Application-driven cross layer optimization for wireless networks using MOS-based utility functions

This paper discusses a Quality of Experience (QoE) driven cross-layer optimization framework for efficient network resource allocation in wireless networks. The proposed scheme jointly optimizes the application layer and the lower layers of the wireless protocol stack with the aim of improving the user´s QoE. The Mean Opinion Score (MOS) is used as a common metric for user-perceived quality in the optimization scheme. Three different QoE-based optimization schemes are compared to a throughput maximization scheme and a non-optimized system. We perform simulations using a software implementation of a developed HSDPA system. Results show that the MOS-based approaches lead to significant improvements of user perceived quality.