An Improved Quality-of-Service Performance Using RED’s Active Queue Management Flow Control in Classifying Networks

Abstract

Traffic classification networks have various applications for data transmissions to ensure quality of service (QoS) for various classes of traffic at the routers. Multi-level random early detection (MRED) scheduling algorithm is used to manage resources at the routers guaranteeing QoS. However, the MRED queue mechanism is insensitive to traffic and difficult to set parameters, for the average queue is sensitive to high congestion level of multi-flow which is a major issue affecting the performance of the queue in the networks. This paper propose a new scheduling algorithm that manages congestion level by increasing the stability of parameters, using dynamic weighted traffic with redefining probability drop traffic in the MRED algorithm. The results present the performance algorithm while utilizing the reference algorithms, improving the bandwidth fairness and average throughput and reduce the average delay and packet drop.