Evaluation of Bandwidth-Aware Scheduling Method for Division-Based Broadcasting with Multiple Video Qualities

In recent years, the traffic generated by the transmission and reception of data such as audio and video has increased rapidly all over the world, which requires video delivery systems that adapt to the network environment. In a video delivery system, broadcasting reduces the server load and overall network traffic by delivering the same video data to many clients simultaneously. On the other hand, clients must wait between a request for video data and the start of playback. Therefore, a number of scheduling methods have been proposed to reduce waiting time by considering the available bandwidth and number of channels required for division-based broadcasting. Conventional division-based scheduling methods prepare a delivery schedule for a single playback rate of video data. However, they do not consider cases where a server delivers multiple types of videos with different playback rates and clients change the video quality based on the network environment. In our previous method, waiting time was reduced by delivering data on the differences among three levels of quality (low, medium, and high) on different channels. In this paper, we propose and evaluate a scheduling method for different data sizes and bandwidths while considering only two quality levels. The evaluation results confirmed that the proposed method reduces waiting time by approximately 81.0% on average compared to simply delivering all three types of video data (high quality, medium quality, and low quality).