A QoE-Based Measurement for DiffServ Multicasting Networks

Multimedia services are typically much more sensitive to throughput, delay, and packet loss than traditional services. These parameters give technical performances of the network but they do not reveal QoS perceived by users. Hence, how to capture users' overall perceptions when they are using network services is an important issue. In this paper, we simulate a streaming video environment in bandwidth broker (BB), and provide the data of measuring the quality of experience (QoE) for genetic algorithm (GA). After GA acquires these data, GA can use the proposed fitness function to measure the multicast group member's QoE, and decide whether to adjust a DiffServ-aware multicast tree (DAMT) according to the measured results. The simulation results have proved that the DAMTs adjusted by GA can satisfy the QoE of multicast group member.