Research on intelligent ship resilient network architecture based on SDN

With the extensive adoption of information and communication technology (ICT) in the maritime field, intelligent ships are increasingly dependent on system integration, control, and data collection from devices. Real-time data transmission is essential for ensuring stable ship system operations. However, communication link failures frequently become key factors impacting data transmission. To this end, we propose an SDN-based intelligent ship network architecture, SDN-Intelligent Ship Network Architecture (SDISN), to simplify network management and enable centralized control of intelligent ships. On this basis, we design a link failure recovery model tailored for different maritime communication services to address the issue of sudden communication link failures. The model begins by collecting the status of the intelligent ship network and pre-defining backup flow rules for different maritime communication service flows. Considering the service flow characteristics, the optimization aims to minimize transmission delay and maximize switch TCAM utilization for life-safety communication flows and ship operational communication flows, respectively. For life-safety communication flows, we introduce a heuristic algorithm that progressively relaxes constraints. Meanwhile, we preload backup flow rules into switches. For ship operational communication flows, we apply a two-stage optimization algorithm, storing the relevant backup flow rules in the controller. Additionally, we propose a backup storage strategy for commercial communication flows based on dynamically adjusting the memory load of switches. Compared to existing approaches, the SDISN satisfies the need for real-time data transmission in intelligent ships while balancing resource consumption and fault response time in its link failure recovery mechanism. Lastly, experiments conduct on a testbed in a real network environment further validate the model's efficacy and efficiency.