QoE-Driven Delay Announcement for Cloud Mobile Media

As a useful tool for improving the user's quality of experience (QoE), delay announcement has received substantial attention recently. However, how to make a simple and efficient delay announcement in the cloud mobile media environment is still an open and challenging problem. Unfortunately, traditional convex and stochastic optimization-based methods cannot address this issue due to the subjective user response with respect to the announced delay. To resolve this problem, this paper analytically studies the characteristics of delay announcement by analyzing the components of the user response and designs a QoE-driven delay announcement scheme by establishing an objective user response function. On the methodology end, the user response associated with the announced delay is approximated in the framework of fluid model, where the interaction between the system performance and delay announcement is well described by a series of mathematical functions. On the technology end, this paper develops a novel state-dependent announcement scheme that is more reliable than the other competing ones and can improve the user's QoE dramatically. Extensive simulation results validate the efficiency of the proposed delay announcement scheme.