ELECTRONIC DEVICE AND METHOD FOR DETECTING ANOMALY OF TELECOMMUNICATION NETWORK BASED ON AUTOENCODER NEURAL NETWORK MODEL

An electronic device and a method for detecting anomaly of a telecommunication network based on AutoEncoder neural network model is provided. The method includes: retrieving a plurality of current signals from the telecommunication network; collecting the plurality of current signals and sorting the...

Full description

Saved in:
Bibliographic Details
Main Authors WANG, SHUN-HSIEN, KAO, CHIH-YUAN, YANG, YAO-TSUNG, HSU, SHIHUN, YU, SHENG-WANG
Format Patent
LanguageChinese
English
Published 21.10.2022
Subjects
CALCULATING
COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
COMPUTING
COUNTING
PHYSICS
Online AccessGet full text

Cover

More Information
Summary:An electronic device and a method for detecting anomaly of a telecommunication network based on AutoEncoder neural network model is provided. The method includes: retrieving a plurality of current signals from the telecommunication network; collecting the plurality of current signals and sorting the plurality of current signals according to time; utilizing the plurality of current signals sorted according to the time to obtain a current feature vector associated with a plurality of characteristics; and inputting the current feature vector to an AutoEncoder neural network model to obtain a current weighted loss value, utilizing the current weighted loss value to determine whether a current operating state of the telecommunications network is normal, and generating an anomaly report in response to determining that the current operating state is abnormal.
Bibliography:Application Number: TW202211102248