QoE-Aware Bitrate Selection in Cooperation With In-Network Caching for Information-Centric Networking

Abstract

Adaptive Bitrate (ABR) algorithms used in MPEG Dynamic Adaptive Streaming over HTTP (MPEG-DASH) can be applied to video streaming over Information-Centric Networks (ICNs). However, in-network cache, which is an inherent and important feature of ICN, might negatively affect the Quality of Experience (QoE) of streaming users due to misestimating download throughput when fetching from the in-network cache. A promising solution to this problem is to use a Server and Network Assisted DASH (SAND)-like approach: ICN routers in the network notify a user application of the available bandwidth. However, with a naive network-assisted approach, the user application cannot fully utilize the cached segments when high-quality video segments are accidentally stored in the router on the user side of the congestion. In this paper, we propose an intelligent QoE-aware ABR selection method that works in cooperation with in-network Caching functions, called QoE-ABC. It is suitable for video streaming over an ICN. In QoE-ABC, QoE-aware adaptation logic running on the user application selects a bitrate that matches the bandwidth of the bottleneck in the original server or of any intermediate router on the download path depending on the situation. Only when the user-perceived QoE is expected to be improved, the user application tries to aggressively download high-quality segments from the in-network cache. Simulation results show that QoE-ABC can obtain high-quality video segments from in-network caches compared with the conventional ABR representatives and achieve high-level QoE performance for users with various preferences.

Keywords

• ICN
• CCN
• NDN
• video streaming
• MPEG-DASH
• adaptive bitrate