Internet of Things (IoT) Intrusion Detection System (IDS) for Home Networks

The surge in remote work, accelerated by the COVID-19 pandemic, alongside the rapid expansion of smart home Internet of Things (IoT) technologies, has transformed home networks into critical infrastructures for both personal and professional use. However, the integration of IoT devices presents significant security challenges due to inherent vulnerabilities such as weak authentication, insecure interfaces, and insufficient firmware updates. This research addresses these challenges by developing an advanced Intrusion Detection System (IDS) specifically tailored for IoT environments in home networks. This research introduces a methodology for identifying attack patterns in home networks by analyzing spectrogram images of IoT telemetry and network flow data using pre-trained CNN models. It evaluates the effectiveness of models like AlexNet, DenseNet, VGG16, EfficientNet, ResNet, and ConvNeXt in threat detection. Results show that spectrogram techniques with pre-trained models, particularly AlexNet, improve IDS capabilities, highlighting the need for customized IDS solutions for home IoT environments and the potential for further research in hybrid models and adaptive learning.

Author

• Deng, Xiaohu

Language

• English

Keyword

• Cybersecurity Analytics
• Cybersecurity
• Machine Learning
• Intrusion Detection System
• Convolutional Neural Network
• Internet of Things

Date created

• 2024

Type of Work

• Dissertation

License

• All rights reserved

Rights statement

• In Copyright

GW Unit

• Cybersecurity Analytics

Degree

• D.Engr.

Advisor

• Islam, Muhammad

Committee Member(s)

• Fossaceca, John
• Sarkani, Shahryar

Persistent URL

•  https://scholarspace.library.gwu.edu/etd/vm40xs415