An Investigation of knowledge-based software defined networking framework for automation of wireless network resource management

Wireless networks represent a significant portion of digital society, employing the packet-switched distributed architecture. Techniques such as deep learning-based traffic classification, reinforcement learning, mobile agents, and the identification of elephant and mice flows have been utilized to handle various network functions, such as traffic identification, routing strategies, and spectrum allocation. Software-defined Networks (SDN) have played a crucial role in addressing the challenges encountered in conventional network management. However, owing to the dynamic nature of networks and the high volume of network traffic, software-defined network architecture is insufficient to ensure optimal system performance in network environments. Hence, this research proposes a framework for integrating analytics and knowledge representation within a software-defined networking architecture, aiming to automatically optimize, diagnose, and troubleshoot wireless networks. A systematic literature review and qualitative comparative analysis were used to investigate knowledge-based Software-Defined Networking (SDN) architectural frameworks in the literature. The findings indicate the proposed framework outperformed other knowledge-based SDN frameworks in network functions and knowledge integration