Abstract:The emergence of 5G technology marks a significant milestone in developing telecommunication networks, enabling exciting new applications such as augmented reality and self-driving vehicles. However, these improvements bring an increased management complexity and a special concern in dealing with failures, as the applications 5G intends to support heavily rely on high network performance and low latency. Thus, automatic self-healing solutions have become effective in dealing with this requirement, allowing a learning-based system to automatically detect anomalies and perform Root Cause Analysis (RCA). However, there are inherent challenges to the implementation of such intelligent systems. First, there is a lack of suitable data for anomaly detection and RCA, as labelled data for failure scenarios is uncommon. Secondly, current intelligent solutions are tailored to LTE networks and do not fully capture the spatio-temporal characteristics present in the data. Considering this, we utilize a calibrated simulator, Simu5G, and generate open-source data for normal and failure scenarios. Using this data, we propose Simba, a state-of-the-art approach for anomaly detection and root cause analysis in 5G Radio Access Networks (RANs). We leverage Graph Neural Networks to capture spatial relationships while a Transformer model is used to learn the temporal dependencies of the data. We implement a prototype of Simba and evaluate it over multiple failures. The outcomes are compared against existing solutions to confirm the superiority of Simba.

What problem does this paper attempt to address?

The paper primarily proposes a novel solution for anomaly detection and root cause analysis (RCA) in 5G Radio Access Networks (5G RAN). ### Main Issues With the development of 5G technology and its applications in fields such as augmented reality and autonomous vehicles, the demand for network performance and low latency is increasing. However, these advancements also bring increased management complexity and special attention to fault handling. To address these issues, automated self-healing solutions become effective means, capable of automatically detecting anomalies and performing root cause analysis based on learning systems. ### Challenges 1. **Lack of Data**: There is insufficient data suitable for anomaly detection and RCA, especially labeled fault scenario data. 2. **Limitations of Existing Solutions**: Current intelligent solutions mainly target LTE networks and cannot fully capture the spatiotemporal characteristics of 5G data. ### Solution The paper proposes Simba, an advanced method combining Graph Neural Networks (GNN) and Transformer models for anomaly detection and root cause analysis in 5G RAN. - **Data Generation**: Researchers used the calibrated simulator Simu5G to generate an open-source dataset under normal operation and fault scenarios. - **Model Design**: The Simba model consists of three main parts: - Graph Learning Module (GL), for learning the graph representation of base stations; - Graph Convolution Module (GCN), for learning the spatial embeddings of base stations; - Transformer Module, for learning the temporal dependencies of base stations. - **Experimental Evaluation**: Researchers conducted extensive evaluations of Simba, demonstrating its superiority over existing solutions in various fault scenarios. ### Core Contributions 1. **Innovative Architecture**: Proposes a new architecture combining time-series Transformer with GNN spatial embeddings, effectively detecting and locating faults in 5G RAN. 2. **Multi-Fault Detection**: For the first time, proposes an RCA model capable of detecting and locating multiple simultaneous faults in 5G networks. 3. **Performance Evaluation**: Conducts a comprehensive evaluation of existing models and compares their results with the proposed model. 4. **Open Resources**: Researchers share the developed model and generated data to promote reproducibility and extensibility. Through the above work, Simba aims to improve the reliability of 5G networks, reduce operational costs, and support the needs of emerging applications.

Root Cause Analysis of Anomalies in 5G RAN Using Graph Neural Network and Transformer

Anomaly Detection and Root Cause Analysis Enabled by Artificial Intelligence

On the Use of Spatial Graphs for Performance Degradation Root-Cause Analysis Toward Self-Healing Mobile Networks

Deep RAN: A Scalable Data-driven platform to Detect Anomalies in Live Cellular Network Using Recurrent Convolutional Neural Network

Neuromorphic AI Empowered Root Cause Analysis of Faults in Emerging Networks

Root cause analysis of network fault based on random forest

A Calibrated and Automated Simulator for Innovations in 5G

A Fault Diagnosis Method for 5G Cellular Networks Based on Knowledge and Data Fusion

Anomaly Detection for Cellular Networks Using Big Data Analytics.

Self-Organizing Cellular Radio Access Network with Deep Learning

Deep Learning-Based Log Anomaly Detection for 5G Core Network.

NetRCA: An Effective Network Fault Cause Localization Algorithm

Deep Learning-based Anomaly Detection and Log Analysis for Computer Networks

AI Empowered Net-RCA for 6G

Machine Learning based Anomaly Detection for 5G Networks

Big Data-driven Automated Anomaly Detection and Performance Forecasting in Mobile Networks

Call Detail Records Driven Anomaly Detection and Traffic Prediction in Mobile Cellular Networks

TRACTOR: Traffic Analysis and Classification Tool for Open RAN

Data-Driven Root-Cause Analysis For Distributed System Anomalies

Network Anomaly Detection in 5G Networks

A Neural Network Approach for Wireless Spectrum Anomaly Detection in 5G-Unlicensed Network