Machine Learning-Enhanced Anomaly Detection and Performance Optimization for Enterprise WiFi Networks

solutions for large-scale WiFi deployments where the performance of the network changes dynamically. The industry partner has an enterprise WiFi solution that collects Key Performance Indicators (KPIs), logs and WiFi configuration parameters in the cloud. There is an immediate need for automation platforms that can use these data to detect anomalies such as sudden performance degradation, understand the reasons of such poor performance and change programmable configuration parameters to mitigate the problem. This project will use machine learning algorithms to detect anomalies, perform automated root cause analysis to bring self-healing features to WiFi, as well as, developing machine learning based self-optimization solutions that will reconfigure the WiFi parameters. The project will bring competitive edge to the industry partner within the emerging market of telemetry and autonomous networking.

Samhita Kuili;Ahmed Omara;Mohammad Sadeghi
Faculty Supervisor: 
Melike Erol-Kantarci;Burak Kantarci
Partner University: