802.11 Wireless Simulation and Anomaly Detection using HMM and UBM

TL;DR

This paper proposes a novel anomaly detection method for 802.11 wireless networks using Hidden Markov Models and Universal Background Models, demonstrating improved detection accuracy over baseline methods through simulation experiments.

Contribution

It introduces a new anomaly detection approach based on HMM and UBM using easily obtainable access point data, enhancing detection performance in WLANs.

Findings

HMM and UBM models outperform baseline methods in detection precision.

Simulation results confirm higher sensitivity of the proposed models.

The approach reduces the need for heavy instrumentation of user devices.

Abstract

Despite the growing popularity of 802.11 wireless networks, users often suffer from connectivity problems and performance issues due to unstable radio conditions and dynamic user behavior among other reasons. Anomaly detection and distinction are in the thick of major challenges that network managers encounter. Complication of monitoring the broaden and complex WLANs, that often requires heavy instrumentation of the user devices, makes the anomaly detection analysis even harder. In this paper we exploit 802.11 access point usage data and propose an anomaly detection technique based on Hidden Markov Model (HMM) and Universal Background Model (UBM) on data that is inexpensive to obtain. We then generate a number of network anomalous scenarios in OMNeT++/INET network simulator and compare the detection outcomes with those in baseline approaches (RawData and PCA). The experimental results show the superiority of HMM and HMM-UBM models in detection precision and sensitivity.