Adaptive Assembly on Delay Fairness in Optical Burst Switched Networks

This paper focuses on burst assembly at edge node in optical burst switching networks, which can be implemented to provide the delay QoS guarantee for the whole network. Based on traffic prediction and delay fairness of different classes, a Prediction and delay QoS Burst Assembly (PQBA) mechanism is proposed to reduce the assembly delay and improve the delay fairness especially for high priority service at edge node. We compare the delay, burst size, burst utilization and length error rate of all service classes under different offered load. Simulation studies are carried out for PQBA and the results show that high priority service with relative small burst size achieves a low assembly delay and high burst utilization. Consequently, PQBA realizes the QoS-based delay fairness and service class differentiation at edge node.