Communication-Efficient Online Detection of Network-wide Anomalies

Overview:

In this project, we design a system and propose a novel approximation scheme for continuously online anomaly detection that dramatically reduces the burden on the production network. Our algorithm leverages intelligent data filtering at distributed monitors and Principal Component Analysis (PCA) for detection at NOC. We derive analytical results based on stochastic matrix perturbation theory to effectively balance the tradeoff between detection accuracy and the amount of data communicated over the network. By avoiding the expensive step of centralizing all traffic data, our solution enables tracking PCA-based anomalies in real time with minimal data communications. This overcomes the key scalability limitations of the state-of-the-art network-wide anomaly detection solution. Experiments with traffic data from an ISP-backbone network demonstrate that our methods yield significant communication benefits while simultaneously achieving high detection accuracy.

Publications:

Talks:

People: