报告详情
Adversarial Robust Swin Transformer-Based Learning for IoT Malware Detection
编号：188
访问权限：仅限参会人
更新：2026-11-10 17:05:35 浏览：9次
Online
报告开始：2026年11月21日 11:00（Asia/Kolkata）
报告时间：15min
所在会场：[S2] Internet of Things and Intelligent Sensing Systems [S2-1] Internet of Things and Intelligent Sensing Systems
暂无文件
摘要
The rapid expansion of the Internet of Things (IoT) has increased the surface area for a wide range of dynamic cyberattacks, necessitating the use of robust security solutions. Current detection methods struggle with sophisticated, complex malware variants, and many Deep Learning (DL) models are too computationally intensive for IoT devices with limited resources. In recent years, researchers have developed a number of malware detection approaches, although precise detection remains difficult. As a result, the goal of this research is to provide a precise and effective Swin Transformer-based framework capable of detecting complicated threats on resource-constrained devices. Initially, data were collected from the BoT-IoT dataset, and adversarial samples were generated using the Projected Gradient Descent (PGD) method, which produces realistic adversarial samples. The features are then extracted using a Lightweight Dilated Convolutional Network with Strip-Style Pooling Attention (LDCN-SSPA), which allows for efficient processing of 1D traffic data. Finally, malware in IoT is detected using a Hybrid Swin Transformer-based BiLSTM with Dynamic Spatial Attention Network (HST-BiLSTM-DSAN), which recognizes localized structures and sequential dependencies to ensure reliable malware recognition in resource-constrained IoT devices. The experimental results show that the proposed model exceeds the state-of-the-art comparison in binary classification, with the maximum accuracy of 99.67%, precision of 99.23%, recall of 99.54%, and F1-score of 99.42%.
关键词
Internet of Things, Malware detection, Network Traffic, Dilated Convolution, Strip-Style Pooling Attention, Adversarial Robustness, Network Traffic Analysis
报告人
Surbhi Prakash
Phd Scholar Indira Gandhi Delhi Technical University Women稿件作者
全部评论
重要日期
-
会议日期
11月20日
2026
至11月22日
2026
-
10月31日 2026
初稿截稿日期
-
11月07日 2026
提前注册日期
-
11月22日 2026
注册截止日期
主办单位
United Societies of Science
承办单位
Pondicherry University
协办单位
IEEE Section
IEEE Madras Section
IEEE Madras Section
联系方式
- ICon INDIA 2026
- ic******@usssociety.org
发表评论