Towards an SDN-based Reconfigurable Edge Architecture for Railway Environment

This paper presents a Software-Defined Networking (SDN)-based reconfigurable edge architecture designed to optimize the allocation of services across multiple edge servers in a railway environment. The proposed system, named the ‘Intelligent Traffic Router,” dynamically assigns primary, secondary, and tertiary edge servers based on real-time load and capacity, ensuring optimal resource utilization and maintaining Quality of Service (QoS) for critical services such as autonomous and teleoperated driving. A novel mathematical model and optimization algorithm are introduced to formalize the service allocation process, considering factors such as service priority, train mobility, and server capacity. The SDN controller continuously monitors server load and capacity, reallocating resources as needed to prevent overload and maintain service continuity. Performance evaluation using Mininet-WiFi and the Open Network Operating System (ONOS) SDN controller demonstrates the system’s effectiveness in dynamically managing server resources, handling edge server disconnections, and prioritizing high-priority services under heavy load conditions. Notably, the system successfully reroutes traffic to available edge servers when CPU usage exceeds 85%, ensuring uninterrupted service. The results highlight the system’s ability to enhance network stability, reliability, and efficiency, making it a viable solution for future railway transport systems.