مراقبة حركة الأحمال في الشبكات المعرفة بالبرمجيات والتحكم بحركتها باستخدام تقنيات التعلم الآلي

Abstract

 The monitoring and classification of traffic loads are among the most significant research areas due to the rapid growth of modern applications and networks. This process offers various benefits، including reducing network congestion، improving network management، and enhancing service quality. With the advancement of Software-Defined Networking (SDN)، which addresses the limitations of traditional networks by simplifying network management، enabling programmability، and providing comprehensive network visibility، SDN has become a promising platform for traffic classification and optimized routing using machine learning techniques.

In this study، an SDN environment was built to monitor and classify traffic loads using machine learning techniques. The Random Forest algorithm، a supervised learning model، was employed to classify network traffic based on application-level features. Traffic loads were categorized into "high" and "low" loads based on extracted flow features. Following classification، the RYU controller was utilized to determine the optimal path for each load. The results demonstrate the feasibility of accurate traffic classification and improved routing efficiency in SDN environments، contributing to enhanced network performance and adaptability to dynamic traffic changes.