Abstract: 

Communications traffic on wireless networks generates large volumes of metadata on a continuous basis across the various servers involved in the communication session. Since these networks are engineered for high reliability, the data is predominantly normal with only a small proportion of the data being anomalous. It is, however, important to detect these anomalies when they occur because such anomalies are indicators of vulnerabilities in the network. In this work we will present the use of neural network based Kohonen Self Organizing Maps (SOM) and visual analytics for network anomaly detection and analysis using data from a 4G wireless network.

Bio: 

TBD