Anomaly detection (AD) is a solution for cellular operators to overcome the difficulty of quality control over the proliferation of cellular phone usage. The telecommunications network monitoring system with anomaly detection enables immediate discovery of problems before they become more complex. Monitoring activities produce logs, which are then analyzed according to the interest, often with the help of statistics and visualizations. Relying on humans for analysis is increasingly difficult due to the immense amount of logs in modern telecommunication networks. This work extends the monitoring to automated anomaly detection by using various ELK modules to form an intelligent monitoring system. A testbed based on OpenAirInterface (OAI) and USRP B210 radiohead is used, which includes the functionalities of HSS, MME, SGW, PGW, eNB, and UE. The proposed system has an average accuracy of 91.5%. This is supported by an average value of the proportion of normal conditions that are correctly predicted at 99.31%. On the other hand, the system can still maintain the functionality of the cellular telecommunications network with an excellent predicate on service quality.
- Anomaly detection
- Elasticsearch-Logstash-Kibana stack
- Long-Term Evolution
- Telecommunication network