A method, device and storage medium for predicting traffic trend based on historical traffic

Abstract

The present invention provides a method, device, and storage medium for predicting traffic trends based on historical traffic, including the steps of: reading historical traffic data information, the historical traffic data includes actual traffic data of multiple historical unit time periods, and the multiple Associate the actual flow data of a historical unit time period; determine the effective historical flow data information; determine the forecast flow data of the forecast unit time period, the forecast flow data of the forecast unit time period and the actual flow data of the multiple historical unit time periods Associated traffic data. The invalid historical flow data is eliminated to eliminate the influence of the invalid historical abnormal flow data on the calculation results, and then the prediction time point is determined and the flow data is predicted to improve the accuracy of the flow forecast.

Description

Technical field: [0001] The present invention relates to the communication field, in particular to a method, device and storage medium for predicting traffic trends based on historical traffic. Background technique: [0002] With the development of network communication technology, there are more and more types of network traffic data and services, and people's demand for network traffic is increasing day by day. Network traffic forecasting is helpful for analyzing network security status, scientifically managing the network and preventing improper network behaviors. Existing technologies generally use the following two methods for forecasting: [0003] 1. Read the historical flow data at the same minute point t in the previous n days, and then calculate the arithmetic mean value of the flow rate at the t-minute point in the previous n days, and use this value as a simple forecast value of the t-minute point flow rate on the n+1 day; [0004] 2. Read the historical traffic ...

AI Technical Summary

Problems solved by technology

[0005] The two technical solutions mentioned above did not consider the processing of historical abnormal traffic data when performing traffic forecasting calculations. The participation of these abnormal traffic data in traffic forecasting calculations will lead to a large deviation between the calculation results and normal traffic, and eventually lead to traffic forecasting. Inaccurate;

Images