360-degree video stream has the problems of unnecessary data transmission and high quality QoE requirement in the AR/VR network field. The project aims to predict immersive video user behavior information and select appropriate video content based on network delay estimation as well as adjusting video bit rate.
- Combining the network delay, we adjust the bit rate and the user’s field of view adaptively;
- A user behavior simulation tool based on Markov model and Beta distribution is implemented;
- A hierachical cache system design for prefetch tiles in high-performace storage and cache the videos in cost-effective storage.