Enhancing IoT security: A Creative Swagger Optimization algorithm for DDoS defence

Area of Science:

• Cybersecurity
• Artificial Intelligence
• Network Security

Background:

• Internet of Things (IoT) systems are vulnerable to security threats due to extensive data exchange.
• Distributed Denial of Service (DDoS) attacks compromise IoT server availability by overwhelming them with traffic.
• Existing DDoS detection methods require enhancement for improved accuracy and robustness in IoT environments.

Purpose of the Study:

• To propose a novel method for detecting and mitigating DDoS attacks in IoT networks.
• To enhance the accuracy and robustness of DDoS attack detection using optimized deep learning models.
• To introduce the Creative Swagger (CS) algorithm for optimizing Deep Convolutional Neural Network (DeepCNN) parameters.

Main Methods:

• Development of a Creative Swagger (CS) algorithm by integrating Swagger behavior with innovative concepts.
• Optimization of Deep Convolutional Neural Network (DeepCNN) parameters using the CS algorithm for improved DDoS detection.
• Implementation of a blacklist table for initial verification of network traffic, including IP address checks.
• Training and validation of the CS-optimized DeepCNN model on the UNSW-NB15 Dataset.

Main Results:

• The CS-optimized DeepCNN model achieved a detection accuracy of 97.07%.
• The model demonstrated high sensitivity (97.23%) and specificity (96.91%) at 80% training data.
• The proposed method showed significant robustness in detecting DDoS attacks on IoT platforms.

Conclusions:

• The CS-optimized DeepCNN presents an effective solution for detecting and mitigating DDoS attacks in IoT environments.
• The fusion of the CS algorithm with DeepCNN significantly enhances detection accuracy and model robustness.
• This research contributes a more secure and reliable approach to safeguarding IoT network communications.