model is generated based on machine learning；
AP.WAN_Q_Busy_TX_Ave Counter is for calculating the percentage that WLAN sends busy average counter；
AP.NEW_STATION indicates the equipment that this AP is added in new website in advance.
As it can be seen that flow rate mode is in a good state of health in actual operating data, then present flow rate mode health alarm is clear
, it means that existing known mass flow mode is normal in the customer traffic specific discharge pattern base of AP equipment.And it always sends
Message byte number 600Kbytes is much smaller than predicted value 9958.886Kbytes, that is, triggers above-mentioned preset condition requirement, i.e. determination is worked as
There is exception in preceding flow rate mode, needs to adjust operation reserve to adapt to above-mentioned exception.
S6: it if exception occurs in the present flow rate mode, is matched in flow rate mode library using mode invocation module new
Flow rate mode, and the operation reserve of the adjustment wifi system is the operation reserve under the new flow mode.
In technical solution provided by the present application, if exception occurs in present flow rate mode, present flow rate mode can not be protected
The normal operation of wifi system is demonstrate,proved, therefore new flow rate mode can be matched in flow rate mode library, to adapt to currently run shape
The variation of state.In technical solution provided by the present application, built-in flow rate mode library, flow rate mode it can be deposited in library in wifi system
Multiple flow rate modes being adapted under different conditions are contained, the flow rate mode may come from the setting of user, can also come
From in the big data of internet.Further, can also be arranged under each flow rate mode stored in the flow rate mode library more
A state adjustment programme.
By taking above-mentioned flow rate mode as an example, if it is determined that abnormal, above-mentioned preset condition requirement occurs in current Monday flow rate mode
It will be triggered, that is, judge AP.WAN_Q_TX/AP.WAN_TX_Q_ML_MODEL whether less than 10 and AP.AP.WAN_Q_
Whether TX is less than whether 15MBPS AP.WAN_Q_Busy_TX_Ave Counter is greater than 50, when the conditions are satisfied,
Reject AP.NEW_STATION instruction will be executed, i.e., any new website for wishing this AP of addition, which will be rejected, executes above-mentioned strategy,
To maintain the normal operation of current wifi system.
In the section Example of the application, as shown in figure 4, if there is abnormal, the side in the present flow rate mode
Method is further comprising the steps of:
S601: worker thread sounds an alarm instruction to exception management module；
S602: the alarm command is sent instrument board by the exception management module, so that the instrument board shows institute
State alarm command；
S603: initial configuration notification rule of the exception management module according to wifi system, signal an alert.
In technical solution provided by the present application, exception management module built in wifi system, exception management module can be directed to
The exception of appearance makes corresponding alarm signal.It therefore, in the present embodiment, can be by exception management module by alarm command
It is sent to instrument board (UI), instrument board (UI) is made to show alarm command.Obviously, the alarm command shown on the dash panel can be
With obvious color signal or other signals.Exception management module can also notify to advise according to the initial configuration of wifi system
Then, Email or message such as are sent to administrator, with the abnormal conditions for notifying administrator to occur.
In the section Example of the application, as shown in figure 5, if there is exception in the present flow rate mode, using mould
Formula calling module matches the step of new flow mode in flow rate mode library, further includes:
S611: judge under the present flow rate mode exception, if need to change flow rate mode；
S612: being the corresponding operation reserve of new flow model selection if necessary to change flow rate mode.
In the present embodiment, it can first judge currently occur under abnormal flow rate mode, if need to change flow rate mode.Example
Such as, if the load of wifi system is superfluous, i.e., current AP can maintain more equipment to access, and it is general to adjust operation reserve at this time
It is that can pass through the plan under present flow rate mode under the premise of not changing flow rate mode at this time using superfluous Internet resources
Adjustment instruction is omited to adjust corresponding operation reserve.And if present flow rate mode is not enough to for operation reserve being adjusted to adapt to work as
Preceding abnormality then needs to change flow rate mode.
It further, is the corresponding operation plan of new flow model selection as shown in fig. 6, if necessary to change flow rate mode
It is slightly the step of, further comprising the steps of:
S6121: the operation reserve under new flow mode described in periodic test；
S6122: judge whether to selected new operation reserve；
S6123: if selected new operation reserve, the operation reserve of selection is executed.
In the present embodiment, it can be determined under new flow mode by the operation reserve under periodic test new flow mode
Adjustment operation reserve whether can adapt to occur unusual condition.Also, it automatically selects or manages under new flow mode
The specified mode of member selects to adapt to the operation reserve of current operating conditions, and after having selected specified operation reserve, executes phase
The operation reserve and the above-mentioned checking process of repetition answered, until selecting most suitable operation reserve.
It further, is the corresponding operation plan of new flow model selection as shown in fig. 7, if necessary to change flow rate mode
Slightly the step of, further include following:
S6124: the policy tag edited is obtained by instrument board；
S6125: according to the policy tag, the corresponding operation reserve of the new flow mode is obtained in policy library；
S6128: enabling the corresponding operation reserve of the new flow mode of acquisition, and disabling present flow rate mode corresponds to
Operation reserve.
In the present embodiment, user can be by UI interface editing policy tag, for example, referring in policy tag settled preceding different
Classification belonging to normal situation, system select corresponding operation reserve further according to the policy tag edited in policy library.This Shen
In the technical solution that please be provide, it by Editing Strategy label, can reduce the matching range of operation reserve, increase matching speed,
Improve Developing Tactics efficiency.After matching corresponding operation reserve, new flow mould can be enabled in the control program of system
The corresponding operation reserve of formula, and the operation reserve run before is disabled.
In addition, obtaining the step of the corresponding operation reserve of the new flow mode in policy library according to the policy tag
Suddenly, comprising the following steps:
S6126: if operation reserve corresponding to the new flow mode has not been obtained from the policy library, according to described
New flow mode creates new strategy；
S6127: the new strategy of creation is updated to the policy library.
In the present embodiment, if operation reserve corresponding to the new flow mode has not been obtained from the policy library,
Determine that control strategy built-in in specified new flow mode is unable to satisfy requirement.At this point, the present embodiment can be new by creating
Strategy adapts to unusual condition, and using new strategy as a kind of policy update under new flow mode to policy library, with after an action of the bowels
It is directly called in continuous analysis.
It, can also be with it should be noted that the health status of AP equipment or wifi system is related to many different performance indicators
Operation reserve adjustment is carried out by adjusting different execution parameter.In technical solution provided by the present application, only with wifi system
AP equipment sends average flow rate and sends busy average counter in system, as the execution parameter of wifi system, for this field
For technical staff, the judgement of unusual condition can also be carried out by other kinds of parameter, or adapts to the execution of unusual condition
Strategy.
Based on above-mentioned wifi system running policy dynamic adjusting method, as shown in figure 8, the application also provides a kind of wifi system
System operation reserve dynamic adjusting device, comprising:
Data collection module 1, for collecting the operation data of current wifi system；
Data aggregate module 2, for carrying out data aggregate to the operation data, and by the operation number after polymerization
According to preservation to database；
Statistical analysis module 3 generates net for doing batch statistics and analysis to the operation data saved in database
Network state of health data and determining present flow rate mode；The network health data include being built according to the operation data
Vertical machine learning model；
Contrast module 4, for comparing the operation data and the machine learning model after polymerizeing；
Abnormal judgment module 5, for determining whether the present flow rate mode exception occurs according to comparing result；
New flow Pattern Matching Module 6, if there is exception for the present flow rate mode, using mode invocation module
New flow mode is matched in flow rate mode library；
Operation reserve adjusts module 7, and the operation reserve for adjusting the wifi system is under the new flow mode
Operation reserve.
From the above technical scheme, the application provides a kind of wifi system running policy dynamic adjusting method and device,
The method first collects the operation data of current wifi system, then carries out data aggregate to operation data, then do to operation data
Batch statistics and analysis generates network health data and determines present flow rate mode, wherein network health data
Including the machine learning model established according to operation data；Operation data and machine learning model after comparing polymerization again, and
According to comparing result, determine whether present flow rate mode exception occurs；If there is exception in present flow rate mode, using mode tune
New flow mode is matched in flow rate mode library with module；The operation reserve of wifi system is adjusted as the operation under new flow mode
Strategy.
Technical solution provided by the present application, by comparing the prediction data and truthful data of machine learning model, dynamic is adjusted
Whole operation reserve, to adapt to the different operating status variation of wifi system, the method for adjustment for solving traditional wifi network system exists
, can not be from preset control strategy when there is new operating status variation, the problem of selecting adaptable control strategy.
Similar portion cross-reference between embodiment provided by the present application, specific embodiment provided above is only
It is several examples under the total design of the application, does not constitute the restriction of the application protection scope.For those skilled in the art
For member, any other embodiment expanded without creative efforts according to application scheme all belongs to
In the protection scope of the application.
Claims (10)
1. a kind of wifi system running policy dynamic adjusting method characterized by comprising
Collect the operation data of current wifi system；
Data aggregate is carried out to the operation data, and the operation data after polymerization is saved to database；
Batch statistics and analysis is done to the operation data saved in database, generates network health data and determination
Present flow rate mode；The network health data include the machine learning model established according to the operation data；
The operation data and the machine learning model after comparison polymerization；
According to comparing result, determine whether the present flow rate mode exception occurs；
If exception occurs in the present flow rate mode, new flow mould is matched in flow rate mode library using mode invocation module
Formula；And the operation reserve of the wifi system is adjusted as the operation reserve under the new flow mode.
2. wifi system running policy dynamic adjusting method according to claim 1, which is characterized in that the method passes through
The Rest API that Swagger is defined collects the operation data of current wifi system, the operation data include at least WAN or
Message byte number and total transmission message byte number are always received on LAN mouth.
3. wifi system running policy dynamic adjusting method according to claim 1, which is characterized in that being protected in database
The operation data deposited does batch statistics and analysis, generates network health data and determines the step of present flow rate mode
Suddenly, comprising:
Scheme program starts the worker thread in worker thread pond；
The data on flows in the operation data by the worker thread, after extracting polymerization；
Corresponding present flow rate mode is determined according to extracted data on flows；
Average value, standard deviation and regression analysis are carried out to the data on flows, establish machine learning model.
4. wifi system running policy dynamic adjusting method according to claim 1, which is characterized in that tied according to comparison
Fruit, determines whether the present flow rate mode abnormal step occurs, comprising:
Extract the calculating parameter in the operation data；
According to the calculating parameter and the machine learning model, model prediction data is determined；
If the operation data and the model prediction data, meets preset condition requirement, judge the present flow rate mode
Occur abnormal；
If the operation data and the model prediction data, it is unsatisfactory for preset condition requirement, judges the present flow rate mould
Formula does not occur exception.
5. wifi system running policy dynamic adjusting method according to claim 1, which is characterized in that if described current
There is exception in flow rate mode, the method also includes:
Worker thread sounds an alarm instruction to exception management module；
The alarm command is sent instrument board by the exception management module, so that the instrument board shows that the alarm refers to
It enables；Alternatively,
Initial configuration notification rule of the exception management module according to wifi system, signal an alert.
6. wifi system running policy dynamic adjusting method according to claim 1, which is characterized in that if described current
There is the step of abnormal, to match new flow mode in flow rate mode library using mode invocation module in flow rate mode, comprising:
Judge under the present flow rate mode exception, if need to change flow rate mode；
It is the corresponding operation reserve of new flow model selection if necessary to change flow rate mode.
7. wifi system running policy dynamic adjusting method according to claim 6, which is characterized in that if necessary to change
The step of flow rate mode, operation reserve corresponding for new flow model selection, comprising:
Operation reserve under new flow mode described in periodic test；
Judge whether to selected new operation reserve；
If selected new operation reserve, the operation reserve of selection is executed.
8. wifi system running policy dynamic adjusting method according to claim 6, which is characterized in that if necessary to change
The step of flow rate mode, operation reserve corresponding for new flow model selection, further includes:
The policy tag edited is obtained by instrument board；
According to the policy tag, the corresponding operation reserve of the new flow mode is obtained in policy library；
Enable the corresponding operation reserve of the new flow mode obtained, and the corresponding operation plan of disabling present flow rate mode
Slightly.
9. wifi system running policy dynamic adjusting method according to claim 8, which is characterized in that according to the strategy
Label, the step of corresponding operation reserve of the new flow mode is obtained in policy library, comprising:
If operation reserve corresponding to the new flow mode has not been obtained from the policy library, according to the new flow mode
Create new strategy；
The new strategy of creation is updated to the policy library.
10. a kind of wifi system running policy dynamic adjusting device characterized by comprising
Data collection module, for collecting the operation data of current wifi system；
Data aggregate module, for being protected to operation data progress data aggregate, and by the operation data after polymerization
It deposits to database；
It is strong to generate network for doing batch statistics and analysis to the operation data saved in database for statistical analysis module
Health status data and determining present flow rate mode；The network health data include being established according to the operation data
Machine learning model；
Contrast module, for comparing the operation data and the machine learning model after polymerizeing；
Abnormal judgment module, for determining whether the present flow rate mode exception occurs according to comparing result；
New flow Pattern Matching Module is being flowed if there is exception for the present flow rate mode using mode invocation module
It measures and matches new flow mode in pattern base；And adjusting the operation reserve of the wifi system is under the new flow mode
Operation reserve.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811167503.2A CN109286526A (en) | 2018-10-08 | 2018-10-08 | A kind of wifi system running policy dynamic adjusting method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811167503.2A CN109286526A (en) | 2018-10-08 | 2018-10-08 | A kind of wifi system running policy dynamic adjusting method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109286526A true CN109286526A (en) | 2019-01-29 |
Family
ID=65176657
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811167503.2A Pending CN109286526A (en) | 2018-10-08 | 2018-10-08 | A kind of wifi system running policy dynamic adjusting method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109286526A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109769280A (en) * | 2019-03-29 | 2019-05-17 | 南京邮电大学 | A kind of WIFI intelligent predicting switching method based on machine learning |
| CN111339048A (en) * | 2020-02-28 | 2020-06-26 | 京东数字科技控股有限公司 | Cache reading amount adjusting method and device, electronic equipment and storage medium |
| CN114257513A (en) * | 2021-12-21 | 2022-03-29 | 中国电信集团系统集成有限责任公司 | Multi-data strategy improvement method and device |
| CN117118907A (en) * | 2023-10-25 | 2023-11-24 | 深圳市亲邻科技有限公司 | Entrance guard flow dynamic monitoring system and method thereof |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101902348A (en) * | 2009-05-25 | 2010-12-01 | 北京启明星辰信息技术股份有限公司 | Network security system and system load automatic adjusting method thereof |
| CN102104611A (en) * | 2011-03-31 | 2011-06-22 | 中国人民解放军信息工程大学 | Promiscuous mode-based DDoS (Distributed Denial of Service) attack detection method and device |
| CN103580960A (en) * | 2013-11-19 | 2014-02-12 | 佛山市络思讯环保科技有限公司 | Online pipe network anomaly detection system based on machine learning |
| CN105721304A (en) * | 2016-04-05 | 2016-06-29 | 网宿科技股份有限公司 | Adaptive routing adjustment method and system and service device |
| EP3046288A1 (en) * | 2013-10-21 | 2016-07-20 | Huawei Technologies Co., Ltd. | Virtual network function network elements management method, device and system |
| CN107612733A (en) * | 2017-09-19 | 2018-01-19 | 杭州安恒信息技术有限公司 | A kind of network audit and monitoring method and its system based on industrial control system |
| CN108574668A (en) * | 2017-03-10 | 2018-09-25 | 北京大学 | A kind of ddos attack peak flow prediction technique based on machine learning |
-
2018
- 2018-10-08 CN CN201811167503.2A patent/CN109286526A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101902348A (en) * | 2009-05-25 | 2010-12-01 | 北京启明星辰信息技术股份有限公司 | Network security system and system load automatic adjusting method thereof |
| CN102104611A (en) * | 2011-03-31 | 2011-06-22 | 中国人民解放军信息工程大学 | Promiscuous mode-based DDoS (Distributed Denial of Service) attack detection method and device |
| EP3046288A1 (en) * | 2013-10-21 | 2016-07-20 | Huawei Technologies Co., Ltd. | Virtual network function network elements management method, device and system |
| CN103580960A (en) * | 2013-11-19 | 2014-02-12 | 佛山市络思讯环保科技有限公司 | Online pipe network anomaly detection system based on machine learning |
| CN105721304A (en) * | 2016-04-05 | 2016-06-29 | 网宿科技股份有限公司 | Adaptive routing adjustment method and system and service device |
| CN108574668A (en) * | 2017-03-10 | 2018-09-25 | 北京大学 | A kind of ddos attack peak flow prediction technique based on machine learning |
| CN107612733A (en) * | 2017-09-19 | 2018-01-19 | 杭州安恒信息技术有限公司 | A kind of network audit and monitoring method and its system based on industrial control system |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109769280A (en) * | 2019-03-29 | 2019-05-17 | 南京邮电大学 | A kind of WIFI intelligent predicting switching method based on machine learning |
| CN109769280B (en) * | 2019-03-29 | 2021-10-08 | 南京邮电大学 | WIFI intelligent prediction switching method based on machine learning |
| CN111339048A (en) * | 2020-02-28 | 2020-06-26 | 京东数字科技控股有限公司 | Cache reading amount adjusting method and device, electronic equipment and storage medium |
| CN114257513A (en) * | 2021-12-21 | 2022-03-29 | 中国电信集团系统集成有限责任公司 | Multi-data strategy improvement method and device |
| CN114257513B (en) * | 2021-12-21 | 2024-02-09 | 中电信数智科技有限公司 | Method and device for perfecting multielement data strategy |
| CN117118907A (en) * | 2023-10-25 | 2023-11-24 | 深圳市亲邻科技有限公司 | Entrance guard flow dynamic monitoring system and method thereof |
| CN117118907B (en) * | 2023-10-25 | 2024-02-02 | 深圳市亲邻科技有限公司 | Entrance guard flow dynamic monitoring system and method thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109286526A (en) | A kind of wifi system running policy dynamic adjusting method and device | |
| CN106708692B (en) | Method and device for establishing filtering alarm model, method and device for filtering alarm and electronic equipment | |
| CN1909516B (en) | Network system, communication relay device, communication terminal device, and program for communication terminal device | |
| EP3136650B1 (en) | Method and system for optimizing network parameters to improve customer satisfaction of network content | |
| CN106468904B (en) | The real-time accumulated quality management and control implementation method of dynamically configurable process | |
| WO2010036885A4 (en) | Wireless mesh network with pinch point and low battery alerts | |
| CN109257304A (en) | A kind of bandwidth adjusting method, device, storage medium and the network equipment | |
| CN103220173A (en) | Alarm monitoring method and alarm monitoring system | |
| CN109918279A (en) | Electronic device, method and storage medium based on daily record data identification user's abnormal operation | |
| CN110891283A (en) | Small base station monitoring device and method based on edge calculation model | |
| CN107148072A (en) | The acquisition methods and system of a kind of target resource parameter of intelligent terminal application | |
| CN109284199A (en) | Server exception processing method, equipment and processor | |
| CN109669835A (en) | MySQL database monitoring method, device, equipment and readable storage medium storing program for executing | |
| CN109840185A (en) | A kind of index correlating method and device | |
| CN107592976A (en) | Interference is modeled | |
| CN104978212B (en) | A kind of unknown crash report method and device | |
| CN108111346A (en) | The definite method, apparatus and storage medium of frequent item set in warning association analysis | |
| CN107547301A (en) | A kind of master/slave device switching method and device | |
| CN105357026B (en) | A kind of resource information collection method and calculate node | |
| CN109974220A (en) | The control method and device of electrical equipment, system, electrical equipment | |
| CN104105097B (en) | Frequency allocation method, frequency configuration device and base station | |
| CN103188730B (en) | System resource load control system, method and device, regulation server equipment | |
| CN109547271A (en) | A kind of network state real time monitoring warning system based on big data | |
| CN103188723A (en) | Measurement scheduling method and device of multimode one-standby user equipment | |
| WO2017059904A1 (en) | Anomaly detection in a data packet access network |
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 |
Application publication date: 20190129 |
|
| RJ01 | Rejection of invention patent application after publication |