CN109995562A - Network traffic prediction technique, device, equipment and medium - Google Patents
Network traffic prediction technique, device, equipment and medium Download PDFInfo
- Publication number
- CN109995562A CN109995562A CN201711493526.8A CN201711493526A CN109995562A CN 109995562 A CN109995562 A CN 109995562A CN 201711493526 A CN201711493526 A CN 201711493526A CN 109995562 A CN109995562 A CN 109995562A
- Authority
- CN
- China
- Prior art keywords
- prediction
- order
- predicted
- network traffic
- prediction result
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 111
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 33
- 230000001932 seasonal effect Effects 0.000 claims abstract description 33
- 238000001914 filtration Methods 0.000 claims abstract description 25
- 230000003044 adaptive effect Effects 0.000 claims abstract description 20
- 238000012545 processing Methods 0.000 claims description 18
- 230000008569 process Effects 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 8
- 230000000737 periodic effect Effects 0.000 claims description 5
- 230000002159 abnormal effect Effects 0.000 claims description 4
- 238000007635 classification algorithm Methods 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 4
- 230000007704 transition Effects 0.000 claims description 4
- 230000000694 effects Effects 0.000 abstract description 12
- 230000004927 fusion Effects 0.000 abstract description 2
- 241001123248 Arma Species 0.000 abstract 1
- 238000004891 communication Methods 0.000 description 20
- 238000010586 diagram Methods 0.000 description 17
- 230000006870 function Effects 0.000 description 10
- 230000002776 aggregation Effects 0.000 description 8
- 238000004220 aggregation Methods 0.000 description 8
- 238000013461 design Methods 0.000 description 8
- 230000008859 change Effects 0.000 description 7
- 238000012937 correction Methods 0.000 description 5
- 238000011161 development Methods 0.000 description 5
- 238000005457 optimization Methods 0.000 description 5
- 238000012360 testing method Methods 0.000 description 5
- 230000000007 visual effect Effects 0.000 description 5
- 230000009471 action Effects 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000001174 ascending effect Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000013178 mathematical model Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000005065 mining Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 238000010521 absorption reaction Methods 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000005311 autocorrelation function Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 235000019580 granularity Nutrition 0.000 description 1
- 230000002401 inhibitory effect Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 210000001503 joint Anatomy 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000007493 shaping process Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
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
- H04L41/14—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
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Algebra (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Mathematical Physics (AREA)
- Probability & Statistics with Applications (AREA)
- Pure & Applied Mathematics (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The embodiment of the invention provides network traffic prediction technique, device, equipment and media, which comprises obtains the historical statistical data in region to be predicted；The historical statistical data is inputted seasonal difference ARMA model to handle, obtains tentative prediction result；The tentative prediction result input adaptive Kalman filter model is handled, the prediction result in the region to be predicted is obtained.The present invention utilizes serial mode fusion application S-ARIMA time series algorithm and the method for adaptive kalman filtering (A-Kalman method) based on Naive Bayes Classification, it is modified using the filtering adjustment quality versus time sequence of A-Kalman method, and the output of the step of Kalman and the N step prediction result seamless connection of time series are got up, the serial relay effect for forming two kinds of algorithms, achievees the effect that multi-step prediction while further increasing Traffic prediction accuracy.
Description
Technical Field
The present invention relates to the technical field of mobile communications, and in particular, to a method, an apparatus, a device, and a medium for predicting network traffic.
Background
With the social development and the continuous improvement of the living standard of people, large-scale gatherings such as increasingly-increasing cultural performances and the like, traveling outside in holidays and the like, a large number of businesses can be generated locally due to the flowing and gathering of a large number of people, a certain impact is caused to a communication network, in order to guarantee the smooth communication experience of users, the guarantee capacity of emergency communication needs to be provided, resources are reasonably allocated according to the change of the business volume, the network is adjusted, and the maximization of the business volume absorption and the minimization of the network congestion degree are guaranteed. Therefore, the service volume prediction is an important premise and basis of emergency communication guarantee, the accurate prediction result can accurately analyze the influence range and the degree, the decision support is provided for reasonably scheduling, managing and distributing emergency guarantee resources, and the emergency communication guarantee is ensured to be orderly and efficiently implemented.
At present, the method commonly used for traffic prediction in actual work is mainly a curve fitting prediction method. The curve fitting prediction method needs to extract a large amount of historical data through OMC, the basic idea of data fitting is carried out by adopting a least square method, the predicted data shows an ascending trend or a descending trend along with the change of time, and no obvious fluctuation exists, and a proper function curve is found to reflect the change trend. Due to the fact that the predicted data show large fluctuation due to the fact that network technology development and market strategy adjustment are combined with the influence of a plurality of external factors, the prediction effect of the fitting curve is poor, and the influence of various emergencies on the development trend is difficult to reflect.
Disclosure of Invention
The embodiment of the invention provides a method, a device, equipment and a medium for predicting network traffic, which are used for solving the problem that the traditional traffic prediction result is inaccurate.
In a first aspect, an embodiment of the present invention provides a method for predicting network traffic, where the method includes:
acquiring historical statistical data of an area to be predicted;
inputting the historical statistical data into a seasonal difference autoregressive moving average model for processing to obtain a preliminary prediction result;
and inputting the preliminary prediction result into an adaptive Kalman filtering model for processing to obtain a prediction result of the area to be predicted.
In a second aspect, an embodiment of the present invention provides a device for predicting network traffic, where the device includes:
the historical statistical data acquisition module is used for acquiring historical statistical data of the area to be predicted;
the preliminary prediction result acquisition module is used for inputting the historical statistical data into a seasonal difference autoregressive moving average model for processing to obtain a preliminary prediction result;
and the final prediction result acquisition module is used for inputting the preliminary prediction result into an adaptive Kalman filtering model for processing to obtain the prediction result of the area to be predicted.
The embodiment of the invention provides a network traffic prediction device, which comprises: at least one processor, at least one memory, and computer program instructions stored in the memory, which when executed by the processor, implement the method of the first aspect of the embodiments described above.
In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which computer program instructions are stored, which, when executed by a processor, implement the method of the first aspect in the foregoing embodiments.
According to the network traffic prediction method, the device, the equipment and the medium provided by the embodiment of the invention, the S-ARIMA time sequence algorithm and the naive Bayesian classification-based adaptive Kalman filtering method (A-Kalman method) are fused and applied in a serial mode, the time sequence is corrected by using the filtering adjustment characteristic of the A-Kalman method, the further output of Kalman and the N-step prediction results of the time sequence are seamlessly linked together, the serial relay effect of the two algorithms is formed, the traffic prediction accuracy is further improved, and the multi-step prediction effect is achieved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required to be used in the embodiments of the present invention will be briefly described below, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a flow chart illustrating a method for predicting network traffic provided by an embodiment of the present invention;
fig. 2 is a flow chart of a method for predicting network traffic according to an embodiment of the present invention;
FIG. 3 is a block diagram illustrating a schematic design flow of a method for predicting network traffic according to an embodiment of the present invention;
fig. 4 is a flowchart illustrating an S-ARIMA algorithm in a network traffic prediction method according to an embodiment of the present invention;
fig. 5 is a flowchart illustrating an a-Kalman algorithm in the network traffic prediction method according to the embodiment of the present invention;
FIG. 6 is a system design architecture diagram illustrating a network traffic prediction method provided by an embodiment of the present invention;
fig. 7 is a system login interface diagram of a network traffic prediction method according to an embodiment of the present invention.
FIG. 8 is a schematic diagram illustrating an output process interface of a network traffic prediction method according to an embodiment of the present invention;
FIG. 9 is a schematic diagram illustrating an output interface of a network traffic prediction method according to an embodiment of the present invention;
fig. 10 is a block diagram of a network traffic prediction apparatus provided by an embodiment of the present invention;
fig. 11 is a schematic diagram illustrating a hardware structure of a network traffic prediction device according to an embodiment of the present invention.
Detailed Description
Features and exemplary embodiments of various aspects of the present invention will be described in detail below, and in order to make objects, technical solutions and advantages of the present invention more apparent, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not to be construed as limiting the invention. It will be apparent to one skilled in the art that the present invention may be practiced without some of these specific details. The following description of the embodiments is merely intended to provide a better understanding of the present invention by illustrating examples of the present invention.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
At present, the method commonly used for traffic prediction in actual work is mainly a curve fitting prediction method. The curve fitting prediction method needs to extract a large amount of historical data through OMC, the basic idea of data fitting is carried out by adopting a least square method, the predicted data shows an ascending trend or a descending trend along with the change of time, and no obvious fluctuation exists, and a proper function curve is found to reflect the change trend. Due to the fact that the predicted data show large fluctuation due to the fact that network technology development and market strategy adjustment are combined with the influence of a plurality of external factors, the prediction effect of the fitting curve is poor, and the influence of various emergencies on the development trend is difficult to reflect.
The main flow of curve fitting traffic prediction is as follows:
determining a curve function to fit the known data: the types of curve functions commonly used include straight lines, polynomials, hyperbolas (one), exponential curves, trigonometric functions and the like, and which curve function is used for data fitting is preliminarily confirmed through a large amount of historical data operation and drawing observation. A large amount of complex data calculation is needed to select a proper curve function, subjective judgment errors exist in drawing observation, and efficient, real-time and accurate prediction is difficult to achieve.
Determining coefficients in the set of fitting functions: and obtaining a linear equation set related to the coefficient by using the necessary condition of the extreme value, and solving the coefficient in the fitting function set.
Data fitting was performed with the help of existing software: and (5) realizing output comparison of the fitting curve and the original data curve through matlab, and judging the accuracy of the prediction result.
The defects of the existing prediction method comprise: 1. the fitting curve generally has the smooth characteristic, and is matched and fitted through a large number of historical data samples, so that the historical data trend is mainly reflected, the accuracy of future data prediction is insufficient, and emergency communication guarantee and daily network optimization cannot be effectively supported; 2. the existing method does not form an automation and visualization system support, lacks an IT means system, and is difficult to realize efficient and real-time prediction and visual presentation.
The invention provides a serial combined traffic prediction method and a serial combined traffic prediction system, which are characterized in that an innovative serial mode is fused with an S-ARIMA time sequence algorithm and an adaptive Kalman filtering method (A-Kalman method) based on naive Bayesian classification, the time sequence is corrected by using the filtering adjustment characteristic of the A-Kalman method, the further output of Kalman and the N-step prediction result of the time sequence are seamlessly connected in an N-step circulation mode, the serial relay effect of the two algorithms is formed, the traffic prediction accuracy is further improved, the multi-step prediction effect is achieved, and effective data support and sufficient time resources are provided for the establishment and implementation of emergency communication guarantee plans. Secondly, automatic and visual innovative design is realized based on the method, real-time presentation and geographical aggregation of prediction services are realized in multiple dimensions and multiple layers, individual prediction fluctuation and group prediction aggregation are included, network bearing conditions and future trends are visually displayed, a network 'fortune telling' system is constructed, and system support is provided for emergency communication guarantee work. Finally, a big data association mining mode is innovated, big data analysis based on traffic prediction is achieved, possible network congestion nodes are located, and a direct optimization basis is provided for emergency communication guarantee and even daily network optimization.
Fig. 1 shows a flowchart of a network traffic prediction method provided by an embodiment of the present invention, and as described in fig. 1, the method includes:
step S10, obtaining historical statistical data of the area to be predicted.
And step S20, inputting the historical statistical data into a seasonal difference autoregressive moving average model for processing to obtain a preliminary prediction result.
And step S30, inputting the preliminary prediction result into an adaptive Kalman filtering model for processing to obtain the prediction result of the area to be predicted.
Specifically, in the embodiment, an S-ARIMA time series algorithm and a naive bayesian classification based adaptive Kalman filtering method (a-Kalman method) are applied in a serial manner in a fusion manner, the time series is corrected by using the filtering adjustment characteristic of the a-Kalman method, and the further output of Kalman is seamlessly linked with the N-step prediction result of the time series in an N-step circulation manner, so that the serial relay effect of the two algorithms is formed, the accuracy of traffic prediction is further improved, the multi-step prediction effect is achieved, and effective data support and sufficient time resources are provided for the establishment and implementation of emergency communication guarantee plans. Fig. 2 is a flowchart illustrating a method for predicting network traffic according to an embodiment of the present invention.
Secondly, automatic and visual innovative design is realized based on the method, real-time presentation and geographical aggregation of prediction services are realized in multiple dimensions and multiple layers, individual prediction fluctuation and group prediction aggregation are included, network bearing conditions and future trends are visually displayed, a network 'fortune telling' system is constructed, and system support is provided for emergency communication guarantee work. Finally, a big data association mining mode is innovated, big data analysis based on traffic prediction is achieved, possible network congestion nodes are located, and a direct optimization basis is provided for emergency communication guarantee and even daily network optimization.
In one possible implementation, obtaining historical statistical data of an area to be predicted includes: acquiring historical original data of an area to be predicted; and deleting abnormal data and missing data in the historical original data to obtain historical statistical data of the area to be predicted.
Specifically, the obtained historical original data of the area to be predicted is sorted, abnormal data and missing data are deleted, and the obtained effective data are used as historical statistical data of the area to be predicted.
In one possible implementation, the seasonal differential autoregressive moving average model is constructed as follows:
wherein t represents time or the output step number of the prediction algorithm;
b is a backshifting operator, subject to Bnyt＝yt-n，
N is a preliminary prediction limiting step number and is a constant;
it is shown that a difference of order d is made,representation versus time series ytD-order difference:
performing a difference process when d is 1, namely commanding
d 2 is subjected to secondary difference processing, namely, order
The seasonal difference of the order D is represented,representation versus time series ytD-order periodic difference and D-order seasonal difference with the length of s are carried out;
and thetaq(B) Representing an autoregressive of order P and a moving average of order Q, the order P being the seasonal autoregressive order, Q being the seasonal moving average order, phiP(Bs) And ΘQ(Bs) Expressing seasonal P-order autoregressive operators and Q-order moving average operators;
μtthe white noise is zero mean white noise, and is set as Gaussian white noise in the algorithm implementation.
In one possible implementation, the adaptive kalman filtering model is:
Pk＝(I-KkHk)Pk，k-1
in the formula:for noise estimation factor, 0<b<1, b is a noise control variable, and is judged and selected to be 0.9 or 0.95 by a Bayes classification algorithm;
representing the optimal estimated value of the network traffic at the k moment;
representing a one-step predicted value of the network traffic predicted by the optimal estimated value at the k-1 moment;
Kkrepresenting the filter gain;
the representation is the difference (i.e. innovation) between the observed value and the predicted value of the network traffic and the mean value of the noise;
representing a system state transition matrix which is a system control parameter;
Pk,k-1representing a network traffic one-step covariance prediction value predicted by the optimal covariance at the k-1 moment;
andall represent noise and represent errors in network traffic due to uncontrollable causes.
Specifically, the variation of the traffic is influenced by various factors, and in order to avoid a large error in prediction, combined prediction is considered as the best method for improving accuracy. The commonly used combined prediction method is that a plurality of methods are applied in parallel, for example, the methods are linearly combined through fixed weight or time-varying weight, so that the weighted average of a plurality of prediction results can be achieved, and the inaccuracy factors of the methods are further weakened, which is the advantage of the combined prediction.
The time sequence prediction algorithm (S-ARIMA) with high prediction accuracy, good real-time performance and low algorithm complexity and the adaptive Kalman filtering method (A-Kalman method) based on naive Bayesian classification are selected for combined prediction, the combined mode of the embodiment innovatively adopts a serial combined mode, the time sequence is corrected by using the filtering adjustment characteristic of the Kalman algorithm, and one-step output of Kalman and an N-step prediction result of the time sequence are seamlessly linked together in an N-step circulation mode to form a serial relay effect of the two algorithms.
Firstly, determining a research object and a variable as mobile network traffic, collecting historical data of the variable by using a network management system, setting a prediction step number range (N < 10 >) according to the divergence degree of a preliminary test and a prediction result, and performing preliminary traffic prediction by adopting an S-ARIMA model; secondly, secondary correction is carried out on the basis of a prediction result, an improved adaptive Kalman filtering algorithm is adopted for filtering and shaping, a state equation and a measurement equation are introduced, system noise and measurement noise are effectively processed, and the prediction accuracy is further improved. Fig. 3 shows a schematic design flow chart of a network traffic prediction method provided by an embodiment of the present invention.
Preliminary prediction using S-ARIMA model
The S-ARIMA model (seasonal differential autoregressive moving average model) is a time sequence prediction analysis method, is derived from the autoregressive moving average model, can adopt a BOX-Jenkins model identification, estimation and prediction program, so that the model can be conveniently adjusted in real time along with the acquisition of more historical data, the prediction precision of the model can be ensured, the model is easily applied to real-time prediction, but the prediction error is gradually increased along with the increase of the prediction steps, a step limit concept is innovatively introduced to avoid error transmission, and the S-ARIMA model is improved through a limiting method of outputting the step number N so as to ensure the output accuracy of the serial combination prediction method.
In the method, firstly, an S-ARIMA model is constructed to carry out preliminary prediction on network traffic, so that the traffic is taken as a research object and a variable (y)t) And acquiring day-granularity scenic spot traffic data of 30 months in total from 2015 month 1 to 2017 month 6 as a historical sample sequence through a network management system. The traffic of scenic spots changes trending and seasonally, and needs to be predictedCarrying out periodic difference and seasonal difference on the data, and eliminating the sequence trend after first-order difference through prediction, wherein d is 1; after the first-order seasonal difference, the seasonal trend is eliminated, and D is 1.
The S-ARIMA model estimates parameters such as P, Q, and P, Q, performs model identification on a provisional initial model (i.e., determining model parameters D-1, P-1, Q-1, and S-12) according to the autocorrelation and partial autocorrelation function of the sequence after the difference, obtains an initial estimation value of the model, and then performs an adaptive test on the remaining sum of squares of the provisional model obtained by estimation to determine whether to accept the provisional model. When the adaptability test shows that the tentative model is not the optimal model, the improved model can be re-fitted according to the information provided by the test about the improvement of the model, and the adaptability test is performed on the improved model, so as to finally obtain the optimal model S-ARIMA (1,1,1) (0,1,2) 12. Compared with the actual value, the model predicts the prediction values from 2016 (1 month) to 2017 (6 months), the prediction errors of 2016 (whole year) and 2017 (last half year) are respectively 3.21% and 5.79%, the errors of the 2016 (whole year) and the 2017 (last half year) are less than 10%, and the prediction errors increase along with the increase of prediction time.
The method constructs an S-ARIMA (P, D, Q) (P, D, Q) S mathematical model as follows:
wherein t represents time or the output step number of the prediction algorithm;
b is a backshifting operator, subject to Bnyt＝yt-nn is a natural number;
it is shown that a difference of order d is made,representation versus time series ytD-order difference:
performing a difference process when d is 1, namely commanding
d 2 is subjected to secondary difference processing, namely, order
The seasonal difference of the order D is represented,representation versus time series ytD-order periodic difference and D-order seasonal difference with the length of s are carried out;
and thetaq(B) Representing an autoregressive of order P and a moving average of order Q, the order P being the seasonal autoregressive order, Q being the seasonal moving average order, phiP(Bs) And ΘQ(Bs) Expressing seasonal P-order autoregressive operators and Q-order moving average operators;
μtthe above algorithm implementation is set to gaussian white noise for zero mean white noise. Fig. 4 shows a flowchart of the S-ARIMA algorithm in the network traffic prediction method provided by the embodiment of the present invention.
Predictive correction using a naive bayesian classification based adaptive Kalman filter model (a-Kalman)
The method is characterized in that a good prediction effect can be obtained by applying an S-ARIMA-based model to real-time prediction of network traffic, but prediction errors can be generated as long as prediction is carried out, and in order to minimize the prediction errors, the method aims at the S-ARIMA prediction result and further corrects the prediction result by using an adaptive Kalman filtering method based on naive Bayesian classification. The method combines naive Bayes classification with a Kalman filtering algorithm based on noise estimation, estimates noise in real time, distinguishes a vacation period from a non-vacation period of an S-ARIMA model prediction result by using the naive Bayes classification, adaptively controls and adjusts a noise estimation factor, dynamically changes the weight of old data, and corrects the statistical characteristics of system noise and observation noise (factors influencing traffic variation, such as uncertainty of user access, randomness of channel quality, difference of user service and the like, are elements forming noise), thereby achieving the purposes of reducing system model errors, inhibiting filtering divergence and improving filtering precision.
The improved mathematical model of the Kalman filtering algorithm is as follows:
the traffic one-step predicted value isThe one-step prediction recurrence formula is as follows:
the estimation error covariance is Pk，k-1The recurrence formula is:
kalman gain of KkThe recursion calculation formula is:
from an observed variable zkUpdating a predicted value:
updating the error covariance: pk＝(I-KkHk)Pk，k-1
Wherein,andall are noise variables, and are obtained by recursive calculation of the following noise estimation formula:
in the formula:for noise estimation factor, 0<b<1, b is a noise control variable, and is judged and selected to be 0.9 or 0.95 by a Bayes classification algorithm.
Representing the optimal estimated value of the network traffic at the k moment;
representing a one-step predicted value of the network traffic predicted by the optimal estimated value at the k-1 moment;
Kkrepresenting the filter gain;
the representation is the difference (i.e. innovation) between the observed value and the predicted value of the network traffic and the mean value of the noise;
representing a system state transition matrix which is a system control parameter;
Pk,k-1representing a network traffic one-step covariance prediction value predicted by the optimal covariance at the k-1 moment;
andall the noise represents errors caused by uncontrollable reasons in network flow;
fig. 5 shows a flowchart of an a-Kalman algorithm in the network traffic prediction method provided by the embodiment of the present invention.
In one possible implementation, the method further includes: and determining an early warning result according to the prediction result and the early warning threshold value.
In one possible implementation, the method further includes:
and performing multi-dimensional presentation on the prediction result and the early warning result, wherein the multi-dimensional presentation comprises presentation of group cell prediction results and presentation of single cell prediction results.
Specifically, in order to visually present the prediction result of the method and efficiently support emergency communication guarantee work, the automatic and visual system is designed, a multi-view integration idea is adopted, the rules and influences of presentation of the same data under different dimensions are displayed, and key data information is conveniently and timely captured. The prediction result of the method is subjected to multi-dimensional and multi-level real-time presentation and geographical aggregation, such as single-cell and single-region individualized prediction service fluctuation presentation and guarantee time interval early warning, multi-cell and continuous region group prediction service aggregation, the network bearing condition and future trend change are visually displayed in real time by adopting a prediction thermodynamic diagram, a network 'fortune telling' system is constructed, and effective system support is provided for emergency communication guarantee work. Fig. 6 is a system design architecture diagram illustrating a network traffic prediction method according to an embodiment of the present invention.
A system design interface demonstration diagram, and fig. 7 shows a system login interface schematic diagram of the network traffic prediction method provided by the embodiment of the invention. As shown in fig. 7, the prediction system of the present invention uses a user account + password to log in for use, so as to ensure the security of the system, information and patent.
Condition input interface: the method supports manual and automatic input modes, single cell and batch cell input, and supports a discrete time sequence of data conversion of various time granularities.
Outputting a process interface: because the method belongs to a combined algorithm and needs a preprocessing process of an early original time sequence, a certain time is needed for prediction operation. Fig. 8 is a schematic diagram illustrating an output process interface of a network traffic prediction method according to an embodiment of the present invention. The output process interface shown in fig. 8 enhances the user's perception experience. Fig. 9 is a schematic diagram illustrating an output interface of a network traffic prediction method according to an embodiment of the present invention.
The invention adopts an innovative serial combination mode to fuse and apply S-ARIMA and an improved Kalman filtering algorithm, can adjust and correct the model in real time along with the acquisition of more historical data, ensures the prediction accuracy and provides a theoretical basis for long-term prediction. The method is characterized in that a time sequence algorithm is adopted for preliminary prediction, accuracy is in a descending trend along with the increase of prediction time, a step limiting concept is provided for reducing error transmission, a limiting variable is set to control the prediction length, and meanwhile butt joint and combination with a subsequent correction algorithm are facilitated. The correction algorithm is an adaptive Kalman filtering algorithm based on naive Bayes classification, and noise estimation factors are adaptively controlled by using the Bayes classification, so that noise is adaptively adjusted, and the correction accuracy of a preliminary prediction result is improved. The invention forms an automatic prediction system through automatic and visual innovative design, realizes real-time presentation and geographical aggregation of prediction services in multiple dimensions and multiple layers, individually predicts service fluctuation of a single cell and a single region, and massively predicts service aggregation of multiple cells and connected regions, and predicts the thermodynamic diagram to visually display the network bearing condition and the future trend in real time.
Fig. 10 is a block diagram of a network traffic prediction apparatus provided in an embodiment of the present invention, and as shown in fig. 10, the apparatus includes:
a historical statistical data obtaining module 61, configured to obtain historical statistical data of the area to be predicted;
a preliminary prediction result obtaining module 62, configured to input the historical statistical data into a seasonal difference auto-regression moving average model for processing, so as to obtain a preliminary prediction result;
and a final prediction result obtaining module 63, configured to input the preliminary prediction result into an adaptive kalman filter model for processing, so as to obtain a prediction result of the region to be predicted.
In a possible implementation manner, the historical statistical data obtaining module 61 includes:
the original data acquisition submodule is used for acquiring historical original data of the area to be predicted;
and the data sorting submodule is used for deleting abnormal data and missing data in the historical original data to obtain historical statistical data of the area to be predicted.
In one possible implementation, the preliminary prediction result obtaining module 62 constructs the seasonal differential autoregressive moving average model as follows:
wherein t represents time or the output step number of the prediction algorithm;
b is a backshifting operator, subject to Bnyt＝yt-n
N is a preliminary prediction limiting step number and is a constant;
it is shown that a difference of order d is made,representation versus time series ytD-order difference:
performing a difference process when d is 1, namely commanding
d 2 is subjected to secondary difference processing, namely, order
The seasonal difference of the order D is represented,representation versus time series ytPerforming D-order periodic difference and D-order length ofsSeasonal differences of (c);
and thetaq(B) Representing an autoregressive of order P and a moving average of order Q, the order P being the seasonal autoregressive order, Q being the seasonal moving average order, phiP(Bs) And ΘQ(Bs) Expressing seasonal P-order autoregressive operators and Q-order moving average operators;
μtthe white noise is zero mean white noise, and is set as Gaussian white noise in the algorithm implementation.
In a possible implementation manner, the final prediction result obtaining module 63 constructs the adaptive kalman filter model as follows:
Pk＝(I-KkHk)Pk，k-1
in the formula:b is a noise estimation factor, b is more than 0 and less than 1, b is a noise control variable, and 0.9 or 0.95 is selected by judgment of a Bayesian classification algorithm;
representing the optimal estimated value of the network traffic at the k moment;
representing a one-step predicted value of the network traffic predicted by the optimal estimated value at the k-1 moment;
Kkrepresenting the filter gain;
the representation is the difference (i.e. innovation) between the observed value and the predicted value of the network traffic and the mean value of the noise;
representing a system state transition matrix which is a system control parameter;
Pk,k-1display unitPredicting the network traffic one-step covariance predicted value through the optimal covariance at the moment of k-1;
andall represent noise and represent errors in network traffic due to uncontrollable causes.
In one possible implementation, the apparatus further includes:
and the early warning module is used for determining an early warning result according to the prediction result and the early warning threshold value.
In one possible implementation, the apparatus further includes:
and the presentation module is used for carrying out multi-dimensional presentation on the prediction result and the early warning result, wherein the multi-dimensional presentation comprises presentation of prediction results of group cells and presentation of prediction results of single cells.
In addition, the network traffic prediction method of the embodiment of the present invention described in conjunction with fig. 1 may be implemented by a network traffic prediction apparatus. Fig. 11 is a schematic diagram illustrating a hardware structure of a network traffic prediction device according to an embodiment of the present invention.
The network traffic prediction device may include a processor 401 and a memory 402 storing computer program instructions.
Specifically, the processor 401 may include a Central Processing Unit (CPU), or an Application Specific Integrated Circuit (ASIC), or may be configured as one or more Integrated circuits implementing embodiments of the present invention.
Memory 402 may include mass storage for data or instructions. By way of example, and not limitation, memory 402 may include a Hard Disk Drive (HDD), floppy Disk Drive, flash memory, optical Disk, magneto-optical Disk, tape, or Universal Serial Bus (USB) Drive or a combination of two or more of these. Memory 402 may include removable or non-removable (or fixed) media, where appropriate. The memory 402 may be internal or external to the data processing apparatus, where appropriate. In a particular embodiment, the memory 402 is a non-volatile solid-state memory. In a particular embodiment, the memory 402 includes Read Only Memory (ROM). Where appropriate, the ROM may be mask-programmed ROM, Programmable ROM (PROM), Erasable PROM (EPROM), Electrically Erasable PROM (EEPROM), electrically rewritable ROM (EAROM), or flash memory or a combination of two or more of these.
The processor 401 may implement any of the network traffic prediction methods in the above embodiments by reading and executing computer program instructions stored in the memory 402.
In one example, the network traffic prediction device may also include a communication interface 403 and a bus 410. As shown in fig. 11, the processor 401, the memory 402, and the communication interface 403 are connected by a bus 410 to complete communication therebetween.
The communication interface 403 is mainly used for implementing communication between modules, apparatuses, units and/or devices in the embodiments of the present invention.
Bus 410 comprises hardware, software, or both coupling the components of the network traffic prediction device to each other. By way of example, and not limitation, a bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hypertransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an infiniband interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a video electronics standards association local (VLB) bus, or other suitable bus or a combination of two or more of these. Bus 410 may include one or more buses, where appropriate. Although specific buses have been described and shown in the embodiments of the invention, any suitable buses or interconnects are contemplated by the invention.
In addition, in combination with the network traffic prediction method in the foregoing embodiment, the embodiment of the present invention may provide a computer-readable storage medium to implement. The computer readable storage medium having stored thereon computer program instructions; the computer program instructions, when executed by a processor, implement any of the network traffic prediction methods in the above embodiments.
It is to be understood that the invention is not limited to the specific arrangements and instrumentality described above and shown in the drawings. A detailed description of known methods is omitted herein for the sake of brevity. In the above embodiments, several specific steps are described and shown as examples. However, the method processes of the present invention are not limited to the specific steps described and illustrated, and those skilled in the art can make various changes, modifications and additions or change the order between the steps after comprehending the spirit of the present invention.
The functional blocks shown in the above-described structural block diagrams may be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, it may be, for example, an electronic circuit, an Application Specific Integrated Circuit (ASIC), suitable firmware, plug-in, function card, or the like. When implemented in software, the elements of the invention are the programs or code segments used to perform the required tasks. The program or code segments may be stored in a machine-readable medium or transmitted by a data signal carried in a carrier wave over a transmission medium or a communication link. A "machine-readable medium" may include any medium that can store or transfer information. Examples of a machine-readable medium include electronic circuits, semiconductor memory devices, ROM, flash memory, Erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, Radio Frequency (RF) links, and so forth. The code segments may be downloaded via computer networks such as the internet, intranet, etc.
It should also be noted that the exemplary embodiments mentioned in this patent describe some methods or systems based on a series of steps or devices. However, the present invention is not limited to the order of the above-described steps, that is, the steps may be performed in the order mentioned in the embodiments, may be performed in an order different from the order in the embodiments, or may be performed simultaneously.
As described above, only the specific embodiments of the present invention are provided, and it can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the system, the module and the unit described above may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again. It should be understood that the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive various equivalent modifications or substitutions within the technical scope of the present invention, and these modifications or substitutions should be covered within the scope of the present invention.
Claims (9)
1. A method for predicting network traffic, the method comprising:
acquiring historical statistical data of an area to be predicted;
inputting the historical statistical data into a seasonal difference autoregressive moving average model for processing to obtain a preliminary prediction result;
and inputting the preliminary prediction result into an adaptive Kalman filtering model for processing to obtain a prediction result of the area to be predicted.
2. The method of claim 1, wherein obtaining historical statistics of the area to be predicted comprises:
acquiring historical original data of an area to be predicted;
and deleting abnormal data and missing data in the historical original data to obtain historical statistical data of the area to be predicted.
3. The method of claim 1, wherein the seasonal differential autoregressive moving average model is constructed as follows:
wherein t represents time or the output step number of the prediction algorithm; b is a backshifting operator, subject to Bnyt＝yt-nN is a preliminary prediction limiting step number and is a constant;
it is shown that a difference of order d is made,representation versus time series ytD-order difference:
performing a difference process when d is 1, namely commanding
d 2 is subjected to secondary difference processing, namely, order
The seasonal difference of the order D is represented,representation versus time series ytD-order periodic difference and D-order seasonal difference with the length of s are carried out;
and thetaq(B) Representing an autoregressive of order P and a moving average of order Q, the order P being the seasonal autoregressive order, Q being the seasonal moving average order, phiP(Bs) And ΘQ(Bs) Expressing seasonal P-order autoregressive operators and Q-order moving average operators;
μtthe white noise is zero mean white noise, and is set as Gaussian white noise in the algorithm implementation.
4. The method of claim 1, wherein the adaptive kalman filter model is:
in the formula:for noise estimation factor, 0<b<1, b is a noise control variable, and is judged and selected to be 0.9 or 0.95 by a Bayes classification algorithm;
representing the optimal estimated value of the network traffic at the k moment;
representing a one-step predicted value of the network traffic predicted by the optimal estimated value at the k-1 moment;
Kkrepresenting the filter gain;
the representation is the difference (i.e. innovation) between the observed value and the predicted value of the network traffic and the mean value of the noise;
representing a system state transition matrix which is a system control parameter;
Pk,k-1representing a network traffic one-step covariance prediction value predicted by the optimal covariance at the k-1 moment;
andall represent noise and represent errors in network traffic due to uncontrollable causes.
5. The method of claim 1, further comprising:
and determining an early warning result according to the prediction result and the early warning threshold value.
6. The method of claim 5, further comprising:
and performing multi-dimensional presentation on the prediction result and the early warning result, wherein the multi-dimensional presentation comprises presentation of group cell prediction results and presentation of single cell prediction results.
7. An apparatus for predicting network traffic, the apparatus comprising:
the historical statistical data acquisition module is used for acquiring historical statistical data of the area to be predicted;
the preliminary prediction result acquisition module is used for inputting the historical statistical data into a seasonal difference autoregressive moving average model for processing to obtain a preliminary prediction result;
and the final prediction result acquisition module is used for inputting the preliminary prediction result into an adaptive Kalman filtering model for processing to obtain the prediction result of the area to be predicted.
8. A network traffic prediction apparatus, comprising: at least one processor, at least one memory, and computer program instructions stored in the memory that, when executed by the processor, implement the method of any of claims 1-6.
9. A computer-readable storage medium having computer program instructions stored thereon, which when executed by a processor implement the method of any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711493526.8A CN109995562B (en) | 2017-12-30 | 2017-12-30 | Network traffic prediction method, device, equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711493526.8A CN109995562B (en) | 2017-12-30 | 2017-12-30 | Network traffic prediction method, device, equipment and medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109995562A true CN109995562A (en) | 2019-07-09 |
| CN109995562B CN109995562B (en) | 2022-11-08 |
Family
ID=67110174
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711493526.8A Active CN109995562B (en) | 2017-12-30 | 2017-12-30 | Network traffic prediction method, device, equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109995562B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110851333A (en) * | 2019-11-14 | 2020-02-28 | 北京金山云网络技术有限公司 | Monitoring method and device of root partition and monitoring server |
| CN110880984A (en) * | 2019-10-18 | 2020-03-13 | 平安科技（深圳）有限公司 | Model-based flow anomaly monitoring method, device, equipment and storage medium |
| CN111475772A (en) * | 2020-03-27 | 2020-07-31 | 微梦创科网络科技（中国）有限公司 | Capacity evaluation method and device |
| CN111723340A (en) * | 2020-06-09 | 2020-09-29 | 武汉理工大学 | Method for correcting phase jump caused by polarization-induced fading in optical fiber phase demodulation |
| CN112019398A (en) * | 2020-07-16 | 2020-12-01 | 中国人民财产保险股份有限公司 | Network flow prediction method and device |
| CN112866129A (en) * | 2021-03-31 | 2021-05-28 | 中国工商银行股份有限公司 | Method, device and system for packing by automatically adjusting intervals according to aging marks |
| CN113672642A (en) * | 2021-07-07 | 2021-11-19 | 德特赛维技术有限公司 | Method and system for predicting mobile network performance based on time sequence |
| CN114760191A (en) * | 2022-05-24 | 2022-07-15 | 咪咕文化科技有限公司 | Data service quality early warning method, system, device and readable storage medium |
| WO2023165145A1 (en) * | 2022-03-02 | 2023-09-07 | 北京沃东天骏信息技术有限公司 | Time sequence traffic prediction method and apparatus, storage medium, and electronic device |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0682080A (en) * | 1992-09-01 | 1994-03-22 | Matsushita Refrig Co Ltd | Load estimating device |
| US20030060692A1 (en) * | 2001-08-03 | 2003-03-27 | Timothy L. Ruchti | Intelligent system for detecting errors and determining failure modes in noninvasive measurement of blood and tissue analytes |
| CN102034350A (en) * | 2009-09-30 | 2011-04-27 | 北京四通智能交通系统集成有限公司 | Short-time prediction method and system of traffic flow data |
| CN102519448A (en) * | 2011-12-16 | 2012-06-27 | 浙江大学 | Quick north-finding method based on optical fiber gyroscope |
| CN102650527A (en) * | 2012-05-25 | 2012-08-29 | 北京航空航天大学 | Temperature compensation method for denoising fiber-optic gyroscope on basis of time series analysis |
| CN103559563A (en) * | 2013-11-20 | 2014-02-05 | 国家电网公司 | Method for predicting wind speed of wind power plant at short term |
| CN103605906A (en) * | 2013-12-03 | 2014-02-26 | 天津工业大学 | Wind speed prediction method based on H filtering |
| CN103745272A (en) * | 2014-01-06 | 2014-04-23 | 国家电网公司 | Power short-term load predicating method based on fast periodic component extraction |
| CN104517159A (en) * | 2014-12-18 | 2015-04-15 | 上海交通大学 | Method for predicting short-time passenger flow of bus |
| CN105321345A (en) * | 2015-09-18 | 2016-02-10 | 浙江工业大学 | Road traffic flow prediction method based on ARIMA model and kalman filtering |
-
2017
- 2017-12-30 CN CN201711493526.8A patent/CN109995562B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0682080A (en) * | 1992-09-01 | 1994-03-22 | Matsushita Refrig Co Ltd | Load estimating device |
| US20030060692A1 (en) * | 2001-08-03 | 2003-03-27 | Timothy L. Ruchti | Intelligent system for detecting errors and determining failure modes in noninvasive measurement of blood and tissue analytes |
| CN102034350A (en) * | 2009-09-30 | 2011-04-27 | 北京四通智能交通系统集成有限公司 | Short-time prediction method and system of traffic flow data |
| CN102519448A (en) * | 2011-12-16 | 2012-06-27 | 浙江大学 | Quick north-finding method based on optical fiber gyroscope |
| CN102650527A (en) * | 2012-05-25 | 2012-08-29 | 北京航空航天大学 | Temperature compensation method for denoising fiber-optic gyroscope on basis of time series analysis |
| CN103559563A (en) * | 2013-11-20 | 2014-02-05 | 国家电网公司 | Method for predicting wind speed of wind power plant at short term |
| CN103605906A (en) * | 2013-12-03 | 2014-02-26 | 天津工业大学 | Wind speed prediction method based on H filtering |
| CN103745272A (en) * | 2014-01-06 | 2014-04-23 | 国家电网公司 | Power short-term load predicating method based on fast periodic component extraction |
| CN104517159A (en) * | 2014-12-18 | 2015-04-15 | 上海交通大学 | Method for predicting short-time passenger flow of bus |
| CN105321345A (en) * | 2015-09-18 | 2016-02-10 | 浙江工业大学 | Road traffic flow prediction method based on ARIMA model and kalman filtering |
Non-Patent Citations (2)
| Title |
|---|
| YUNXIANG TIAN,QUNYING LIU: "Wind speed forecasting bsed on Time series-Adaptive Kalman filtering algorithm", 《2014 IEEE FAR EAST FORUM ON NONDESTRUCTIVE EVALUATION/TESTING》 * |
| 李媛媛: "中长期负荷预测模型研究及系统实现", 《中国优秀硕士学位论文数据库》 * |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110880984A (en) * | 2019-10-18 | 2020-03-13 | 平安科技（深圳）有限公司 | Model-based flow anomaly monitoring method, device, equipment and storage medium |
| WO2021072890A1 (en) * | 2019-10-18 | 2021-04-22 | 平安科技（深圳）有限公司 | Traffic abnormality monitoring method and apparatus based on model, and device and storage medium |
| CN110851333A (en) * | 2019-11-14 | 2020-02-28 | 北京金山云网络技术有限公司 | Monitoring method and device of root partition and monitoring server |
| CN110851333B (en) * | 2019-11-14 | 2023-09-01 | 北京金山云网络技术有限公司 | Root partition monitoring method and device and monitoring server |
| CN111475772A (en) * | 2020-03-27 | 2020-07-31 | 微梦创科网络科技（中国）有限公司 | Capacity evaluation method and device |
| CN111475772B (en) * | 2020-03-27 | 2023-12-15 | 微梦创科网络科技（中国）有限公司 | Capacity assessment method and device |
| CN111723340B (en) * | 2020-06-09 | 2023-07-25 | 武汉理工大学 | Correction method for phase jump caused by polarization induced fading in optical fiber phase demodulation |
| CN111723340A (en) * | 2020-06-09 | 2020-09-29 | 武汉理工大学 | Method for correcting phase jump caused by polarization-induced fading in optical fiber phase demodulation |
| CN112019398A (en) * | 2020-07-16 | 2020-12-01 | 中国人民财产保险股份有限公司 | Network flow prediction method and device |
| CN112866129A (en) * | 2021-03-31 | 2021-05-28 | 中国工商银行股份有限公司 | Method, device and system for packing by automatically adjusting intervals according to aging marks |
| CN112866129B (en) * | 2021-03-31 | 2023-03-21 | 中国工商银行股份有限公司 | Method, device and system for packing by automatically adjusting intervals according to aging marks |
| CN113672642A (en) * | 2021-07-07 | 2021-11-19 | 德特赛维技术有限公司 | Method and system for predicting mobile network performance based on time sequence |
| WO2023165145A1 (en) * | 2022-03-02 | 2023-09-07 | 北京沃东天骏信息技术有限公司 | Time sequence traffic prediction method and apparatus, storage medium, and electronic device |
| CN114760191A (en) * | 2022-05-24 | 2022-07-15 | 咪咕文化科技有限公司 | Data service quality early warning method, system, device and readable storage medium |
| CN114760191B (en) * | 2022-05-24 | 2023-09-19 | 咪咕文化科技有限公司 | Data service quality early warning method, system, equipment and readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109995562B (en) | 2022-11-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109995562B (en) | Network traffic prediction method, device, equipment and medium | |
| US10725008B2 (en) | Automatic siting for air quality monitoring stations | |
| CN112285807B (en) | Meteorological information prediction method and device | |
| CN111222776B (en) | Satellite network coordination situation assessment method and system based on convolutional neural network | |
| CN110175690A (en) | A kind of method, apparatus, server and the storage medium of scenic spot passenger flow forecast | |
| CN114666224B (en) | Method, device, equipment and storage medium for dynamically allocating service resource capacity | |
| CN112529234A (en) | Surface water quality prediction method based on deep learning | |
| CN112787878B (en) | Network index prediction method and electronic equipment | |
| CN109981234A (en) | Self-adapting regulation method, device, equipment and the medium of dual carrier and carrier wave polymerization | |
| CN118175037A (en) | Method and device for predicting and deciding by integrating calculation network and arranging real-time analysis | |
| CN118037000A (en) | Urban resource dynamic scheduling method and system based on digital economy | |
| CN116934507B (en) | Intelligent claim settlement method and system based on big data driving | |
| Lee et al. | A machine learning approach that meets axiomatic properties in probabilistic analysis of LTE spectral efficiency | |
| Shisher et al. | On the monotonicity of information aging | |
| CN104579850A (en) | Quality of service (QoS) prediction method for Web service under mobile Internet environment | |
| CN108537322A (en) | Neural network interlayer activation value quantization method and device | |
| CN112101313A (en) | Machine room robot inspection method and system | |
| CN113095608A (en) | Index analysis method, device, equipment and storage medium | |
| CN110572356B (en) | Computing power migration method and system based on edge gateway data quality evaluation | |
| CN114970495A (en) | Name disambiguation method and device, electronic equipment and storage medium | |
| Sameh et al. | Seasonal Autoregressive Integrated Moving Average for Climate Change Time Series Forecasting | |
| Yang et al. | Historical load curve correction for short-term load forecasting | |
| Kousias et al. | Long short term memory networks for bandwidth forecasting in mobile broadband networks under mobility | |
| CN109993388B (en) | Method, device, equipment and medium for judging cause of deteriorated cell | |
| CN114818467B (en) | Accurate analysis method for satellite communication system situation |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |