计算机科学
带宽(计算)
带宽分配
计算机网络
电子工程
光通信
光开关
集成光学
无源光网络
光突发交换
传播延迟
实时计算
动态带宽分配
频率分配
分组交换
时分复用
调度(生产过程)
电信网络
光交叉连接
钥匙(锁)
抖动
光传送网
光学性能监测
信道分配方案
光学滤波器
光子学
作者
Tiziana Fiori,Francesco Giacinto Lavacca,Vincenzo Eramo
摘要
Passive optical networks are emerging as a promising solution for industrial communication systems, thanks to their high bandwidth, cost efficiency, and inherent reliability. However, the stringent latency and jitter requirements of industrial applications demand deterministic scheduling mechanisms capable of addressing real-time constraints. Recent research has proposed advanced scheduling strategies for deterministic PONs, often based on current optical access standards that include ranging and delay equalization procedures, but these approaches assume symmetric network conditions, do not mention or consider equalization, and ignore propagation-delay asymmetries, which are common in realistic industrial scenarios. For the first time, to our knowledge, this work explicitly analyzes delay equalization, here referred to as EqD-DetBA (Equalization of Propagation Delays with Deterministic Bandwidth Allocation), in deterministic PONs under asymmetric conditions. We then propose a Propagation Delay Aware Deterministic Bandwidth Allocation (PDADetBA) framework that explicitly accounts for such asymmetries during the scheduling process, which provides improved latency performance, zero jitter, and greater transmission manageability compared to relying solely on delay equalization. The proposed method is formulated as a mixed integer linear programming (MILP) problem to achieve optimal frame alignment and prevent collisions. A heuristic based on max-plus algebra and heap modeling ensures scalability and real-time applicability. Numerical results confirm that PDADetBA reduces latency, avoids collisions, and enhances transmission manageability under asymmetric conditions while remaining computationally efficient for large-scale deployments.
科研通智能强力驱动
Strongly Powered by AbleSci AI