QoS measures and implementations based on various models for real-time communications

Quality of service (QoS) is a set of service requirements to be met by the network while transporting a data flow in real-time communication systems. Many real-time sources generate traffic with variable rates. We studied a class of soft real-time applications whose bandwidth requirements are random variables. We defined two models for soft real-time communications: the dropping model and the keeping model. We measure the QoS of the whole system, not individual applications. We defined three measures for the system´s performance: system feasibility, packet transferability, and weighted packet transferability. Different scheduling algorithms are studied. Some related issues were also studied, such as queue size and gain functions. Finally we implemented a simulator and some typical simulation results and conclusions are presented