Sparse Spectra Graph Convolution Neural for Detecting IoT-Botnet Cyber Attacks in Wireless Sensor Networks.
Saved in:
| Title: | Sparse Spectra Graph Convolution Neural for Detecting IoT-Botnet Cyber Attacks in Wireless Sensor Networks. |
|---|---|
| Authors: | Deepa Priya, Balakrishnan Nair Sreedevi1 (AUTHOR) deepapriya@bitsathy.ac.in, Thangavel, Thiruvenkadam2 (AUTHOR) mailone.thiru@gmail.com, Jothilakshmi, Rajendran3 (AUTHOR) rjothilakshmi@gmail.com, Abdul Kader, Mohideen4 (AUTHOR) mohideenammu.m@gmail.com |
| Source: | IETE Journal of Research. Jul2025, Vol. 71 Issue 7, p2344-2352. 9p. |
| Subjects: | Botnets, Internet of things, Graph neural networks, Feature selection, Wireless sensor networks, Machine learning, Cyberterrorism |
| Abstract: | The widespread use of smart digital devices and Internet of Things (IoT) systems has led to the perception of being the target of network attacks. Botnets are buttoned-up vectors that let attackers take control of IoT systems and engage in malicious behavior. Many existing solutions are focussed on detecting established attack types, ignoring the dynamic and adaptable nature of IoT botnets, which can exploit vulnerabilities in real-time. In this manuscript, a Sparse Spectra Graph Convolution Neural Network for Detecting IoT-Botnet Cyber Attacks (SSGCNN-IoT-BAD-WSN) is proposed. Here, the input data is collected from the UNSW-NB15 dataset. Then the obtained data is fed to Sub-Aperture Keystone Transform Matched Filtering (SAKTMF) for removing noise and cleaning the data. Then, the Honey Badger Algorithm (HBA) is used to select the feature subset selection and subset size from the dataset. Building Ensemble techniques using Sparse Spectra Graph Convolution Neural Network (SSGCNN), Enhanced Attentive Dual Residual Generative Adversarial Network (EADRGAN), and Memory-Augmented Deep Unfolding Network (MADUN) for detecting IoT-Botnet cyber-attacks as Fuzzers, analysis, backdoors, DoS, exploits, generic, reconnaissance, shell code, and worms. Then, based on the voting classifier SSGCNN gives a higher result. The Crisscross Harris Hawks Optimization (CHHO) algorithm is proposed to improve the weight parameters of SSGCNN for effectively detecting IoT-Botnet cyber-attacks. The proposed approach is applied and analyzed utilizing performance metrics like accuracy, precision, and computational time. The proposed SSGCNN-IoT-BAD-WSN achieves 20.16%, 23.31%, and 24.56% higher accuracy respectively compared with existing techniques. [ABSTRACT FROM AUTHOR] |
| Copyright of IETE Journal of Research is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
|
Full text is not displayed to guests.
Login for full access.
|
|
| Abstract: | The widespread use of smart digital devices and Internet of Things (IoT) systems has led to the perception of being the target of network attacks. Botnets are buttoned-up vectors that let attackers take control of IoT systems and engage in malicious behavior. Many existing solutions are focussed on detecting established attack types, ignoring the dynamic and adaptable nature of IoT botnets, which can exploit vulnerabilities in real-time. In this manuscript, a Sparse Spectra Graph Convolution Neural Network for Detecting IoT-Botnet Cyber Attacks (SSGCNN-IoT-BAD-WSN) is proposed. Here, the input data is collected from the UNSW-NB15 dataset. Then the obtained data is fed to Sub-Aperture Keystone Transform Matched Filtering (SAKTMF) for removing noise and cleaning the data. Then, the Honey Badger Algorithm (HBA) is used to select the feature subset selection and subset size from the dataset. Building Ensemble techniques using Sparse Spectra Graph Convolution Neural Network (SSGCNN), Enhanced Attentive Dual Residual Generative Adversarial Network (EADRGAN), and Memory-Augmented Deep Unfolding Network (MADUN) for detecting IoT-Botnet cyber-attacks as Fuzzers, analysis, backdoors, DoS, exploits, generic, reconnaissance, shell code, and worms. Then, based on the voting classifier SSGCNN gives a higher result. The Crisscross Harris Hawks Optimization (CHHO) algorithm is proposed to improve the weight parameters of SSGCNN for effectively detecting IoT-Botnet cyber-attacks. The proposed approach is applied and analyzed utilizing performance metrics like accuracy, precision, and computational time. The proposed SSGCNN-IoT-BAD-WSN achieves 20.16%, 23.31%, and 24.56% higher accuracy respectively compared with existing techniques. [ABSTRACT FROM AUTHOR] |
|---|---|
| ISSN: | 03772063 |
| DOI: | 10.1080/03772063.2025.2485400 |
