IoT Network Security Enhancement: Leveraging Deep Learning for Intrusion Detection and Addressing Data Imbalance

• Item type: SIG Technical Reports(1)
• 公開日: 2023-10-31
• タイトル: IoT Network Security Enhancement: Leveraging Deep Learning for Intrusion Detection and Addressing Data Imbalance
• タイトル / 言語: en
• タイトル: IoT Network Security Enhancement: Leveraging Deep Learning for Intrusion Detection and Addressing Data Imbalance
• 言語: eng
• キーワード / 主題Scheme: Other
• キーワード / 主題: ポスター
• 資源タイプ / 資源タイプ識別子: http://purl.org/coar/resource_type/c_18gh
• 資源タイプ: technical report
• 著者所属: Tokyo Institute of Technology
• 著者所属: Tokyo Institute of Technology
• 著者所属(英): en
• 著者所属(英): Tokyo Institute of Technology
• 著者所属(英): en
• 著者所属(英): Tokyo Institute of Technology
• 著者名: Qingyu, Zeng; Yuko, Hara-Azumi
• 著者名(英): Qingyu, Zeng; Yuko, Hara-Azumi
• 論文抄録 / 内容記述タイプ: Other
• 論文抄録 / 内容記述: In the era of the Internet of Things (IoT), ensuring robust network security is important. This work studies an innovative approach to enhance IoT network security by deep learning-based intrusion detection while addressing data imbalance challenges. We comprehensively integrate advanced deep learning models with tailored data imbalance mitigation strategies. We also employ novel synthetic data generation techniques to bolster the model's resilience against real-world adversarial attacks. Extensive evaluations on established datasets, including UNSW-NB15, NSL-KDD, and IoT23, demonstrate significant improvements in intrusion detection accuracy, especially in detecting minority-class threats. Additionally, our research extends to deploying models on resource-constrained IoT devices, such as Raspberry Pi and Nvidia Jetson, improving real-time intrusion detection efficiency in these small-scale devices.
• 論文抄録(英) / 内容記述タイプ: Other
• 論文抄録(英) / 内容記述: In the era of the Internet of Things (IoT), ensuring robust network security is important. This work studies an innovative approach to enhance IoT network security by deep learning-based intrusion detection while addressing data imbalance challenges. We comprehensively integrate advanced deep learning models with tailored data imbalance mitigation strategies. We also employ novel synthetic data generation techniques to bolster the model's resilience against real-world adversarial attacks. Extensive evaluations on established datasets, including UNSW-NB15, NSL-KDD, and IoT23, demonstrate significant improvements in intrusion detection accuracy, especially in detecting minority-class threats. Additionally, our research extends to deploying models on resource-constrained IoT devices, such as Raspberry Pi and Nvidia Jetson, improving real-time intrusion detection efficiency in these small-scale devices.
• 書誌レコードID / 収録物識別子タイプ: NCID
• 書誌レコードID / 収録物識別子: AA11451459
• 書誌情報: 研究報告システムとLSIの設計技術（SLDM）; 巻 2023-SLDM-203, 号 14, p. 1-3, 発行日 2023-10-31
• ISSN / 収録物識別子タイプ: ISSN
• ISSN / 収録物識別子: 2188-8639
• Notice: SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc.
• 出版者 / 言語: ja
• 出版者: 情報処理学会