QoE-based Cross-Layer Optimization for video delivery in Long Term Evolution mobile networks

In this paper, we apply a Quality of Experience (QoE) based Cross-Layer Optimization (CLO) framework for efficiently allocating the network resources for video delivery in LTE mobile networks. Conventionally, the objective function aims to maximize the average user-perceived quality of all users by jointly optimizing the application layer and the lower layers of the radio communication system. We extend the conventional objective function by a new constraint that allows the network operator to set its policy to provide the video service with a target average video quality. The Mean Opinion Score (MOS) is used to indicate the user satisfaction of the video service. A comparison between our approach, an Application-Layer Rate Adaptation (ALRA) and a non-optimized LTE system is done by performing simulations using an OPNET LTE simulator. Results show that our QoE-based approaches lead to significant improvements of user-perceived quality, and additionally to significant bandwidth savings in the case of target mean MOS approach.