Optimal Flow Rate Control for SDN-Based Naval Systems

It is essential for a naval ship system (NSS) to transmit important information with very low delay for successful network-centric warfare under given time constraints. From this perspective, quality of service (QoS) control and low delay transmission are the most important components to improve efficient network utilization in NSSs. In general networks, conventional network algorithms have been optimized to manage network devices such as gateway, switch, bridge, hub, and router in a distributed manner for ease of plug-in and play operation, but which may not be optimal for the NSS. This is because, unlike the general network environment, the NSS is equipped with various nodes, which need to be operated through a centralized hierarchical network structure in accordance with their priorities. Thereby, it is difficult to apply new network algorithms to the NSS due to the closed network environment. To overcome this limitation, we propose a novel NSS employing software defined network (SDN) technology. A novel algorithm called real-time transmission via flow rate control (RTF) is proposed for the SDN-based NSS to optimize the specific environment of NSS. In the proposed algorithm, the optimal flow rates of nodes and network devices are obtained via dual decomposition in conjunction with requirements of delay and QoS. The proposed algorithm obtains the optimal solution iteratively while measuring the performance with traffic models based on real data measured in NSS. Through simulation using Floodlight, we demonstrate that RTF significantly contributes to improving throughput in real-time transmission compared with conventional algorithms. The proposed algorithm improves the average throughput by up to 11.8% while guaranteeing real-time transmission.