Malware detection using Conventional Neural Network and Regression on smartwatches

Venkateswara, Raakesh babu (2022) Malware detection using Conventional Neural Network and Regression on smartwatches. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (656kB) | Preview
Preview	PDF (Configuration manual) Download (498kB) | Preview

Abstract

Wireless wearable gadgets have become more widely known. Despite the fact that all these gadgets have minimal computing capability, the data they gather is sensitive. Wearable Internet of Things (WIoT) gadgets may function as both a stand-alone platform as well as an extension of smartphones and other mobile devices. Because the WIoT has computing, storage, and capacity restrictions, such gadgets can indeed be interconnected with an intrusion detection system. Such WIoT may pose a serious danger in the near future simply because they are open and easily available to both the customer and the intruder, which makes them more vulnerable. An intrusion into smart wearable gadgets might have disastrous consequences for a victim or a massive network. Wearable devices are largely neglected as a crucial element of a network. Nonetheless, this might have a negative impact on privacy and data security. This research will give an intrusion detection method for wearable smart IoT (WIoT) gadgets by evaluating the WSN-DS dataset by using a Convolutional Neural Network (CNN) and regression model that may be utilized to identify normal and unusual network activity with great accuracy and minimal complication. In addition to the Wireless Body Area Network (WBAN), the Wireless Sensor Network (WSN) represents the most intimately linked network to smart wearable IoT devices (WBAN). A compact model is built and compared with a single CNN model using a database developed on Wireless Sensor Network (WSN) to identify unusual cases of attack with limited processing capabilities that can be applied in tiny IoT devices or portable devices.CNN uses a regression function instead of a classifying method because it works better with output that changes over time. By finding outliers in the network, this model made the single CNN better and got a high accuracy of 97.96%.

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
Divisions:	School of Computing > Master of Science in Cyber Security
Depositing User:	Tamara Malone
Date Deposited:	05 Jan 2023 16:45
Last Modified:	07 Mar 2023 11:38
URI:	https://norma.ncirl.ie/id/eprint/6069

Actions (login required)

View Item