Embodiment
The invention will be further described with the following Examples.
Referring to Fig. 1, a kind of efficient network apparatus management system of the present embodiment, for being managed to the network equipment,
Including smart mobile phone 1 and network device management platform 2, the smart mobile phone 1 passes through wireless network and network device management platform 2
Connection, the network device management platform 2 are connected by wireless network with the network equipment, and the smart mobile phone 1 is used to pass through net
The download management interactive software of network device management platform 2, and network is obtained from network device management platform 2 by the interactive software
Condition Prediction of Equipment information and network equipment information, and send administration order, the network equipment to network device management platform 2
For network equipment information to be sent into network device management platform 2, and receive the management life of the transmission of network device management platform 2
Order and execution corresponding operating, the network device management platform 2 are used to be predicted the state of the network equipment, receive network and set
The administration order that the network equipment information and smart mobile phone 1 that preparation is sent are sent, and administration order is sent to the network equipment.
The present embodiment facilitates user to be managed the network equipment, realizes the efficient management of the network equipment.
Preferably, the wireless network is 3G network or 4G networks.
This preferred embodiment improves communication speed, so as to improve network device management efficiency.
Preferably, the network device management platform 2 is carried out by network device state subsystem to network device state
Prediction, the network device state predicting subsystem include the first prediction module, the second prediction module, hybrid predicting module and pre-
Evaluation module is surveyed, first prediction module is used to carry out tentative prediction to the state of the network equipment, obtains tentative prediction result,
Second prediction module is used to carry out re prediction to the state of the network equipment according to tentative prediction result, obtains re prediction
As a result, the hybrid predicting module obtains the final prediction of network device state according to tentative prediction result and re prediction result
As a result, the forecast assessment module is used to evaluate status predication effect.
This preferred embodiment realizes the prediction of network device state and the assessment of prediction effect.
Preferably, the state to the network equipment carries out tentative prediction, carries out in the following ways:
Step 1, assume that time series input is A=(a1,a2,…,al), true desired output is Bt=(b1,b2,…,
bt), the first anticipation function is established based on autoregressive moving-average model time series is predicted:
In formula,Tentative prediction result is represented, n represents Autoregressive, and m represents moving average exponent number, βiRepresent certainly
Regression coefficient, γiRepresent moving average coefficient, { ctRepresent white noise sequence;
Step 2, using AIC criterion to Autoregressive and moving average Order- reduction, using least square method to from returning
Return coefficient and moving average coefficient to be estimated, tentative prediction result is solved:
In formula,Autoregressive estimate is represented,Moving average Order- reduction value is represented,Represent autoregressive coefficient
Estimate,Represent moving average coefficient estimate.
This preferred embodiment carries out tentative prediction to the state of the network equipment, on the one hand, the first prediction module not only can be with
The rule of dynamic data is disclosed, predicts its future value, but also the relevant characteristic of system, the opposing party can be studied from many aspects
Face, the first prediction module can accurately be caught to the linear segment of time series, realize the Accurate Prediction of linear segment.
Preferably, the state to the network equipment carries out re prediction, carries out in the following ways:
The anticipation function of Establishment of Neural Model second is based on the basis of tentative prediction result:
In formula,Represent re prediction result, wqAnd wpqThe connection weight of neutral net is represented, p and q represent nerve
Network input layer and the number of nodes in intermediate layer, d0And d0jRepresent bias term.
The prediction module of this preferred embodiment second can be predicted the non-linear partial of time series well, realize
The Accurate Prediction of non-linear partial.
Preferably, the final prediction knot that network device state is obtained according to tentative prediction result and re prediction result
Fruit, carry out in the following ways:
In formula,Represent final prediction result.
This preferred embodiment hybrid predicting module overcomes generalization ability difference of conventional forecast model during predicting etc. and asked
Topic, the first prediction module and the second prediction module are combined, improve precision of prediction.
Preferably, the forecast assessment module includes early warning submodule and assesses submodule, and the early warning submodule is used for
Network equipment abnormal conditions are carried out with early warning, and the assessment submodule is used to status predication effect is commented according to early warning situation
Valency;
It is described that early warning is carried out to network equipment abnormal conditions, be specially:Shape using hybrid predicting module to the network equipment
State is monitored, and prediction result is obtained by network equipment historical data, when the running status of the network equipment deviates prediction result
Reach certain value, then send the abnormal early warning of the network equipment;
The assessment submodule includes the first evaluation unit, the second evaluation unit and overall merit unit, and described first comments
Valency unit is used to determine the first evaluation points, and second evaluation unit is used to determine the second evaluation points, the overall merit
Unit is used to evaluate status predication effect according to the first evaluation points and the second evaluation points;
The first evaluation points of the determination, are carried out in the following ways:Establish the first evaluation function:
In formula, F1The first evaluation points are represented, N represents all monitoring numbers, N1Represent the number of false-alarm;
The second evaluation points of the determination, are carried out in the following ways:Establish the second evaluation function:
In formula, F1Represent the first evaluation points, N2Represent the number of false dismissal;
It is described that status predication effect is evaluated, carry out in the following ways:Establish composite evaluation function:
In formula, F represents the prediction and evaluation factor, and the prediction and evaluation factor is bigger, and prediction result is more accurate.
This preferred embodiment forecast assessment module realizes the status monitoring to the network equipment according to status predication situation, and
Evaluated using prediction and evaluation factor pair status predication effect, ensure that the accuracy and reliability of prediction effect.
The network equipment is managed using efficient network apparatus management system of the invention, chooses 5 network equipments, point
Not Wei the network equipment 1, the network equipment 2, the network equipment 3, the network equipment 4, the network equipment 5, the efficiency of management and management cost are entered
Row statistics, is compared compared with network apparatus management system, caused to have the beneficial effect that shown in table:
|
The efficiency of management improves |
Management cost reduces |
The network equipment 1 |
29% |
21% |
The network equipment 2 |
27% |
23% |
The network equipment 3 |
26% |
25% |
The network equipment 4 |
25% |
27% |
The network equipment 5 |
24% |
29% |
Finally it should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention, rather than the present invention is protected
The limitation of scope is protected, although being explained with reference to preferred embodiment to the present invention, one of ordinary skill in the art should
Work as understanding, technical scheme can be modified or equivalent substitution, without departing from the reality of technical solution of the present invention
Matter and scope.