Hybrid HTTP Adaptive Augmented Reality Streaming System over Wireless Networks

Abstract

This dissertation presents a hybrid HTTP adaptive augmented reality streaming system over wireless network. First, we present a cooperative HTTP adaptive streaming system to maximize overall video streaming service quality of all the clients over SDN-enabled Wi-Fi APs. To archive the goal, the proposed system employs multipath technology to overcome the limitation of a Wi-Fi AP, and SDN technology to effectively harmonize multiple Wi-Fi APs. During the streaming services, SDN controller periodically monitors the traffic information collected from the APs, and then determines video bitrate and segment transmission schedule based on this information. Second, we present a HTTP adaptive augmented reality (AR) streaming system over wireless networks. In the proposed system, we employ a progressive mesh technique to enhance the scalability of AR streaming services, and adopt the metafile structure to support various attributes of progressive meshes necessary for rendering, displaying, and scheduling according to wireless network conditions. Based on this metafile, the proposed system requests the next chunk among AR contents to improve human visual perceptual quality considering the displaying scale values of AR contents on the screen. Thirdly, based on the firstly and secondary studies, we propose a mobile edge computing (MEC)-assisted hybrid HTTP adaptive animating AR streaming system over 5G network. In the proposed system, the mobile AR device determines the rendering priority according to the position and movement of the animating AR contents. The 3D meshes of high priority AR content are directly delivered to the mobile AR device and rendered. Otherwise, 3D meshes of low priority AR contents are rendered and converted to image/video at the MEC and then streamed to the mobile AR device. Finally, the mobile AR device combines the 3D meshes and the image/video in real-time.