Abstract:
With the increasing adoption of IoT technologies, securing IoT networks has
become increasingly critical. Traditional intrusion detection systems face
challenges in IoT environments due to resource constraints and the complexity
of these networks. This tpes of research develops, evaluates, and implements an
intrusion detection system incorporating deep learning algorithms, blockchain
technology, and a hybrid placement approach within a multi-agent framework.
The system comprises modules for response, analysis, data management, and
data collection. Testing is conducted using the NSL-KDD dataset from the
National Security Lab-Knowledge Discovery and Data Mining. The findings
indicate that deep learning approaches are viable for intrusion detection in IoT
networks.
