Hit Ratio and Content Quality Tradeoff for Adaptive Bitrate Streaming in Edge Caching Systems

This paper addresses the tradeoff problem between hit ratio and content quality in edge caching systems for multiuser adaptive bitrate streaming (ABS) services. A dynamic policy for cache decision and quality level selection for each ABS content during every cache cycle is proposed. Achieving this policy is NP-complete. For this, the considered problem is transformed into a nested multidimensional 0/1 knapsack optimization problem which is then resolved by a cooperative transfer learning-accelerated genetic algorithm. Performance evaluation demonstrates an adaptation of the proposed algorithm on various video stream popularity models in terms of algorithmic convergence and cache balancing.

I. Introduction

In a modern mobile video streaming, heterogeneous user demands and preferences are supported by an adaptive bitrate streaming (ABS) service. The ABS service enables networks to dynamically adjust the quality level of content chunks (represented by video bitrate) to respond to the change of environmental conditions and resource availability [1]. To ensure efficient cooperation between network elements in the ABS systems, the International Organization for Standardization (ISO) publishes the ISO/IEC 23009-5:2017 standard [2] which defines a functional architecture, namecoded SAND. The SAND architecture enables network-assisted video streaming ability along with dynamic adaptive streaming over HTTP (DASH) protocol utilization. Here, content chunks are transcoded and/or cached on the delivery path by the network elements closer to users. These in-network services are referred to as the edge caching systems (ECSs).