科研速览 · Science Skim继续刷下去 · Keep skimming →
◆ IEEE Internet of Things Journal2025-10-03· Computer science

100 Gb/s per Channel Dynamic Adaptive Granularity-Aware Onboard Optical Switching

Zhao Li, Hui Yang, Qiuyan Yao, Yun Teng, Zixiao Wang, Jie Zhang

原始摘要(英文原文)· Original abstract
As satellite network rapidly develops, traditional on-board switching technologies can no longer meet the demands of high-reliability, low-latency, and multi-granularity satellite networks. On-board electronic switching systems encounter the ’electro-optical conversion bottleneck’, limiting their ability to efficiently handle high-bandwidth traffic. The static wavelength routing optical switching architecture struggles to handle multi-granularity traffic requirements between satellites due to its slow switching speed and lack of flexibility, resulting in congestion and packet loss in large-scale constellations. To address these issues, we designed an on-board optical switching prototype based on Ultra-fast Optical Packet Switching (UOPS) technology. This prototype supports 100Gbps transmission per channel and achieves sub-microsecond switching latency. Meanwhile, we found that in timeslot-based optical switching, traditional fixed timeslot configurations face challenges in achieving a reasonable trade-off: longer timeslots can lead to high packet loss rates, while shorter timeslots impose throughput limitations. Therefore, we propose a Dynamic Adaptive Granularity-Aware (DAGA) scheduling algorithm. By precisely sensing the time derivative of queue occupancy to capture the traffic granularity characteristics in dynamic satellite networks, enabling flexible timeslot allocation adjustments. Experimental results demonstrate that the DAGA scheduling algorithm significantly reduces packet loss rates and improves throughput, particularly under high-load and dynamic traffic conditions.
读原文 · Read the paper ↗

AI 追问PRO

登录后使用 AI 追问

讨论区

登录后参与讨论

相关论文 · Related

100 Gb/s per Channel Dynamic Adaptive Granularity-Aware Onboard Optical Switching — 科研速览 Science Skim