An Improved ARIMA-Based Traffic Anomaly Detection Algorithm for Wireless Sensor Networks

Traffic anomaly detection is emerging as a necessary component as wireless networks gain popularity. In this paper, based on the improved Autoregressive Integrated Moving Average (ARIMA) model, we propose a traffic anomaly detection algorithm for wireless sensor networks (WSNs) which considers the particular imbalanced, nonstationary properties of the WSN traffic and the limited energy and computing capacity of the wireless sensors at the same time. We systematically analyze the characteristics of WSN traffic, the causes of WSN abnormal traffic, and the latest related research and development. Specifically, we improve the traditional time series ARIMA model to make traffic prediction and judge the traffic anomaly in a WSN. Simulated and real WSN traffic data gathered from University of North Carolina are used to carry out simulations on Matlab. Simulation results and comparative analyses demonstrate that our proposed WSN traffic anomaly detection scheme has better anomaly detection accuracy than traditional traffic anomaly detection algorithms.