CN102111284A - Method and device for predicting telecom traffic - Google Patents
Method and device for predicting telecom traffic Download PDFInfo
- Publication number
- CN102111284A CN102111284A CN200910243569XA CN200910243569A CN102111284A CN 102111284 A CN102111284 A CN 102111284A CN 200910243569X A CN200910243569X A CN 200910243569XA CN 200910243569 A CN200910243569 A CN 200910243569A CN 102111284 A CN102111284 A CN 102111284A
- Authority
- CN
- China
- Prior art keywords
- carrying capacity
- traffic carrying
- prediction
- traffic
- sample
- 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
Images
Abstract
The invention provides a method and device for predicting telecom traffic. The method comprises the following steps: determining the prediction granulation of the telecom traffic; selecting a historical sample and a prediction sample; respectively calculating the growth rates and the initial values of the telecom traffics of the historical sample and the prediction sample by using a unary linear regression model; calculating the prediction traffics of the historical sample and the prediction sample; reading the actual traffic of the historical sample; and adjusting the second prediction traffic according to the deviation between the traffic of first prediction and the actual traffic to obtain the prediction traffic adjusted by a user. The behavior mode of the user is obtained through analyzing the deviation between the linear prediction value and the actual value in the historical sample, and accordingly, the linear prediction value of future telecom traffic data is adjusted to obtain more accurate telecom network traffic, thereby providing decision support for the user to plan, organize and manage network more accurately.
Description
Technical field
The present invention relates to the telecommunication network technology field, particularly relate to a kind of telecommunication traffic Forecasting Methodology and device.
Background technology
Telecommunication traffic is the telecom information quantity that expression needs transmission, is to adjust the main foundation that all kinds of network element devices, producers and tissue are produced, for example telephone traffic, multimedia message amount, note amount etc.The characteristics of telecommunication service are that traffic carrying capacity changes greatly, and are busy not busy uneven, influenced significantly by festivals or holidays.
The telecommunication traffic prediction, be from the relevant historical record material, to obtain rule and the characteristic that the telecommunication system traffic carrying capacity changes, set up the Mathematical Modeling that to describe the telecommunication traffic variation characteristic, and then under certain required precision, utilize this mathematical model prediction telecommunication traffic of following certain specific period.The general time series analysis method that adopts of the telecommunication traffic prediction of prior art, time series analysis method is a kind of only based on the linear analysis method of telecommunication traffic data, for example, commonly used have simple linear regression analysis method, an autoregressive moving-average model (arma modeling, Auto Regression Moving Average Model) analytic approach etc., these methods all be based upon to future traffic carrying capacity change and suppose on the basis relatively slowly.But bigger for professional variable quantity, be subjected to influence festivals or holidays significant communication network simultaneously, these methods are difficult to obtain the prediction of degree of precision.
Summary of the invention
Technical problem to be solved by this invention provides a kind of telecommunication traffic Forecasting Methodology, can solve traffic carrying capacity and change greatly, is subjected to influence festivals or holidays bigger telecommunication traffic especially and predicts, thereby provide decision support for the telecommunication network management of specific period.
The present invention also provides a kind of telecommunication traffic prediction unit, to guarantee said method application in practice.
In order to address the above problem, the invention discloses a kind of telecommunication traffic Forecasting Methodology, comprising: the prediction granularity of determining telecommunication traffic; Choose the historical sample and the forecast sample of telecommunication traffic; Utilize the one-variable linear regression forecast model, calculate the first traffic carrying capacity growth rate and the first traffic carrying capacity initial value according to above-mentioned historical sample; Calculate the second traffic carrying capacity growth rate and the second traffic carrying capacity initial value according to above-mentioned forecast sample; And, calculate the first prediction traffic carrying capacity of the corresponding predicted time of historical sample institute and forecast sample corresponding predicted time second predict traffic carrying capacity; Read historical sample the actual volume of corresponding predicted time; According to the deviation of the actual volume of the corresponding predicted time of described first prediction traffic carrying capacity and the historical sample institute, the described second prediction traffic carrying capacity is adjusted, obtain after the adjustment prediction traffic carrying capacity at predicted time; Wherein, read actual volume step step any one step execution before in the end of historical sample.
Preferably, said method also comprises: generate traffic carrying capacity early warning report according to adjusted prediction traffic carrying capacity.
Preferably, said method also comprises: be presented on the computer traffic carrying capacity early warning report or printout.
Preferably, above-mentioned prediction granularity is selected to determine by the user.
Preferably, in carrying out the computer system of said method, be preset with ad hoc rules, automatically perform above-mentioned telecommunication traffic prediction flow process during condition in satisfying preset rules.
Preferably, according to described first prediction traffic carrying capacity and the historical sample the deviation of actual volume of corresponding predicted time, the described second prediction traffic carrying capacity is adjusted, adjusting the back is specially in the method for the prediction traffic carrying capacity of predicted time: the actual volume and first of the corresponding predicted time of historical sample institute is predicted the poor of traffic carrying capacity, multiply by the ratio of the second traffic carrying capacity growth rate and the first traffic carrying capacity growth rate, add the second prediction traffic carrying capacity, obtain adjusted prediction traffic carrying capacity.
Preferably, above-mentioned one-variable linear regression forecast model is specially:
The traffic growth rate
The business initial value
Prediction traffic carrying capacity Y
t=b+kt
Wherein, n represents sample size, y
iThe traffic carrying capacity of representing i sample, t
iThe time series of representing i sample, k are represented the traffic carrying capacity growth rate, and b represents traffic carrying capacity initial value, Y
tPrediction traffic carrying capacity when the express time ordinal number is t; The described first traffic carrying capacity growth rate, the second traffic carrying capacity growth rate, the first traffic carrying capacity initial value, the second traffic carrying capacity initial value, the first prediction traffic carrying capacity and the second prediction traffic carrying capacity can adopt above-mentioned formula to calculate respectively and obtain.
Preferably, above-mentioned communication network is a mobile communications network, and network element device comprises mobile switching centre, Mobile Switching Center Server, base station manager, radio network controller, sms center or MMS center;
Preferably, above-mentioned telecommunication service is voice service, short message service, MMS or Internet service; Above-mentioned telecommunication traffic is telephone traffic, note quantity, multimedia message quantity, internet data flow or the Internet line duration.
According to another preferred embodiment of the invention, a kind of telecommunication traffic prediction unit is also disclosed, comprise: prediction granularity determining unit, sample are chosen unit, traffic carrying capacity predicting unit, traffic carrying capacity adjustment unit, and wherein: prediction granularity determining unit is used for determining the prediction granularity of telecommunication service; Sample is chosen the unit and is used for telecommunication traffic prediction granularity according to the output of prediction granularity determining unit, chooses the historical sample and the forecast sample of telecommunication traffic, and, obtain historical sample the actual volume of corresponding predicted time; The traffic carrying capacity predicting unit is used for according to the one-variable linear regression forecast model, based on historical sample and the forecast sample of originally choosing unit output, calculate the first traffic carrying capacity growth rate and the first traffic carrying capacity initial value of historical sample respectively, and the second traffic carrying capacity growth rate of forecast sample and the second traffic carrying capacity initial value, and, calculate the first prediction traffic carrying capacity of the corresponding predicted time of historical sample institute and forecast sample corresponding predicted time second predict traffic carrying capacity; The output result that the traffic carrying capacity adjustment unit is used for choosing according to sample unit and traffic carrying capacity predicting unit adjusts the prediction traffic carrying capacity.
Preferably, said apparatus also comprises traffic carrying capacity early warning report generation unit, and the prediction traffic carrying capacity at predicted time after the adjustment that is used for generating according to the traffic carrying capacity adjustment unit generates traffic carrying capacity early warning report.
Preferably, said apparatus also comprises the output unit that predicts the outcome, and is used for showing or printout adjusting the prediction traffic carrying capacity or the traffic carrying capacity early warning report of back at predicted time.
Preferably, said apparatus also comprises prediction execution unit, this prediction execution unit is preset with ad hoc rules, in satisfying ad hoc rules during preset condition, call prediction granularity determining unit automatically, sample is chosen unit, traffic carrying capacity predicting unit, traffic carrying capacity adjustment unit, traffic carrying capacity early warning report generation unit and the output unit that predicts the outcome and carried out the telecommunication traffic forecasting process.
Preferably, the concrete grammar of above-mentioned traffic carrying capacity adjustment unit adjustment prediction traffic carrying capacity is: adjust the ratio that difference that the back equals the actual volume of historical sample and the first prediction traffic carrying capacity in the prediction traffic carrying capacity of predicted time multiply by the second traffic carrying capacity growth rate and the first traffic carrying capacity growth rate, add the second prediction traffic carrying capacity that abovementioned steps obtains.
Compared with prior art, the present invention has the following advantages:
By the linear predictor of analysis of history sample traffic carrying capacity and the deviation of actual value, obtain the subscribers to telecommunication network behavior pattern, and in view of the above the linear predictor of following telecommunication traffic data is adjusted, obtain telecommunication network service amount more accurately, for the user carries out the network planning more accurately and organization and administration provide decision support.
In addition, the prediction scheme that the present invention proposes is simple and easy to usefulness, is specially adapted to the telecommunication network service amount prediction of very big festivals or holidays of traffic carrying capacity amplitude of variation, and precision of prediction is more accurate.
Description of drawings
Fig. 1 is an embodiment flow chart of telecommunication traffic Forecasting Methodology of the present invention;
Fig. 2 is an example structure block diagram of telecommunication traffic prediction unit of the present invention.
Embodiment
For above-mentioned purpose of the present invention, feature and advantage can be become apparent more, the present invention is further detailed explanation below in conjunction with the drawings and specific embodiments.
One of core idea of the present invention is, according to the linear predictor of historical traffic data and the deviation of actual value, the linear predictor of the traffic data that predict adjusted, and the accurate service amount predicts the outcome to obtain more.
With reference to Fig. 1, show the flow process of an embodiment of telecommunication traffic Forecasting Methodology of the present invention, specifically may further comprise the steps:
Step 101: the prediction granularity of determining telecommunication traffic;
Different prediction granularities can satisfy the user traffic carrying capacity on certain class network element device in different spaces and the different time is predicted.This prediction granularity is made of time interval, geographical space and three key elements of network element device type, according to determined prediction granularity, can predict the telecommunication traffic on certain class network element device of being administered in the following a certain sometime communication network zone.Can determine according to user's selection, also can in system, preset some rules and select.Described communication network can be fixed-line telephone network, data communication network, also can be mobile radio communication.Related telecommunication service can be voice service, short message service, MMS or Internet service etc.
With the mobile radio communication is example, the geographical space that the user selects can be the whole nation, certain province or somewhere are as the spatial granularity of prediction, the time interval of selecting can be one day, several days, waited time granularity in one month as prediction, take place professional based on network element device can select (the MSC of mobile switching centre, MobileSwitch Center), Mobile Switching Center Server (MSS, MSC Server), base station manager (the BSC of second generation network, Base Station Controller), radio network controller (the RNC of third generation network, Radio Network Controller), sms center (SMSC, Short Messaging Service Center) or MMS center (MMSC, Multimedia Messaging Service Center) etc.Certainly, can directly will determine that good prediction granularity is stored in the system, system carries out follow-up operation according to the prediction granularity of acquiescence.
The prediction granularity can define voluntarily according to user's needs, and general time granularity is no more than one month.
Step 102: historical sample and the forecast sample of choosing described telecommunication traffic;
According to the prediction granularity that step 101 is determined, the traffic carrying capacity of choosing a period of time before predicted time (as prediction day) from the communication network database is as forecast sample; The traffic carrying capacity of choosing a period of time before corresponding predicted time in cycle reference time is as historical sample.
A time cycle before the above-mentioned place time cycle predicted time of index futures week reference time.If with 1 year as a time cycle, with one day time granularity as prediction, also can select the first two years as reference time as the reference time the previous year of just selecting predicted day place time.As, when the user wish to predict A in 2009 economize " five. one " during voice service amount situation during International Labour Day, can be with the traffic data in 1 day~April 23 April in 2009 as forecast sample, with the traffic data in 1 day~April 23 April in 2008 as historical sample.Need to prove that cycle reference time is near more from predicted day, its precision that predicts the outcome is high more.
Step 103: utilize the one-variable linear regression forecast model, calculate the traffic carrying capacity growth rate and the traffic carrying capacity initial value of historical sample and forecast sample respectively;
Adopt one-variable linear regression forecast model Y
t=b+kt calculates, and wherein b is constant (a traffic carrying capacity initial value), and k is regression parameter (a traffic carrying capacity growth rate), and t represents the time series in the monobasic linear session sequence, for example: 0,1,2,3,4... ...-3 ,-2 ,-1,0,1,2,3..., Y
tTraffic data when the express time ordinal number is t;
Wherein, the computing formula of one-variable linear regression forecast model constant b (business initial value) and regression parameter k (traffic growth rate) is as follows:
In the above-mentioned formula, n represents sample size, y
iThe traffic carrying capacity of representing i sample, t
iThe time series of representing i sample, k are represented the traffic carrying capacity growth rate, and b represents traffic carrying capacity initial value, Y
tPrediction traffic carrying capacity when the express time ordinal number is t.
Corresponding to historical sample, traffic carrying capacity growth rate and traffic carrying capacity initial value are used k respectively
1And b
1Expression, corresponding to forecast sample, traffic carrying capacity growth rate and traffic carrying capacity initial value are used k respectively
2And b
2Expression.
Step 104: calculate historical sample in the traffic carrying capacity of predicted time and forecast sample traffic carrying capacity at predicted time;
The traffic carrying capacity growth rate k of the historical sample that obtains according to step 103
1With traffic carrying capacity initial value b
1, utilize one-variable linear regression anticipation function Y
T1=b
1+ k
1Traffic carrying capacity predicted value when t calculates the time series try to achieve historical sample and is t is Y
T1
The traffic carrying capacity growth rate k of the forecast sample that obtains according to step 103
2With traffic carrying capacity initial value b
2, utilize one-variable linear regression anticipation function Y
T2=b
2+ k
2Traffic carrying capacity predicted value when t calculates the time series try to achieve forecast sample and is t is Y
T2
Step 105: obtain historical sample the actual volume Y of corresponding predicted time
T1';
As, when the user wish to predict A in 2009 economize " five. one " during voice service amount situation during International Labour Day, can promptly obtain actual volume with historical sample at the actual traffic data in 30 days~May 3 April in 2008 comparison other as the reference time with reference to corresponding predicted time of time.
Step 106:, adjust the traffic carrying capacity predicted value of forecast sample at predicted time according to the prediction traffic carrying capacity of historical sample and the deviation of actual volume;
Adjusted predicted time traffic carrying capacity predicted value Y
T2' equal the actual volume Y of the corresponding predicted time of historical sample
T1' corresponding predicted time traffic carrying capacity predicted value Y with historical sample
T1Poor, multiply by forecast sample traffic growth rate k
2With historical sample traffic growth rate k
1Ratio, add the corresponding predicted time traffic carrying capacity predicted value Y of forecast sample
T2Computing formula is as follows:
Y
t2′=Y
t2+(Y
t1′-Y
t1)×k
2/k
1
In another preferred embodiment of the inventive method embodiment, can also comprise: above-mentioned predicting the outcome shown, or printout the decision-making foundation of providing convenience for the user.In addition, can also predict the outcome and early warning baseline generation traffic carrying capacity early warning report according to above-mentioned, described early warning baseline is to be provided with according to the situation of change of traffic carrying capacity, telephone traffic early warning baseline during red-letter day will be higher than telephone traffic early warning baseline at ordinary times, can set when being higher than telephone traffic early warning baseline when predicting the outcome and will report to the police, showing has the unexpected variation of traffic carrying capacity to take place.Like this, just for the telecommunication network service management, particularly huge especially operation management festivals or holidays of traffic growth amount provides good technical support.
In order to reduce operation personnel's workload, reduce the accident that human factor causes, some ad hoc ruless can also be set in system, in satisfying ad hoc rules, during preset condition, automatically perform the traffic carrying capacity forecasting process.For example, carry out the traffic carrying capacity prediction important the last week festivals or holidays automatically.
In the forecasting process of said method embodiment, the linear predictor by analyzing the traffic carrying capacity of predicted time in cycle reference time and the deviation of actual value, can obtain the user behavior pattern of predicted time, and in view of the above the linear predictor of predicted time traffic data is adjusted, obtain traffic carrying capacity of predicted time more accurately, for communication network planning and management provide decision-making foundation more accurately.In addition, this method embodiment is specially adapted to the telecommunication network service amount prediction of very big festivals or holidays of traffic carrying capacity amplitude of variation.
Method example one: traffic forecast
Below, with A in 2009 economize " five. one " voice service during International Labour Day is example, specifies telecommunication traffic Forecasting Methodology of the present invention and effect thereof.Specifically comprise the steps:
Step 1: choose A and economize during April 30 to May 3 in 2009, the telephone traffic that produce in mobile switching centre every day is as the prediction granularity.
Step 2: choose on April 23rd, 1 day 1 April in 2009 A and economize the telephone traffic that produces in mobile switching centre every day as forecast sample, as shown in table 1:
Date | Telephone traffic | Date | Telephone traffic | Date | Telephone traffic |
April 1 | 465496.6495 | April 9 | 467208.2000 | April 17 | 486043.2632 |
April 2 | 462500.8755 | April 10 | 473943.7504 | April 18 | 427970.0125 |
April 3 | 471012.9628 | April 11 | 427970.0125 | April 19 | 413297.4748 |
April 4 | 402057.8877 | April 12 | 413297.4748 | April 20 | 467536.9374 |
April 5 | 385927.7499 | April 13 | 460312.0751 | April 21 | 456247.0752 |
April 6 | 399570.1747 | April 14 | 463109.8213 | April 22 | 467407.2002 |
April 7 | 467355.2371 | April 15 | 466978.1499 | April 23 | 433946.9542 |
April 8 | 465943.2166 | April 16 | 453871.0002 |
Table 1,1 day~April 23 April in 2009, A economized the telephone traffic (unit: Ireland) of every day
Obtain on April 23rd, 1 day 1 April in 2008 A and economize the traffic data that produces in mobile switching centre every day as historical sample, as shown in table 2.
Date | Telephone traffic | Date | Telephone traffic | Date | Telephone traffic |
April 1 | 452675.2452 | April 9 | 430203.5111 | April 17 | 460921.5798 |
April 2 | 450358.5456 | April 10 | 451109.8811 | April 18 | 451093.7980 |
April 3 | 457055.9756 | April 11 | 449540.9667 | April 19 | 406664.9031 |
April 4 | 397783.8744 | April 12 | 415569.6622 | April 20 | 398874.4792 |
April 5 | 369161.5567 | April 13 | 402873.3033 | April 21 | 445134.8655 |
April 6 | 379882.7867 | April 14 | 445406.0622 | April 22 | 441288.4802 |
April 7 | 440147.9667 | April 15 | 436469.4611 | April 23 | 440408.8611 |
April 8 | 427023.0967 | April 16 | 438826.2600 |
Table 2,1 day~April 23 April in 2008, A economized the telephone traffic (unit: Ireland) of every day
Step 3: calculate the traffic carrying capacity growth rate and the traffic carrying capacity initial value of historical sample, and forecast sample traffic carrying capacity growth rate and traffic carrying capacity initial value.
For convenience of calculation, getting the seasonal effect in time series interlude usually is initial point, gets April 12 here respectively for initial point, i.e. t during April 12
12=0; T during April 11
11=-1; T during April 13
13=1 or the like.Have this moment:
The computing formula of k and b can be reduced to:
According to above-mentioned formula, with 23, every day of 1 day~2008 on April April in 2008 telephone traffic as historical sample, calculate traffic carrying capacity growth rate k
1=471.7358991, traffic carrying capacity initial value b
1=429933.701;
With 23, every day of 1 day~2009 on April April in 2009 telephone traffic as the traffic carrying capacity growth rate k of forecast sample
2=504.0300679, traffic carrying capacity initial value b
2=447782.7894.
Step 4: according to one-variable linear regression anticipation function Y
T1=b
1+ k
1T calculates telephone traffic 3, every day of 30 days~2008 on May April in 2008, as the prediction traffic carrying capacity Y of historical sample
T1As shown in table 3:
Date | The prediction traffic carrying capacity Y of historical sample t1(unit: Ireland) |
On April 30th, 2008 | 438424.9472 |
On May 1st, 2008 | 438896.6831 |
On May 2nd, 2008 | 439368.419 |
On May 3rd, 2008 | 439840.1549 |
Table 3, the prediction telephone traffic that on May 3rd, 30 days 1 April in 2008, A economized
In like manner, according to one-variable linear regression anticipation function Y
T2=b
2+ k
2T calculates telephone traffic 3, every day of 30 days~2009 on May April in 2009, as the prediction traffic carrying capacity Y of forecast sample
T2, as shown in table 4:
Date | The prediction traffic carrying capacity Y of forecast sample t2(unit: Ireland) |
On April 30th, 2009 | 456855.3306 |
On May 1st, 2009 | 457359.3607 |
On May 2nd, 2009 | 457863.3908 |
On May 3rd, 2009 | 458367.4208 |
Table 4, the prediction telephone traffic that 30 days~May 3 April in 2009, A economized
Step 5: obtain telephone traffic Y in 30 days~2008 April in 2008 actual every day on May 3, according to historical sample
T1', as shown in table 5.
Date | The actual telephone traffic Y of historical sample t1' (unit: Ireland) |
On April 30th, 2008 | 492298.9383 |
On May 1st, 2008 | 435705.0411 |
On May 2nd, 2008 | 384150.1022 |
On May 3rd, 2008 | 378436.0682 |
Table 5, the actual telephone traffic that 30 days~May 3 April in 2008, A economized
Step 6: historical sample actual telephone traffic Y on May 3,30 days~2008 April in 2008
T1' with on May 3,30 days~2008 April in 2008 prediction telephone traffic Y
T1Poor, multiply by forecast sample traffic carrying capacity growth rate k
2With historical sample traffic carrying capacity growth rate k
1Ratio, add on May 3,30 days~2009 April in 2009 prediction telephone traffic Y
T2, the final adjustment back prediction telephone traffic Y that obtains
T2'.Computing formula is as follows:
Y
t2′=Y
t2+(Y
t1′-Y
t1)×k
2/k
1
Result of calculation is as shown in table 6:
Date | Adjusted prediction telephone traffic Y t2' (unit: Ireland) |
On April 30th, 2009 | 514417.4357 |
On May 1st, 2009 | 453949.2248 |
On May 2nd, 2009 | 398864.9299 |
On May 3rd, 2009 | 392759.7238 |
It is as shown in table 7 that table 6,30 days~May 3 April in 2009, A economized adjusted prediction traffic forecast resultant error analysis:
Date | Actual telephone traffic in 2009 | One-variable linear regression prediction telephone traffic Y t2 | The one-variable linear regression predicated error | The present invention predicts telephone traffic Y 2 | Predicated error % of the present invention |
April 30 | 503887.9877 | 456855.3306 | 9.33% | 514417.4357 | 2.09% |
May 1 | 433409.8879 | 457359.3607 | 5..53% | 453949.2248 | 4.74% |
May 2 | 387213.7627 | 457863.3908 | 18.25% | 398864.9299 | 3.01% |
May 3 | 383643.538 | 458367.4208 | 19.48% | 392759.7238 | 2.38% |
The present invention that table 7,30 days~May 3 April in 2009, A economized predict telephone traffic (unit: Ireland),
One-variable linear regression prediction telephone traffic and actual telephone traffic comparative result
As can be seen from Table 7, adopt forecast method error rate of the present invention to be controlled in 5%, minimal error 2.09%, worst error 4.74%; And adopt the direct predicated error change of one-variable linear regression Forecasting Methodology bigger, minimum 5.52%, maximum 19.48%; The Forecasting Methodology that the present invention's employing is described is more accurate, more effective.
Method example two: note amount prediction
For further specifying the effect of the inventive method embodiment, economizing the note quantitative forecast in 30 days~May 3 April in 2009 with A below is that example describes.Calculation procedure is identical with previous examples, lists related data at this and describes.
One, about historical sample and forecast sample
On April 23,1 day~2008 April in 2008 and note incremental data tabulation on April 23rd, 1 day 1 April in 2009 are as shown in table 8, unit: 1,000,000:
Date | 2008 | 2009 | Date | 2008 | 2009 |
April 1 | 204.43 | 239.86 | April 13 | 166.94 | 231.21 |
April 2 | 174.54 | 228.72 | April 14 | 174.65 | 230.88 |
April 3 | 186.04 | 232.84 | April 15 | 177.21 | 225.07 |
April 4 | 181.18 | 228.32 | April 16 | 173.12 | 224.66 |
April 5 | 168.99 | 220.19 | April 17 | 173.75 | 226.57 |
April 6 | 168.03 | 218.75 | April 18 | 172.69 | 222.83 |
April 7 | 173.01 | 220.25 | April 19 | 171.43 | 219.79 |
April 8 | 180.4 | 238.49 | April 20 | 168.23 | 223.9 |
April 9 | 173.69 | 229.91 | April 21 | 176.42 | 222.2 |
April 10 | 172.42 | 232.6 | April 22 | 171.48 | 220.83 |
April 11 | 179.05 | 232.19 | April 23 | 173.18 | 225.13 |
April 12 | 168.76 | 225.28 |
Table 8, the note quantity that A economized during 1~April 23 April in 2008 and 2009
Two, the business information that takes place about reality
On May 3,30 days~2008 April in 2008 and on May 3,30 days~2009 April in 2009, the actual note quantity tabulation that takes place was as shown in table 9, unit: 1,000,000:
Date | 2008 | 2009 |
April 30 | 195.57 | 249.66 |
May 1 | 201.98 | 239.97 |
May 2 | 178.5 | 224.27 |
May 3 | 170.85 | 215.64 |
Table 9, the actual note quantity that A economized during 30~May 3 April in 2008 and 2009
Three, about traffic carrying capacity growth rate k and traffic carrying capacity initial value b
As shown in table 10 according to k and b that table 8 historical sample and forecast sample calculate:
Parameter | 2008 | 2009 |
k | -0.535573123 | -0.388369565 |
b | 175.2017391 | 226.9769565 |
The traffic growth rate of table 10, historical sample and forecast sample and traffic carrying capacity initial value parameter
Four, the prediction traffic carrying capacity that calculates according to historical sample and forecast sample
According to the one-variable linear regression forecast model, the note incremental data tabulation on 30~May 3 April that prediction obtains is as shown in table 11, unit: 1,000,000:
Date | 2008 | 2009 |
April 30 | 173.52 | 219.46 |
May 1 | 172.98 | 219.08 |
May 2 | 172.45 | 218.69 |
May 3 | 171.92 | 218.30 |
Table 11, the prediction note quantity that A economized during 30~May 3 April in 2008 and 2009
Five, adjusted predicting the outcome
The Forecasting Methodology according to the present invention, predicting the outcome of drawing at last is as shown in table 12, unit: 1,000,000:
Date | Adjusted prediction traffic carrying capacity in 2009 |
April 30 | 243.02 |
May 1 | 250.05 |
May 2 | 225.15 |
May 3 | 217.16 |
Prediction note quantity after table 12, A economizes during 30 days~May 3 April in 2009 the adjustment
Six, the error analysis that predicts the outcome is as shown in table 13:
Date | Actual note quantity in 2009 | One-variable linear regression prediction note quantity | Monobasic is linear to return the present invention and predicts that the present invention predicts and return predicated error note quantitative error |
April 30 | 249.66 | 219.46 | 12.09% 243.02 2.66% |
May 1 | 239.97 | 219.08 | 8.71% 250.05 4.20% |
May 2 | 224.27 | 218.69 | 2.49% 225.15 0.39% |
May 3 | 215.64 | 218.30 | 1.23% 217.16 0.70% |
The present invention that table 13,30 days~May 3 April in 2009, A economized predict the note number (unit: the bar number),
One-variable linear regression prediction note number and actual note are counted comparative result
As can be seen from Table 13, it is more more accurate than adopting one-variable linear regression method prediction result to adopt Forecasting Methodology prediction result of the present invention.
For aforesaid each method embodiment, simple in order to describe, so it all is expressed as a series of combination of actions, but those skilled in the art should know, the present invention is not subjected to the restriction of described sequence of movement, because according to the present invention, some step can adopt other orders or carry out simultaneously.Secondly, those skilled in the art should know that also said method embodiment all belongs to preferred embodiment, and related action and module might not be that the present invention is necessary.
With reference to Fig. 2, show the structured flowchart of telecommunication traffic prediction unit one embodiment of the present invention, specifically comprise with lower unit:
Prediction granularity determining unit 21: the prediction granularity that is used for determining telecommunication service.
Apparatus of the present invention embodiment communication network applicatory can be a fixed-line phone network, also can be mobile radio communication.When determining the prediction granularity, according to the business field (as voice service, short message service, MMS or Internet service etc.) that the user paid close attention to, determine prediction granularity based on time interval, geographical space and three key elements formations of network element device type, the time granularity of prediction can be one day, several days, one month etc., the spatial granularity of prediction can be the communication network zone, as Beijing area, Liaoning Province, Daliang City etc.; Take place professional based on network element device can be MSC, MSS, BSC, RNC, SMSC or MMSC or the like.About the confirmation method of prediction granularity, can also can in prediction granularity determining unit 21, preset the prediction granularity by predicting that granularity determining unit 21 prompting users select or input.
Sample is chosen unit 22: be used for the result according to 21 outputs of prediction granularity determining unit, choose the historical sample and the forecast sample of telecommunication service; And, obtain the actual volume of the pairing predicted time of historical sample.
According to the prediction granularity that prediction granularity determining unit 21 is determined, the traffic carrying capacity of choosing a period of time before predicted time (as prediction day) from the communication network database is as forecast sample; The traffic carrying capacity of choosing a period of time before corresponding predicted time in cycle reference time is as historical sample.
As, when the user wish to predict A in 2009 economize " five. one " during voice service amount situation during International Labour Day, can be with the traffic data in 1 day~April 23 April in 2008 as historical sample, with the traffic data in 30 days~May 3 April in 2008 comparison other as the reference time, with the traffic data in 1 day~April 23 April in 2009 as forecast sample.
Traffic carrying capacity predicting unit 23: be used for according to one-variable linear regression forecast model Y
t=b+kt chooses historical sample and the forecast sample that unit 22 is exported based on sample, calculates the traffic carrying capacity growth rate and the traffic carrying capacity initial value of historical sample respectively, and the traffic carrying capacity growth rate of forecast sample and traffic carrying capacity initial value; And, be used for further calculating historical sample in the prediction traffic carrying capacity of corresponding predicted time and forecast sample prediction traffic carrying capacity at corresponding predicted time;
Wherein b is constant (a traffic carrying capacity initial value), and k is regression parameter (a traffic carrying capacity growth rate), and t represents the time series in the monobasic linear session sequence, for example: 0,1,2,3,4... ...-3 ,-2 ,-1,0,1,2,3..., Y
tTraffic data when the express time ordinal number is t.
Wherein, the computing formula of one-variable linear regression forecast model constant b (business initial value) and regression parameter k (traffic growth rate) is as follows:
In the above-mentioned formula, n represents sample size, y
iThe traffic carrying capacity of representing i sample, t
iThe time series of representing i sample, k are represented the traffic carrying capacity growth rate, and b represents traffic carrying capacity initial value, Y
tPrediction traffic carrying capacity when the express time ordinal number is t.
Traffic carrying capacity adjustment unit 24: the output result who is used for choosing according to sample unit 22 and traffic carrying capacity predicting unit 23 adjusts the prediction traffic carrying capacity, obtains the traffic carrying capacity at predicted time.
Wherein, adjust the prediction traffic carrying capacity Y of back at predicted time
T2' equal the actual volume Y of the corresponding predicted time of historical sample
T1' with the prediction traffic carrying capacity Y of the corresponding predicted time of historical sample
T1Poor, multiply by the traffic carrying capacity growth rate k of forecast sample
2Traffic carrying capacity growth rate k with historical sample
1Ratio, add the prediction traffic carrying capacity of the forecast sample of traffic carrying capacity predicting unit 23 output at predicted time.
Computing formula is as follows:
Y
t2′=Y
t2+(Y
t1′-Y
t1)×k
2/k
1
In another preferred embodiment of apparatus of the present invention embodiment, also comprise the early warning report generation unit 25 and the output unit 26 that predicts the outcome, wherein, early warning report generation unit 25 is used for the adjusted prediction traffic carrying capacity according to 24 generations of traffic carrying capacity adjustment unit, generates traffic carrying capacity early warning report; The output unit 26 that predicts the outcome is used for the described back of adjusting is shown or printout in the prediction traffic carrying capacity or the traffic carrying capacity early warning report of predicted time.
In addition, apparatus of the present invention embodiment can also comprise prediction execution unit 27, be preset with ad hoc rules in this prediction execution unit 27, during preset condition, prediction execution unit 27 automatic sequence calls prediction granularity determining units 21, sample are chosen parts such as unit 22, traffic carrying capacity predicting unit 23, traffic carrying capacity adjustment unit 24, early warning report generation unit 25 and the output unit 26 that predicts the outcome and are carried out the telecommunication traffic forecasting process in satisfying ad hoc rules.
Need to prove that said apparatus embodiment belongs to preferred embodiment, related unit might not be that the present invention is necessary.
Each embodiment in this specification all adopts the mode of going forward one by one to describe, and what each embodiment stressed all is and the difference of other embodiment that identical similar part is mutually referring to getting final product between each embodiment.For device embodiment of the present invention, because it is similar substantially to method embodiment, so description is fairly simple, relevant part gets final product referring to the part explanation of method embodiment.
More than a kind of telecommunication traffic Forecasting Methodology provided by the present invention and device are described in detail, used specific case herein principle of the present invention and execution mode are set forth, the explanation of above embodiment just is used for helping to understand method of the present invention and core concept thereof; Simultaneously, for one of ordinary skill in the art, according to thought of the present invention, the part that all can change in specific embodiments and applications, in sum, this description should not be construed as limitation of the present invention.
Claims (12)
1. a telecommunication traffic Forecasting Methodology is characterized in that, described method is carried out by computer, and described method comprises:
Determine the prediction granularity of described telecommunication traffic;
Choose the historical sample and the forecast sample of described telecommunication traffic;
Utilize the one-variable linear regression forecast model, calculate the first traffic carrying capacity growth rate and the first traffic carrying capacity initial value according to described historical sample; Calculate the second traffic carrying capacity growth rate and the second traffic carrying capacity initial value according to described forecast sample; And, calculate described historical sample corresponding predicted time the corresponding predicted time of first prediction traffic carrying capacity and the forecast sample institute second predict traffic carrying capacity;
Read historical sample the actual volume of corresponding predicted time;
According to the deviation of the actual volume of the corresponding predicted time of described first prediction traffic carrying capacity and the historical sample institute, the described second prediction traffic carrying capacity is adjusted, obtain after the adjustment prediction traffic carrying capacity at predicted time;
Wherein, described read historical sample the actual volume step of corresponding predicted time can be in the end any one step before step carry out.
2. the method for claim 1 is characterized in that, described method also comprises: generate traffic carrying capacity early warning report according to described adjusted prediction traffic carrying capacity.
3. method as claimed in claim 2 is characterized in that, described method also comprises: described traffic carrying capacity early warning report is presented on the described computer, or printout.
4. as the described method of one of claim 1~3, it is characterized in that:
Described prediction granularity is selected to determine by the user; Or
In described computer system, be preset with ad hoc rules, automatically perform above-mentioned telecommunication traffic prediction flow process during condition in satisfying preset rules.
5. the method for claim 1, it is characterized in that, the deviation of described actual volume according to the corresponding predicted time of described first prediction traffic carrying capacity and the historical sample institute is adjusted the described second prediction traffic carrying capacity, obtains after the adjustment to be specially in the method for the prediction traffic carrying capacity of predicted time:
Historical sample the actual volume of corresponding predicted time and the first prediction traffic carrying capacity poor, multiply by the ratio of the second traffic carrying capacity growth rate and the first traffic carrying capacity growth rate, add the second prediction traffic carrying capacity, obtain adjusted prediction traffic carrying capacity at predicted time.
6. the method for claim 1 is characterized in that, described one-variable linear regression forecast model is specially:
The traffic growth rate
The business initial value
Prediction traffic carrying capacity Y
t=b+kt
Wherein, n represents sample size, y
iThe traffic carrying capacity of representing i sample, t
iThe time series of representing i sample, k are represented the traffic carrying capacity growth rate, and b represents traffic carrying capacity initial value, Y
tPrediction traffic carrying capacity when the express time ordinal number is t;
The described first traffic carrying capacity growth rate, the second traffic carrying capacity growth rate, the first traffic carrying capacity initial value, the second traffic carrying capacity initial value, the first prediction traffic carrying capacity and the second prediction traffic carrying capacity can adopt above-mentioned formula to calculate respectively and obtain.
7. the method for claim 1 is characterized in that:
Described communication network is a mobile communications network, and described network element device comprises mobile switching centre, Mobile Switching Center Server, base station manager, radio network controller, sms center or MMS center;
And/or
Described telecommunication service is voice service, short message service, MMS or Internet service; Described telecommunication traffic is telephone traffic, note quantity, multimedia message quantity, internet data flow or the Internet line duration.
8. a telecommunication traffic prediction unit is characterized in that, described device comprises that prediction granularity determining unit, sample choose unit, traffic carrying capacity predicting unit, traffic carrying capacity adjustment unit, wherein:
Described prediction granularity determining unit is used for determining the prediction granularity of described telecommunication traffic;
Described sample is chosen the unit and is used for predicting granularity according to the telecommunication traffic of described prediction granularity determining unit output, chooses the historical sample and the forecast sample of described telecommunication traffic; And, obtain historical sample the actual volume of corresponding predicted time;
Described traffic carrying capacity predicting unit is used for according to the one-variable linear regression forecast model, choose the historical sample and the forecast sample of unit output based on described sample, calculate the first traffic carrying capacity growth rate and the first traffic carrying capacity initial value of described historical sample respectively, and the second traffic carrying capacity growth rate of described forecast sample and the second traffic carrying capacity initial value; And, calculate described historical sample corresponding predicted time the corresponding predicted time of first prediction traffic carrying capacity and the forecast sample institute second predict traffic carrying capacity;
Described traffic carrying capacity adjustment unit is used for adjusting described prediction traffic carrying capacity according to the result of calculation that described sample is chosen unit and traffic carrying capacity predicting unit, obtains the prediction traffic carrying capacity at predicted time.
9. device as claimed in claim 8 is characterized in that, described device also comprises early warning report generation unit, and the prediction traffic carrying capacity at predicted time after the adjustment that is used for generating according to described traffic carrying capacity adjustment unit generates traffic carrying capacity early warning report.
10. device as claimed in claim 8 is characterized in that described device also comprises the output unit that predicts the outcome, and is used for described adjusted prediction traffic carrying capacity or traffic carrying capacity early warning report are shown or printout.
11. device as claimed in claim 10, it is characterized in that, described device also comprises prediction execution unit, described prediction execution unit is preset with ad hoc rules, calls described prediction granularity determining unit during preset condition automatically in satisfying described ad hoc rules, sample is chosen unit, traffic carrying capacity predicting unit, traffic carrying capacity adjustment unit, early warning report generation unit and the output unit that predicts the outcome and carried out the telecommunication traffic forecasting process.
12. device as claimed in claim 8, it is characterized in that, the method of described traffic carrying capacity adjustment unit adjustment prediction traffic carrying capacity is specially: describedly adjust the ratio that difference that the back equals the actual volume of the corresponding predicted time of historical sample institute and the first prediction traffic carrying capacity in the prediction traffic carrying capacity of predicted time multiply by the second traffic carrying capacity growth rate and the first traffic carrying capacity growth rate, add that second predicts traffic carrying capacity.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN 200910243569 CN102111284B (en) | 2009-12-28 | 2009-12-28 | Method and device for predicting telecom traffic |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN 200910243569 CN102111284B (en) | 2009-12-28 | 2009-12-28 | Method and device for predicting telecom traffic |
Publications (2)
Publication Number | Publication Date |
---|---|
CN102111284A true CN102111284A (en) | 2011-06-29 |
CN102111284B CN102111284B (en) | 2013-09-04 |
Family
ID=44175318
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN 200910243569 Active CN102111284B (en) | 2009-12-28 | 2009-12-28 | Method and device for predicting telecom traffic |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102111284B (en) |
Cited By (25)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102711177A (en) * | 2012-04-26 | 2012-10-03 | 北京邮电大学 | Service prediction based load balancing method |
CN103024762A (en) * | 2012-12-26 | 2013-04-03 | 北京邮电大学 | Service feature based communication service forecasting method |
CN103178990A (en) * | 2011-12-20 | 2013-06-26 | 中国移动通信集团青海有限公司 | Network device performance monitoring method and network management system |
CN103338461A (en) * | 2013-06-18 | 2013-10-02 | 中国联合网络通信集团有限公司 | Method and device for network planning based on prediction of volume of business |
CN103514506A (en) * | 2012-06-29 | 2014-01-15 | 国际商业机器公司 | Automatic event analysis method and system |
CN103856344A (en) * | 2012-12-05 | 2014-06-11 | 中国移动通信集团北京有限公司 | Alarm event information processing method and device |
CN104125584A (en) * | 2013-04-27 | 2014-10-29 | 中国移动通信集团福建有限公司 | Service index realization prediction method aiming at network service and apparatus thereof |
WO2014190923A1 (en) * | 2013-05-29 | 2014-12-04 | 华为技术有限公司 | Method and system for predicting traffic of cellular network |
CN104766144A (en) * | 2015-04-22 | 2015-07-08 | 携程计算机技术(上海)有限公司 | Order forecasting method and system |
CN104978604A (en) * | 2014-04-04 | 2015-10-14 | 中国移动通信集团福建有限公司 | Simulation method based on service capability model and simulation device based on service capability model |
CN105138691A (en) * | 2015-09-18 | 2015-12-09 | 北京百度网讯科技有限公司 | Method and system for analyzing user traffic |
CN105468774A (en) * | 2015-07-07 | 2016-04-06 | 中华电信股份有限公司 | Data usage prediction apparatus, method and computer program product thereof |
CN105634787A (en) * | 2014-11-26 | 2016-06-01 | 华为技术有限公司 | Evaluation method, prediction method and device and system for network key indicator |
CN105792231A (en) * | 2014-12-23 | 2016-07-20 | 中国移动通信集团上海有限公司 | Telephone traffic forecasting method and device |
CN105979538A (en) * | 2016-07-19 | 2016-09-28 | 浪潮软件集团有限公司 | Method for analyzing lost telephone traffic and flow during cell interruption |
CN106164864A (en) * | 2014-03-31 | 2016-11-23 | 微软技术许可有限责任公司 | The prediction load of service is flexible |
WO2017095344A1 (en) | 2015-12-04 | 2017-06-08 | Turkcell Teknoloji Arastirma Ve Gelistirme Anonim Sirketi | A network traffic estimation system |
CN107092973A (en) * | 2016-11-25 | 2017-08-25 | 口碑控股有限公司 | The Forecasting Methodology and device of a kind of portfolio |
CN107426759A (en) * | 2017-08-09 | 2017-12-01 | 广州杰赛科技股份有限公司 | The Forecasting Methodology and system of newly-increased base station data portfolio |
CN107992988A (en) * | 2017-10-18 | 2018-05-04 | 阿里巴巴集团控股有限公司 | The personnel of client service center have holidays by turns management-control method and device |
CN108668311A (en) * | 2017-03-29 | 2018-10-16 | 中国移动通信集团四川有限公司 | Cell area flow method for early warning and device |
CN108829718A (en) * | 2018-05-07 | 2018-11-16 | 北京奇艺世纪科技有限公司 | A kind of method and apparatus of data processing |
CN109740799A (en) * | 2018-12-18 | 2019-05-10 | 平安科技(深圳)有限公司 | Budget data generation method, device, computer equipment and storage medium |
CN109981327A (en) * | 2017-12-28 | 2019-07-05 | 中移信息技术有限公司 | A kind of prediction technique and system of portfolio |
CN112766529A (en) * | 2019-11-05 | 2021-05-07 | 顺丰科技有限公司 | Logistics daily measurement prediction optimization method, device, equipment and storage medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6246880B1 (en) * | 1998-11-13 | 2001-06-12 | Telefonaktiebolaget Lm Ericsson (Publ) | Determining subscriber demands on a communications system |
US7269599B2 (en) * | 2005-12-01 | 2007-09-11 | International Business Machines Corporation | Method and system for predicting user activity levels associated with an application |
CN101052067A (en) * | 2007-05-22 | 2007-10-10 | 杨苹 | Intelligent traffic predicting method |
CN101541016A (en) * | 2009-05-06 | 2009-09-23 | 华为技术有限公司 | Method for predicting data and equipment |
CN101541030A (en) * | 2009-05-06 | 2009-09-23 | 华为技术有限公司 | Method for predicting data based on support vector machine and equipment thereof |
-
2009
- 2009-12-28 CN CN 200910243569 patent/CN102111284B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6246880B1 (en) * | 1998-11-13 | 2001-06-12 | Telefonaktiebolaget Lm Ericsson (Publ) | Determining subscriber demands on a communications system |
US7269599B2 (en) * | 2005-12-01 | 2007-09-11 | International Business Machines Corporation | Method and system for predicting user activity levels associated with an application |
CN101052067A (en) * | 2007-05-22 | 2007-10-10 | 杨苹 | Intelligent traffic predicting method |
CN101541016A (en) * | 2009-05-06 | 2009-09-23 | 华为技术有限公司 | Method for predicting data and equipment |
CN101541030A (en) * | 2009-05-06 | 2009-09-23 | 华为技术有限公司 | Method for predicting data based on support vector machine and equipment thereof |
Cited By (39)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103178990A (en) * | 2011-12-20 | 2013-06-26 | 中国移动通信集团青海有限公司 | Network device performance monitoring method and network management system |
CN102711177A (en) * | 2012-04-26 | 2012-10-03 | 北京邮电大学 | Service prediction based load balancing method |
CN103514506A (en) * | 2012-06-29 | 2014-01-15 | 国际商业机器公司 | Automatic event analysis method and system |
CN103856344A (en) * | 2012-12-05 | 2014-06-11 | 中国移动通信集团北京有限公司 | Alarm event information processing method and device |
CN103856344B (en) * | 2012-12-05 | 2017-09-15 | 中国移动通信集团北京有限公司 | A kind of alarm event information processing method and device |
CN103024762A (en) * | 2012-12-26 | 2013-04-03 | 北京邮电大学 | Service feature based communication service forecasting method |
CN103024762B (en) * | 2012-12-26 | 2015-04-15 | 北京邮电大学 | Service feature based communication service forecasting method |
CN104125584A (en) * | 2013-04-27 | 2014-10-29 | 中国移动通信集团福建有限公司 | Service index realization prediction method aiming at network service and apparatus thereof |
WO2014190923A1 (en) * | 2013-05-29 | 2014-12-04 | 华为技术有限公司 | Method and system for predicting traffic of cellular network |
CN104219691A (en) * | 2013-05-29 | 2014-12-17 | 华为技术有限公司 | Traffic prediction method and system for cellular network |
CN103338461B (en) * | 2013-06-18 | 2016-03-02 | 中国联合网络通信集团有限公司 | Based on network plan method and the device of Traffic prediction |
CN103338461A (en) * | 2013-06-18 | 2013-10-02 | 中国联合网络通信集团有限公司 | Method and device for network planning based on prediction of volume of business |
CN106164864A (en) * | 2014-03-31 | 2016-11-23 | 微软技术许可有限责任公司 | The prediction load of service is flexible |
CN104978604A (en) * | 2014-04-04 | 2015-10-14 | 中国移动通信集团福建有限公司 | Simulation method based on service capability model and simulation device based on service capability model |
CN104978604B (en) * | 2014-04-04 | 2019-09-17 | 中国移动通信集团福建有限公司 | A kind of analog simulation method and device based on professional ability model |
CN105634787A (en) * | 2014-11-26 | 2016-06-01 | 华为技术有限公司 | Evaluation method, prediction method and device and system for network key indicator |
CN105634787B (en) * | 2014-11-26 | 2018-12-07 | 华为技术有限公司 | Appraisal procedure, prediction technique and the device and system of network key index |
CN105792231A (en) * | 2014-12-23 | 2016-07-20 | 中国移动通信集团上海有限公司 | Telephone traffic forecasting method and device |
CN105792231B (en) * | 2014-12-23 | 2019-03-22 | 中国移动通信集团上海有限公司 | A kind of telephone traffic prediction method and device |
CN104766144A (en) * | 2015-04-22 | 2015-07-08 | 携程计算机技术(上海)有限公司 | Order forecasting method and system |
CN105468774A (en) * | 2015-07-07 | 2016-04-06 | 中华电信股份有限公司 | Data usage prediction apparatus, method and computer program product thereof |
CN105468774B (en) * | 2015-07-07 | 2020-11-17 | 中华电信股份有限公司 | Data usage prediction apparatus, method and system |
CN105138691B (en) * | 2015-09-18 | 2018-10-02 | 北京百度网讯科技有限公司 | Analyze the method and system of subscriber traffic |
CN105138691A (en) * | 2015-09-18 | 2015-12-09 | 北京百度网讯科技有限公司 | Method and system for analyzing user traffic |
WO2017095344A1 (en) | 2015-12-04 | 2017-06-08 | Turkcell Teknoloji Arastirma Ve Gelistirme Anonim Sirketi | A network traffic estimation system |
CN105979538A (en) * | 2016-07-19 | 2016-09-28 | 浪潮软件集团有限公司 | Method for analyzing lost telephone traffic and flow during cell interruption |
CN107092973A (en) * | 2016-11-25 | 2017-08-25 | 口碑控股有限公司 | The Forecasting Methodology and device of a kind of portfolio |
US11443251B2 (en) | 2016-11-25 | 2022-09-13 | Koubei Holding Limited | Method and device for predicting traffic |
CN108668311B (en) * | 2017-03-29 | 2021-07-30 | 中国移动通信集团四川有限公司 | Cell area traffic early warning method and device |
CN108668311A (en) * | 2017-03-29 | 2018-10-16 | 中国移动通信集团四川有限公司 | Cell area flow method for early warning and device |
CN107426759A (en) * | 2017-08-09 | 2017-12-01 | 广州杰赛科技股份有限公司 | The Forecasting Methodology and system of newly-increased base station data portfolio |
CN107426759B (en) * | 2017-08-09 | 2020-06-19 | 广州杰赛科技股份有限公司 | Method and system for predicting data traffic of newly added base station |
CN107992988A (en) * | 2017-10-18 | 2018-05-04 | 阿里巴巴集团控股有限公司 | The personnel of client service center have holidays by turns management-control method and device |
CN109981327A (en) * | 2017-12-28 | 2019-07-05 | 中移信息技术有限公司 | A kind of prediction technique and system of portfolio |
CN108829718B (en) * | 2018-05-07 | 2021-04-06 | 北京奇艺世纪科技有限公司 | Data processing method and device |
CN108829718A (en) * | 2018-05-07 | 2018-11-16 | 北京奇艺世纪科技有限公司 | A kind of method and apparatus of data processing |
CN109740799A (en) * | 2018-12-18 | 2019-05-10 | 平安科技(深圳)有限公司 | Budget data generation method, device, computer equipment and storage medium |
CN109740799B (en) * | 2018-12-18 | 2024-02-02 | 平安科技(深圳)有限公司 | Budget data generation method, device, computer equipment and storage medium |
CN112766529A (en) * | 2019-11-05 | 2021-05-07 | 顺丰科技有限公司 | Logistics daily measurement prediction optimization method, device, equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN102111284B (en) | 2013-09-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102111284B (en) | Method and device for predicting telecom traffic | |
CN101159968B (en) | GSM mobile network extension calculating method | |
Zhang et al. | Citywide cellular traffic prediction based on densely connected convolutional neural networks | |
US10733642B2 (en) | Service management system that enables subscriber-driven changes to service plans | |
Yamakawa et al. | The diffusion of mobile telephones: An empirical analysis for Peru | |
CN102270363B (en) | Intelligent queuing system | |
US9331892B2 (en) | System and method for automatic sub-panel creation and management | |
CN103150696A (en) | Method and device for selecting potential customer of target value-added service | |
Lee et al. | Forecasting mobile broadband traffic: Application of scenario analysis and Delphi method | |
US8762359B2 (en) | Method of analyzing data traffic in a telecommunication network | |
CN105657201A (en) | Method and system for processing call based on decision tree model | |
CN107908680A (en) | Management method, electronic device and the computer-readable recording medium of wechat public platform | |
Aboolian et al. | Location and allocation of service units on a congested network | |
US11716422B1 (en) | Call center load balancing and routing management | |
Hohwald et al. | User modeling for telecommunication applications: Experiences and practical implications | |
CN110233741A (en) | Service charging method, device, equipment and storage medium | |
CN110378781A (en) | Data monitoring method, device, computer equipment and storage medium | |
CN102436506A (en) | Mass data processing method and device for network server | |
CN106302770B (en) | The based reminding method and system of business information | |
CN106412796B (en) | A kind of recommended method and system | |
CN113852723B (en) | Number scheduling method, equipment and storage medium | |
CN101080037B (en) | Method and device for preparing mobile content | |
Jha et al. | Diffusion and Forecast of Mobile Service Generations in Germany, UK, France and Italy–A Comparative Analysis Based on Bass, Gompertz and Simple Logistic Growth Models | |
CN108428009A (en) | A kind of channel distribution method and device | |
Sultan et al. | Mobile phone price as a proxy for socio-economic indicators |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right | ||
TR01 | Transfer of patent right |
Effective date of registration: 20190509 Address after: 100093 Haidian District, Haidian District, Xing Shi Kou Road, No. 1 20302, 20302 Patentee after: Beijing BOCO Inter-Telecom Technology Co., Ltd. Address before: 100093 AB Block 99 Xingshikou Road, Haidian District, Beijing Co-patentee before: Yiyang Xintong Co., Ltd. Patentee before: Beijing BOCO Inter-Telecom Technology Co., Ltd. |