QoE Driven Resource Allocation in Next Generation Wireless Networks

Abstract

With the proliferation of mobile video services in next-generation wireless networks, network optimization is progressing in the direction of improving QoE. This phenomenon is due to the fact that traditional QoS-oriented network optimization for video services fails to effectively configure network resources and enhance user satisfaction. Recently, QoE-driven resource allocation methods in different scenarios have been corroborated to outperform traditional QoS-oriented methods. It has been realized that reliable QoE models and effective optimization schemes are crucial to these methods. In this article, we consider a learning based QoE model from real-world data and apply it to network resource optimization. It is worth mentioning that human perception limits of video viewing are studied in the process of QoE modeling. Based on this model, the upper-layer optimization method suitable for the general wireless communication scenario and the lower-layer optimization method with the caching framework as the kernel are investigated respectively. The goal of this article is to provide an overview of the QoE-driven resource allocation architecture and to have a glimpse of the methodology employed in QoE modeling and network optimization.