Advances to IoT security using a GRU-CNN deep learning model trained on SUCMO algorithm.

Area of Science:

• Cybersecurity
• Artificial Intelligence
• Network Security

Background:

• The proliferation of Internet of Things (IoT) devices has introduced significant security vulnerabilities.
• Emerging threats include Denial of Service (DoS) attacks and Botnets, compromising IoT environments.
• Effective detection mechanisms are crucial for maintaining IoT integrity and user trust.

Purpose of the Study:

• To propose a novel hybrid deep learning model for classifying IoT security threats.
• To enhance the accuracy and efficiency of detecting sophisticated network attacks in IoT ecosystems.
• To address the limitations of existing methods in analyzing complex network data.

Main Methods:

• A hybrid deep learning architecture integrating Convolutional Neural Network (CNN) for spatial feature extraction and Gated Recurrent Units (GRUs) for temporal dependency analysis.
• Implementation of the self-upgraded Cat and Mouse Optimization (SUCMO) algorithm for hyperparameter tuning of the deep learning model.
• Validation of the proposed model using two benchmark datasets: UNSW-NB15 and BoT-IoT.

Main Results:

• The hybrid CNN-GRU model demonstrated superior performance in classifying IoT security threats compared to traditional methods.
• The SUCMO algorithm effectively optimized the deep learning model's hyperparameters, leading to improved classification accuracy.
• Experimental results on both UNSW-NB15 and BoT-IoT datasets confirmed the model's effectiveness against state-of-the-art approaches.

Conclusions:

• The proposed hybrid deep learning model offers a robust solution for detecting and classifying IoT security threats.
• The integration of CNN and GRUs, optimized by SUCMO, provides a powerful tool for securing the expanding IoT landscape.
• This research contributes a significant advancement in the field of IoT cybersecurity, enhancing threat detection capabilities.