QoE-aware Optimization of Video Stream Downlink Scheduling Over LTE Networks Using RNNs and Genetic Algorithm

Long Term Evolution (LTE) is the initial version of fourth-generation (4G) networks which provides ubiquitous broadband access. LTE supports multimedia Quality of Service (QoS) traffic with high data transfer speed, fast communication connectivity, and high security. Multimedia traffic over LTE networks is one of the highest percentages of mobile traffic and it has been growing rapidly in recent years. Our approach focuses on the development of Quality of Experience (QoE) aware optimization downlink scheduling video traffic flow. QoE is the overall acceptability of a service or application, as perceived subjectively by end users. In this work we aim to maximise QoE of video traffic streaming over LTE networks. This work introduces a novel integration framework between genetic algorithm (GA) and random neural networks (RNN) applied to QoE-aware optimization of video stream downlink scheduling. The proposed framework has been applied and evaluated using an open source simulation tool for LTE networks (LTE-Sim). A comparison between our framework and state-of-the-art LTE downlink scheduling algorithms (FLS, EXP-rule, and LOG-rule) has been done under different network conditions. Simulation results have shown that our scheduler can achieve better performance in terms of QoE (∼10% increase), throughput and fairness.