Anomaly Detection and Early Warning Model for Latency in Private 5G Networks

Different from previous generations of communication technology, 5G has tailored several modes especially for industrial applications, such as Ultra-Reliable Low-Latency Communications (URLLC) and Massive Machine Type (mMTC). The private networks require high performance latency, bandwidth, reliability, while the deployment environment is usually complicated, causing network problems difficult to identify. This poses a challenge operation maintenance (O&M) networks. It needed quickly diagnose or predict faults based on high-dimensional data services reduce impact services. paper proposes ConvAE-Latency model anomaly detection, which enhances correlation between target indicators hidden features by multi-target learning. Meanwhile, transfer learning applied prediction in proposed LstmAE-TL solve problem unbalanced samples. Based China Telecom platform, models are deployed tested an Automated Guided Vehicles (AGVs) application scenario. results have been improved compared existing research.