Ripple20 vulnerabilities detection using featureless deep learning model

Hulayyil, Sara and Li, Shancang 2023. Ripple20 vulnerabilities detection using featureless deep learning model. Presented at: TrustCom 2023 : The 22nd IEEE International Conference on Trust, Security and Privacy in Computing and Communications, 1-3 November 2023.

Preview

PDF - Accepted Post-Print Version Available under License Creative Commons Attribution. Download (2MB) | Preview

Abstract

Inherent vulnerabilities create new security risks and challenges that leave Internet of Things (IoT) systems open to cyber attacks. Featureless deep learning shows great potential in vulnerabilities detection without relying on explicit feature engineering in the IoT. Featureless deep learning models provide a low-cost and memory time-series analysis of network traffic. This paper proposes a featureless deep learning procedure in a 1D CNN model to carry out Rippl20 detection. The experimental results demonstrate the effectiveness of proposed solution with it is beneficial for decreasing the time spent on feature engineering. Specifically, this proposed featureless model achieved $99\%$ of accuracy and a $F_1$ score as $0.9991$ with less time than traditional methods.

Item Type:	Conference or Workshop Item (Paper)
Status:	In Press
Schools:	Computer Science & Informatics
Date of First Compliant Deposit:	11 October 2023
Last Modified:	04 Nov 2023 02:30
URI:	https://orca.cardiff.ac.uk/id/eprint/162618

Actions (repository staff only)

Edit Item