Local area network traffic prediction method based on improved ConvLSTM deep learning model

The invention discloses a local area network link traffic prediction method based on an improved ConvLSTM deep learning model, and relates to the field of network traffic prediction. The method aims at solving the problems that a target network flow prediction method based on a traditional linear mo...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: LAI JUNYU, CHEN ZHIYONG, YU CHANGJIANG, PI CHANGHONG, ZHU JUNHONG, MA WANYI
Format: Patent
Sprache:chi ; eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The invention discloses a local area network link traffic prediction method based on an improved ConvLSTM deep learning model, and relates to the field of network traffic prediction. The method aims at solving the problems that a target network flow prediction method based on a traditional linear model and a non-linear model is generally low in prediction accuracy, target network flow space features cannot be predicted at the same time, and all link flows of a target network cannot be predicted through a single model. The invention provides a local area network traffic prediction method based on an improved ConvLSTM deep learning model. A similar residual structure module and a Squeeze-and-Excitation module based on an attention mechanism are added into the model, so that the purpose of accurately and quickly simulating and twinning target network link traffic is finally achieved. According to the method, the precision of local area network time-space traffic matrix prediction is improved, the number of train