Using Debt Mechanism to Achieve Proportional Delay and Loss Differentiation in a Wireless Network with a Multi-state Channel

Abstract

Many algorithms for providing proportional delay differentiation and proportional loss differentiation have been proposed under the environment of wired networks. However, these algorithms suffer from difficulties when meet some distinct characteristics, such as location dependent, high error-rate or time varying channel capacity, which exist in wireless networks. This paper proposes a novel algorithm, Debt-aware Proportional Delay and Loss differentiation (DPDL), including a debt-aware scheduler and a debt-aware dropper. It can achieve proportional delay and loss by considering the channel state and debt information. From simulation results, DPDL actually achieves proportional delay differentiation, proportional loss differentiation, lower queuing delay and loss and higher throughput, compared with the other methods in wireless environment.