Performance improvement of adaptive active queue management based on the variation of queue length

AQM(Active Queue Management) starts dropping packets earlier to notify traffic sources about the incipient stages of congestion. The AQM improves fairness between response flow(like TCP) and non-response flow(like UDP), it can provide high throughput and link efficiency. In this paper, we suggest an algorithm that QVARED(Queue Variation Adaptive RED) is based on the variation of a queue per hour, it can respond to bursty traffic more actively. It is possible more smoothness of average queue length by the method that considered the variation of current queue compared to RED and ARED. This means higher adaptability based on a new congestion collapse. Comparing the drop rate of QVARED to those of RED and ARED, and it was decreased by 80%, 40% each. This result in shorter end-to-end delays by the lower number of packet retransmits. QVARED reduces a bias of exceeding 18% higher than the drop-tail, therefore packets are transmitted stably with respect to bursty traffic.