SSHS: SDN seamless handover system among LAN access points

In recent years, artificial intelligence techniques, such as software-defined networks (SDNs), machine learning classification (ML classification), and mobility models (MMs), have become vital in developing networks. Furthermore, communication methodologies, such as handover, directly affect network performance. In this paper, we propose a new system named SSHS, SDN Seamless Handover System, that combines SDN with an ML classifier to administer the network connection of mobile nodes. Through the SSHS system, the SDN will centralize the control to enable comprehensive management over the network, coupled with a decision tree (DT) classifier in the RYU controller to bring intelligence to the SDN application by enabling data analysis and prediction among mobile nodes generated by the RSSGM model. We present the SSHS model's effectiveness in providing a seamless communication handover among multiple access points (APs). The results of this study revealed that the SSHS provided a seamless handover among APs by improving the throughput by 26%, and decreasing the delay of arriving packets by 73% to standard SDN handover system.