Implications and Application of Artificial Intelligence and Machine Learning Concepts on Software Defined Network and Its Future Prospects.
Abstract
21st Century came with many discoveries and inventions that set the creation of modern world into motion. One of the noteworthy invention was the internet a realm of interconnected machines sharing information. Approximate of 4.39 billion internet users around the world which is around 58% of the worlds populous. With an insurgence of network users, there is an exponential surge in the devices connecting and accessing the network, interconnected intelligent IoT devices interacting over the internet, introduction to the 5G Mobile Wireless Networks, the problems faced by the network are not only increasing exponentially but also time constraints of process have to be reduced too. A solution proposed for these prolonged problems for last couple of decades is Software Defined Network (SDN), an effective platform to craft network for more agile, efficient and at an astonishing pace. Artificial Intelligence (AI) and one of its aspects Machine Learning (ML) has an overgrown demand in the industry due to its effectiveness in solving the problem on the machines end reducing than the human interaction. Utilizing the concepts presented by the AI and algorithms used for fabricating the learning aspect of machines the SDN could be made even more agile, robust and efficient and could configure itself as per requirement. Major problem faced with the SDN could be resolved using Machine Learning Algorithm like Load Balance and Flow Routing. SDN’s Security aspect could be made even more robust with an introduction to the AI-infused protection to the network. Future aspects are with combined forces of SDN and AI is quite prominent, as potential, it could be seen with AI-Agent. AI-Agent Works on several layers of SDN creating a synergy between the several aspects of SDN and enhancing the Quality of Experience (QoE). A model devised for QoE inclined cognitive network shows a prominent development for the combined fields of ML and SDN.
