A service prediction method, network element device, and computer-readable storage medium

Abstract

The invention discloses a service prediction method, network element equipment, and a computer-readable storage medium, wherein the method includes: obtaining a first service load value according to historical service load data; obtaining a second service load value according to the current day service load data ;According to the first business load value and the second business load value, the business load forecast value is obtained; when the business load forecast value is the forecast value of the previous period of the day that is not this period, the forecast value of the previous period Perform difference calculation with the obtained actual business load to obtain the difference; if the difference is less than or equal to the threshold value, use the operation parameters of the previous period to predict the business load in this period, otherwise, the The operation parameters in the previous period are adjusted, and the business load is predicted in this period according to the adjusted operation parameters.

Description

technical field [0001] The present invention relates to communication technology, in particular to a service prediction method, network element equipment, and a computer-readable storage medium. Background technique [0002] With the development of mobile communication technology and the increasing number of mobile communication users in recent years, the service types and service rates provided by mobile communication to users have also increased significantly. Taking 4G network as an example, the instantaneous downlink rate that users can enjoy has exceeded 100M. The improvement of user experience and the increase in the number of users have also led to an increase in network service load. According to statistics, by 2020, the average speed of the global mobile network will increase by 3.2 times compared with 2015 (2.0Mbps), reaching 6.5Mbps. Global 4G adoption is the main catalyst for the increase in mobile speeds. The analysis also shows that by 2020: 4G connections wi...

AI Technical Summary

Problems solved by technology

[0006] The disadvantages of using the above business forecasting method are: 1) In the multiple linear regression operation, it is considered that all points in the input sequence have the same weight, ignoring the characteristics of periodic changes in business volume, and the regression analysis prediction method is based on the input sequence For one-way forecasting, there is no adaptive and self-learning correction of the forecast results. When the forecast results deviate, it cannot be automatically adjusted to obtain better forecast accuracy

However, in the same way, the forecasting object is always unable to break through the historical peak or trough during the forecast period

The existence of this defect will also produce larger and larger errors for the sequence prediction of a single trend

Images