Traffic Verification for Network Anomaly Detection in Sensor Networks

The traffic that is being injected to the network is increasing every day. It can be either normal or anomalous. Anomalous traffic is variation in the communication pattern from the normal one and hence anomaly detection is an important procedure in ensuring network resiliency. Probabilistic models can be used to model traffic for anomaly detection. In this paper, we use Gaussian Mixture Model for traffic verification. The traffic is captured and is given to the model to verification. Traffic which obeys the model is normal and those which disobey are anomalies. Analysis shows that the proposed system has better performance in terms of delay, throughput and packet delivery ratio