Enhanced Classification For Intrusion Detection In WSNS Using Bi Directional RNNS

Abstract

This paper presents a robust Intrusion Detection System (IDS) for Wireless Sensor Networks (WSNs)based on a Bi-directional Recurrent Neural Network (Bi-RNN) model that takes advantage of the temporaldependencies inherent to network traffic to enhance anomaly detection. Although other models can beused to examine time series data of traffic, The Bi-RNN architecture examines input sequences in bothforward and backward fashions. This negates potential subsumption and allows the system to recognizemore unique patterns and contextual backgrounds