A Hybrid Deep Learning-Driven SDN Enabled Mechanism for Secure Communication in Internet of Things (IoT).

Danish Javeed,Tianhan Gao,Muhammad Taimoor Khan,Ijaz Ahmad

doi:10.3390/s21144884

Abstract

The Internet of Things (IoT) has emerged as a new technological world connecting billions of devices. Despite providing several benefits, the heterogeneous nature and the extensive connectivity of the devices make it a target of different cyberattacks that result in data breach and financial loss. There is a severe need to secure the IoT environment from such attacks. In this paper, an SDN-enabled deep-learning-driven framework is proposed for threats detection in an IoT environment. The state-of-the-art Cuda-deep neural network, gated recurrent unit (Cu- DNNGRU), and Cuda-bidirectional long short-term memory (Cu-BLSTM) classifiers are adopted for effective threat detection. We have performed 10 folds cross-validation to show the unbiasedness of results. The up-to-date publicly available CICIDS2018 data set is introduced to train our hybrid model. The achieved accuracy of the proposed scheme is 99.87%, with a recall of 99.96%. Furthermore, we compare the proposed hybrid model with Cuda-Gated Recurrent Unit, Long short term memory (Cu-GRULSTM) and Cuda-Deep Neural Network, Long short term memory (Cu- DNNLSTM), as well as with existing benchmark classifiers. Our proposed mechanism achieves impressive results in terms of accuracy, F1-score, precision, speed efficiency, and other evaluation metrics.

Highlights

IntroductionThere has been an enormous growth in the Internet of Things (IoT), described as a global network of interconnected devices that are assigned unique addresses
In recent years, there has been an enormous growth in the Internet of Things (IoT), described as a global network of interconnected devices that are assigned unique addresses.IoT devices use different communication protocols and sensing features
For a thorough performance evaluation of our proposed hybrid model, we made the comparison of our model with our constructed two deep learning (DL)-driven hybrid models, i.e., Cu-GRULSTM and DNNLSTM, and with existing literature

Summary

Introduction

There has been an enormous growth in the Internet of Things (IoT), described as a global network of interconnected devices that are assigned unique addresses. IoT devices use different communication protocols and sensing features. These devices have computational abilities to analyze data and provide services. IoT contains heterogeneous and homogeneous networks with networking devices that use different types of protocols. It means that vulnerabilities can produce an imperceptible threat to IoT devices and the entire system. Cybersecurity exploits numerous concerns in the dynamic features of these devices in the form of different attacks, i.e., DoS attacks, DDoS attacks, and some other types of malware [4]. IoT security remains a significant challenge and poses a severe need for security

Objectives

Methods

Results

Conclusion