核心网用户面异常流量智能化检测创新方案与实践

打开文本图片集

关键词：核心网；ResNet；LSTM；simpleCNN；异常检测；自动化

doi：10.3969/J.ISSN.1672-7274.2025.04.034

中图分类号：TN929.53 文献标志码：B 文章编码：1672-7274（2025）04-0099-03

Abstract： With the rapid development of big data，cloud computing，artificial intellgence，and 5G technology， network applications have become increasingly widespread and convenient.By 2O24，the number of Internet users inthe World willreach6bilion，anincreaseof3.3billionover2014，and the Internet penetrationratehas reached 76% .This huge scale of internet users has brought a massive amount of online user traffc，including both normal and abnormal traffic.Therefore，the detection technologyof abnormal user trafc is crucial for improving network information security. As a mainstream core network equipment provider， ZTE is commited to providing innovative technologies and solutions for global telecommunications operators，consumers，and more.By combining ResNet （Residual Network），LSTM（Long Short Term Memory），and SimpleCNN（Simple Convolutional Neural Network） neural network models，ZTE has successfully achieved automated anomaly detection of user plane data streams， which is expected to provide some help for the digitization of core network delivery and operation.

Keywords： core network; ResNet; LSTM; simpleCNN; anomaly detection; automation

研究背景

动化技术显得尤为迫切。（剩余3935字）