Deep Learning for Intrusion Detection

10Deep Learning-Based Intrusion Detection in Wireless Networks

Rahila Rahim and Mohammad Ahsan Chishti

Department of Computer Science and Technology, National Institute of Technology Srinagar, Srinagar Jammu & Kashmir, India

10.1 Introduction

Wireless networks have already become an integral part of modern communication systems, providing seamless connections between various locations. Wireless networks transmit data over radio frequency waves instead of through physical wires, as is done in wired networks. While that distinction creates tremendous freedom, it also creates unique security challenges, requiring imaginative solutions, especially as cyber threats increasingly evolve.

Focusing on security applications, deep learning (DL) techniques with remarkable quality of self-learning are beneficial to develop the intrusion detection models that are used to discover malicious nodes or network anomalies [9]. These DL models result in low false rates and high accuracy as compared to traditional machine learning techniques. Similarly, recurrent neural networks (RNNs) and long short-term memory (LSTM) networks come under the category of supervised sequence learning methods and are the most popular methods used for detecting malware activities with self-learning techniques. These analyze temporal data to detect abnormalities, such as routing attacks (e.g., black hole, grey hole, or wormhole) in MANETS. Artificial neural networks (ANNs), deep neural networks (DNNs), convolutional ...