Analyzing and forecasting network traffic

Patent metadata

Field	Value
Publication number	US-9451473-B2
Application number	US-201414247798-A
Country	US
Kind code	B2
Filing date	Apr 8, 2014
Priority date	Apr 8, 2014
Publication date	Sep 20, 2016
Grant date	Sep 20, 2016

Abstract

Systems and methods for analyzing and forecasting network traffic are disclosed. In some implementations, multiple points are received. Each point represents a time value and a network traffic value. For each point in the multiple points, the network traffic value is decomposable into a trend component, a seasonality component, a burst component, and a random error component. A trend equation is generated for calculating the trend component based on the time value. A seasonality equation is generated for calculating the seasonality component based on the time value. A burst equation is generated for calculating the burst component based on the time value. A random error equation is generated for calculating the random error component. A network traffic equation is generated for calculating the network traffic value based on the time value by combining the trend equation, the seasonality equation, the burst equation, and the random error equation.

First claim

What is claimed is:

1. A method comprising:
receiving, at a computing device, a plurality of points, each point representing a time value and a network traffic value representative of traffic on a communication network, the network traffic values and the time values represented by the plurality of points decomposable into a trend component, a seasonality component, a burst component, and a random error component;
generating, using machine learning or data mining techniques, a trend equation for calculating the trend component, the trend component taking into account a long-term trend of the network traffic values and one or more sharp changes in network traffic, the sharp changes followed by stable periods and caused by predictable bursts or seasonal network traffic patterns but not accounted for in the seasonality component due to a periodicity of the seasonality component and not accounted for in the burst component due to a granularity of the burst component, the sharp changes taken into account in the trend component by a compensation formula applied to the trend component;
generating, using machine learning or data mining techniques, a seasonality equation for calculating the seasonality component;
generating, using machine learning or data mining techniques, a burst equation for calculating the burst component;
generating a random error equation for calculating the random error component;
generating a network traffic equation for predicting future network traffic values corresponding to future time values by combining the trend equation, the seasonality equation, the burst equation, and the random error equation; and
providing, via the computing device to a network administrator responsible for the communication network:
an indication of a current network capacity,
an indication of a projected network usage based on the generated network traffic equation,
an indication that the communication network is expected to operate above capacity at a future time corresponding to at least one of the future time values based on a difference between the current network capacity and the projected network usage, and
a recommendation including an action for the network administrator to take to improve the current network capacity to meet the projected network usage prior to the future time.

2. The method of claim 1 , wherein a granularity of network traffic associated with the network traffic equation comprises one or more of a granularity per: cell, NodeB base station, radio network controller (RNC), RNC cluster, mobile switching center (MSC) or entire cellular network.

3. The method of claim 1 , further comprising determining, using the generated network traffic equation, that a predicted future network traffic value exceeds a network capacity, wherein the action included within the recommendation includes increasing the network capacity to a level sufficient to handle the predicted future network traffic value.

4. The method of claim 1 , wherein:
the trend component corresponds to a long-term trend of the network traffic value;
the seasonality component corresponds to a periodicity of the network traffic value; and
the burst component corresponds to a predictable outlier of the network traffic value from a combination of the trend component and the seasonality component.

5. The method of claim 4 , wherein the long-term trend is a monthly trend or a yearly trend.

6. The method of claim 4 , wherein the periodicity of the network traffic value is seven days.

7. The method of claim 1 , wherein the random error component cannot be explained using the trend component, the seasonality component, and the burst component.

8. The method of claim 1 , wherein the random error component among the plurality of points has a normal distribution.

9. The method of claim 1 , wherein the network traffic value comprises a long term evolution (LTE) network traffic value.

10. A non-transitory computer-readable medium comprising instructions which, when executed by a computer, cause the computer to:
receive a plurality of points, each point representing a time value and a network traffic value representative of traffic on a communication network, the network traffic values and the time values represented by the plurality of points decomposable into a trend component, a seasonality component, a burst component, and a random error component;
generate, using machine learning or data mining techniques, a trend equation for calculating the trend component, the trend component taking into account a long-term trend of the network traffic values and one or more sharp changes in network traffic, the sharp changes followed by stable periods and caused by predictable bursts or seasonal network traffic patterns but not accounted for in the seasonality component due to a periodicity of the seasonality component and not accounted for in the burst component due to a granularity of the burst component, the sharp changes taken into account in the trend component by a compensation formula applied to the trend component;
generate, using machine learning or data mining techniques, a seasonality equation for calculating the seasonality component;
generate, using machine learning or data mining techniques, a burst equation for calculating the burst component;
generate a random error equation for calculating the random error component;
generate a network traffic equation for predicting future calculating the network traffic values corresponding to future time values by combining the trend equation, the seasonality equation, the burst equation, and the random error equation; and
provide, to a network administrator responsible for the communication network:
an indication of a current network capacity,
an indication of a projected network usage based on the generated network traffic equation,
an indication that the communication network is expected to operate above capacity at a future time corresponding to at least one of the future time values based on a difference between the current network capacity and the projected network usage, and
a recommendation including an action for the network administrator to take to improve the current network capacity to meet the projected network usage prior to the future time.

11. The computer-readable medium of claim 10 , further comprising instructions which, when executed by the computer, cause the computer to determine, using the generated network traffic equation, that a predicted future network traffic value exceeds a network capacity, wherein the action included within the recommendation includes increasing the network capacity to a level sufficient handle the predicted future network traffic value.

12. The computer-readable medium of claim 10 , wherein:
the trend component corresponds to a long-term trend of the network traffic value;
the seasonality component corresponds to a periodicity of the network traffic value; and
the burst component corresponds to a predictable outlier of the network traffic value from a combination of the trend component and the seasonality component.

13. The computer-readable medium of claim 12 , wherein the long-term trend is a monthly trend or a yearly trend.

14. The computer-readable medium of claim 12 , wherein the periodicity of the network traffic value is seven days.

15. The computer-readable medium of claim 10 , wherein the random error component cannot be explained using the trend component, the seasonality component, and the burst component.

16. The computer-readable medium of claim 10 , wherein the random error component among the plurali

Assignees

• Cellco Partnership Dba Verizon

Inventors

• Ouyang Ye
• Becht Carol
• Venkatasubramaniam Gopinath

Classifications

• H04W16/00

  Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures · CPC title

• H04W16/22 Primary

  Traffic simulation tools or models · CPC title

• H04L47/127

  by using congestion prediction · CPC title