An enhanced self-learning-based clustering scheme for real-time traffic data distribution in wireless networks.

Area of Science:

• Computer Science
• Network Engineering
• Data Mining

Background:

• Network traffic analysis is crucial for identifying wireless network performance abnormalities.
• Effective traffic classification is essential for distinguishing real-time from non-real-time data.
• Existing algorithms often leave network traffic classification unaddressed for effective analysis.

Purpose of the Study:

• To enhance network traffic classification accuracy and efficiency in wireless networks.
• To address the gap in existing algorithms for effective network traffic analysis.
• To improve real-time traffic data distribution using advanced clustering and classification.

Main Methods:

• Utilized data mining clustering and classification algorithms for traffic data categorization.
• Proposed an Enhanced Self-Learning-based Clustering Scheme (ESLCS).
• Employed an enhanced unsupervised algorithm and adaptive seeding approach within ESLCS.

Main Results:

• The ESLCS model demonstrated enhanced clustering accuracy and True Positive Rate (TPR).
• Achieved a substantial reduction in Classification Time (CT) and Communication Overhead (CO).
• Outperformed peer-existing routing techniques in test-bed evaluations.

Conclusions:

• The ESLCS effectively classifies real-time and non-real-time traffic data in wireless networks.
• The proposed scheme significantly improves classification accuracy while reducing processing time and overhead.
• ESLCS offers a promising solution for efficient network traffic analysis and performance enhancement.