A Data Enhancement Algorithm for DDoS Attacks Using IoT.

Area of Science:

• Cybersecurity
• Machine Learning
• Internet of Things (IoT)

Background:

• The proliferation of Internet of Things (IoT) devices has led to an increase in botnet-driven cyberattacks, particularly distributed denial-of-service (DDoS) attacks.
• Traditional intrusion detection systems struggle with low accuracy due to the small percentage of attack packets in IoT environments, a problem exacerbated by data imbalance.

Purpose of the Study:

• To propose and evaluate a novel oversampling algorithm, KG-SMOTE, designed to mitigate data imbalance in IoT-based DDoS attack detection.
• To enhance the accuracy and effectiveness of intrusion detection systems in IoT environments.

Main Methods:

• Development of the KG-SMOTE algorithm, which utilizes Gaussian distribution for synthetic sample insertion and K-means clustering to balance minority class samples.
• Application of the KG-SMOTE algorithm to generate a balanced dataset for IoT intrusion detection.

Main Results:

• The KG-SMOTE algorithm effectively increased the density and amount of minority class samples.
• Experiments demonstrated that the balanced dataset significantly improved intrusion detection accuracy across all categories.
• The proposed method successfully addressed the critical data imbalance issue in IoT security.

Conclusions:

• KG-SMOTE provides a robust solution for data imbalance in IoT cybersecurity, particularly for detecting DDoS attacks.
• The algorithm's ability to generate synthetic samples and increase minority class data density leads to superior intrusion detection performance.