Self-tuning of scheduling parameters for balancing the quality of experience among services in LTE

Abstract

Improving the Quality of Experience (QoE) offered to subscribers has become a major issue for mobile network operators. In this paper, a self-tuning algorithm for adjusting parameters in a multi-service packet scheduler of a radio base station based on network statistics is proposed to balance QoE across services in Long Term Evolution (LTE). The aim of the algorithm is to ensure that all users achieve the same average QoE regardless of the type of service. For this purpose, the proposed heuristic algorithm iteratively changes service priority parameters to re-prioritize services so that those services with the lowest QoE increase their priority. Unlike previous approaches, the proposed algorithm takes QoE (and not Quality of Service) into account. Method assessment is carried out in a dynamic system-level LTE simulator. Simulation results in a typical scenario show that tuning service priority parameters can significantly increase the QoE of the worst service without affecting the overall network QoE.