Intrusion Detection in IoT devices using Zero Bias DNN

Shajan, Aksa Anna (2021) Intrusion Detection in IoT devices using Zero Bias DNN. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (780kB) | Preview
Preview	PDF (Configuration manual) Download (694kB) | Preview

Abstract

By providing smart services in a way that would otherwise be inconceivable, the Internet of Things has become an unavoidable aspect of today's life. IoT devices generally have only one privilege level, both externally and internally, so once this privilege is achieved, it becomes easy for attackers to exploit the vulnerabilities. Because of the huge demand and privilege weaknesses, attackers aim to exploit flaws to launch attacks like DDoS attacks, so ensuring security becomes difficult. It's vital to identify malicious activities effectively. Insecure IoT devices can cause a tremendous impact in a variety of ways. Among the various techniques used for intrusion detection, machine learning approaches proves to be best with better results. Developing an efficient and effective intrusion detection system is challenging as attackers come with a variety of attack patterns each day. This study proposes a new machine learning method - zero bias deep neural network for identifying intrusions in IoT devices. A dataset named MQTT-IoT IDS 2020, a latest dataset on IoT normal and attack patterns, from IEEEPort, is used in our model. The previous research papers on IoT IDS have used datasets that are used in all general intrusion cases of network attacks, not particularly for IoT. SMOTE oversampling technique has been used to resolve the data imbalance and Random Forest classifier algorithm was used for feature elimination. Our experiment shows that zero bias DNN produces a satisfactory result with an accuracy of 92% on our dataset.

Item Type:	Thesis (Masters)
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > QA Mathematics > Computer software > Computer Security T Technology > T Technology (General) > Information Technology > Computer software > Computer Security T Technology > TK Electrical engineering. Electronics. Nuclear engineering > Telecommunications > Computer networks > Internet of things Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Cyber Security
Depositing User:	Tamara Malone
Date Deposited:	29 Dec 2022 15:54
Last Modified:	07 Mar 2023 12:14
URI:	https://norma.ncirl.ie/id/eprint/6057

Actions (login required)

View Item