Modeling and analysis of AF service with correlated source and channel dynamics

Modeling differentiated services mainly has been achieved in the wired network. However, most DiffServ models neither represent nor investigate time-varying wireless channel characteristics. In this paper, we focus on the modeling and analysis of the assured forwarding (AF) service that depends on the RIO queue management. Our goal is to examine how the correlation between source and channel affects the performance of the AF service. We quantify the performance measure of the AF service over wireless channel. Our analysis shows that joint dynamics of source and channel has a significant impact, and identifies the performance inversion, when the load (ρ) exceeds 1.