Using Intuitionistic Fuzzy Set for Anomaly Detection of Network Traffic from Flow Interaction

TL;DR

This paper introduces a novel anomaly detection method for network traffic based on complex network modeling of flow interactions, utilizing intuitionistic fuzzy sets and ensemble techniques to improve detection accuracy.

Contribution

The paper proposes a new approach combining complex network analysis with intuitionistic fuzzy sets and ensemble methods for anomaly detection in network traffic.

Findings

Outperforms state-of-the-art anomaly detection methods.

Effectively detects anomalies in various network traffic datasets.

Demonstrates robustness against inconsistent network behavior.

Abstract

We present a method to detect anomalies in a time series of flow interaction patterns. There are many existing methods for anomaly detection in network traffic, such as number of packets. However, there is non established method detecting anomalies in a time series of flow interaction patterns that can be represented as complex network. Firstly, based on proposed multivariate flow similarity method on temporal locality, a complex network model (MFS-TL) is constructed to describe the interactive behaviors of traffic flows. Having analyzed the relationships between MFS-TL characteristics, temporal locality window and multivariate flow similarity critical threshold, an approach for parameter determination is established. Having observed the evolution of MFS-TL characteristics, three non-deterministic correlations are defined for network states (i.e. normal or abnormal). Furthermore, intuitionistic fuzzy set (IFS) is introduced to quantify three non-deterministic correlations, and then a anomaly detection method is put forward for single characteristic sequence. To build an objective IFS, we design a Gaussian distribution-based membership function with a variable hesitation degree. To determine the mapping of IFS's clustering intervals to network states, a distinction index is developed. Then, an IFS ensemble method (IFSE-AD) is proposed to eliminate the impacts of the inconsistent about MFS-TL characteristic to network state and improve detection performance. Finally, we carried out extensive experiments on several network traffic datasets for anomaly detection, and the results demonstrate the superiority of IFSE-AD to state-of-the-art approaches, validating the effectiveness of our method.