Anomaly Classification in Internet of Things Using Temporal Convolutional Network with Spectral Unit based Attention Mechanism

Read full text Summarise with VIRAT
Authors: WT Valavan, Nalini Joseph
Cited by: 0
References: 27

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

A fast growth of Internet of Things (IoT) devices has considerably increased their susceptibility to a wide range of cyber-attacks.Thus, highlighted the requirement for a vigorous Intrusion Detection System (IDS) to ensure the security of the IoT networks.However, existing IoT anomaly detection approaches demonstrate poor handling of temporal patterns, weakening the model of sequential dependencies in device behaviour.Thus, this research considers a significant approach called Spectral Mix 1D layer-based Temporal Convolutional Network (STCN) for the classification of various intrusions in IoT.The proposed TCN encoder with dilated causal convolutions is specifically designed to capture long-term temporal dependencies without vanishing gradients.By performing in the frequency domain (Spectral Layer), the TCN determines the retelling patterns, periodic anomalies, and hidden correlations that time-domain-only models often miss.This study considered four different datasets, namely IoTID20, Bot-IOT, UNSW-NB15, and NFToN-IoT, to estimate a significance of proposed approach.The experimental discoveries establish that proposed STCN method achieves an optimal accuracy of 99.88% on the Bot-IoT dataset compared with the existing method of Swin transformer-based Long Short-Term Memory (LSTM).

Fields of Study & Subjects

Network Security and Intrusion Detection Network Packet Processing and Optimization Internet Traffic Analysis and Secure E-voting

Publication details

Journal / source
International journal of intelligent engineering and systems
Volume
19
Issue
4
Pages
501–515
Publication date
23 March 2026
Language
EN