A queue scheduling approach to quality of service support in diff-serv networks using fuzzy logic

Abstract

In this paper, the fuzzy logic-based decision algorithm is proposed for queue scheduling. This algorithm is employed to determine the optimal values for achieving fairness control for each queue-type serviced. It takes into account the dynamic nature of the internet traffic as regards to its time-varying packet arrival process that affects the network states and performance. In order that the fairness control is achieved some predetermined threshold values are set so as to enforce some level of control in the router for traffics of different quality of service requirements. The threshold values are set for queue priority, drop priority and the number of waiting packets in queue (queue limit). The fuzzy scheduling mechanism is proposed for implementation at the output port of the internet router. Three (3) queues are assumed, low priority queue, medium priority queue and high priority queue, which takes the form q0, q1 and q2, respectively. All the queues are subject to priority, that is, packets are placed in the queues according to their priority class by the classifier. The choice of prioritizing packets influences how queues are served but unlike the priority queue (PQ) scheduling mechanism, which only schedules packets based on priority, the fuzzy scheduler utilizes queue priority, drop priority and the queue limit as parameters for queue scheduling. Through simulation experiments, it is found that the effectiveness of the proposed method proves to offer better improvement for network performance in terms of traffic dropped and throughput.