Huber-AQMRD algorithm:Performance improvement against network congestion
With the vigorous development of the information age,information transmission has permeated every aspect of people′ daily lives and commercial activities and become an indispensable part.However,this massive growth in information transmission has also brought about a series of problems,of which the network congestion has become increasingly common.In addressing the issue of network congestion,the active queue management(AQM)algorithms play a crucial role,including the algorithm of random early detection(RED)and the algorithm of adaptive mean queue size and its rate of change:queue management with random dropping(AQMRD).However,although these algorithms have worked to some extent,there is still room for improvement in throughput and quality of service.In view of the shortcomings of the existing algorithms,this paper proposes an improved algorithm based on AQMRD,namely the Huber-AQMRD algorithm.This algorithm assesses the differences between queue size and expected value more accurately by introducing the loss function ″Huber″,so as to optimize the design of packet dropping function.It is verified by ns3 simulation experiments that the Huber-AQMRD algorithm improves network throughput and the quality of service while reducing packet loss rate,so it holds significant importance in eliminating the network congestion under the background of massive growth of information transmission and enhancing the network performance and user experience.
network congestionAQMHuber-AQMRDns3packet loss ratethroughput
万方数据
维普
