Title: IoT security using deep learning algorithm: intrusion detection model using LSTM

Authors: Abitha V.K. Lija; R. Shobana; J. Caroline Misbha; S. Chandrakala

Addresses: Department of CSE, Meenakshi College of Engineering, Chennai, 78, India ' Department of CSE, S.A. Engineering College, Thiruverkadu Post, Chennai – 600077, India ' Department of CSE, Arunachala College of Engineering, Tamil Nadu, 629203, India ' Department of Computer Science and Engineering, Meenakshi College of Engineering, Chennai, 78, India

Abstract: Internet of things (IoT) and the integration of many gadgets is rapidly becoming a reality. IoT devices, particularly edge devices, are particularly vulnerable to cyberattacks as a result of the proliferation of device-to-device (D2D) connectivity Advanced network security measures are required to do real-time traffic analysis and to mitigate malicious traffic. These mechanisms must also be able to detect malicious traffic. We describe a game-changing approach to detect and classify new malware in record time. This will allow us to handle the difficulty that has been presented (zero-day malware). This article puts out the idea of a hybrid deep learning (DL) model for the detection of cyber attacks. Long short-term memory (LSTM) and gated recurrent unit are the foundations of the model that has been suggested (GRU). The results of the experiments are quite encouraging, revealing an accuracy rate of 94.50% for the identification of malware traffic.

Keywords: deep learning; gated recurrent units; internet of things; IoT; long short-term memory; LSTM; machine learning.

DOI: 10.1504/IJESDF.2025.143479

International Journal of Electronic Security and Digital Forensics, 2025 Vol.17 No.1/2, pp.283 - 293

Received: 17 Sep 2023
Accepted: 13 Nov 2023
Published online: 23 Dec 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article