Abstract
A stream of video is composed of frames, which can be categorized into frame types of different importance. In this research, we aim to improve the Quality of Experience (QoE) perceived by user via improving the rate of successful transmission for important video frames. Our method is designed for 4G LTE-A networks. Different modulation and coding schemes (MCSs) are assigned to frames based on the degree of the importance. The downlink radio resource scheduling problem is formulated with the considerations of frame type and reference relation between frames of the same group of picture (GOP). The formulation aims to maximize the average expected weighted successful transmission size, while considering the delay constraint for video frames. However, this formulation is a mixed integer nonlinear problem (MINLP), which is proved to be NP-hard. Therefore, we propose a heuristic algorithm to determine the transmission order and the MCS level for each video frame. In simulation, we measure Peak Signal Noise Ratio (PSNR) to evaluate the performance of our proposed method comparing to the methods proposed in existing literature. The evaluation shows that the PSNR value of the proposed method is higher than those of the conventional adaptive method 2.35 dB and 5.27 dB on average in the scenario with the same resources and in the sufficient bandwidth condition respectively.