Caching, transcoding, delivery and learning for advanced video streaming services

Abstract

Online video services are responsible for a large amount of global data traffic, which is why wireless video caching has become a major focus in order to deal with the increasing amount of data. While conventional caching methods have been successful in reducing file transmission latency and data traffic on the core network, there is still a lot of potential to improve wireless caching for streaming users’ satisfaction, especially as video streaming technology continues to improve. To improve caching, it is important to consider the distinct characteristics of a video file, video encoding, dynamic video streaming, and even 3D video services such as virtual reality (VR) and augmented reality (AR). This paper investigates the distinct video features that can be used to design video caching policies. Additionally, recent studies on wireless video caching and delivery are surveyed, with a focus on four key characteristics: (1) partitioning a video file, (2) encoding a video file into multiple quality versions, (3) recognizing that video file popularity is dynamic and heterogeneous, and (4) understanding that video recommendation affects a series of user requests.

Keywords

• Wireless caching
• Video streaming
• Content delivery network
• Adaptive bitrate streaming