Estratégia de adaptação de fluxo de vídeo baseada em Fatores de QoE

Abstract

Due to increasing demand of HTTP Adaptive Streaming (HAS) in the last years, se-veral studies have been performed to develop multimedia content distribution techniques to the Internet, aiming to improve the users Quality of Experience (QoE). Techniques such as algorithms that automatically adjust the bit rates to network conditions during a video session and algorithms that assist in the preparation of multimedia content for access by customers. Although there are a large number of studies in this area, many have not explored network resource sharing scenarios between multiple clients and little have investigated issues relating to preparation of the content multimedia. Making it necessary to study more about this complex scenario for the future to ensure a fairer distribution of network resources, furthermore users can experience similar qualities. In addition, pre-paration of the content should be viewed as an important process such as the adaptive algorithms for better distribution of the content and to improve the QoE. This work pre-sents a video stream adaptation strategy that takes into account factors related to QoE, as the playback interruption frequency. The goal is to minimize the amount of video playback interruption, thereby improving its continuity. The experiments are performed in different network scenarios for one client and in extreme resource scarcity scenarios on a share across multiple clients. Furthermore, we experienced versioning videos with different granularity and FPS (frames per second). In the results, the used metrics are average video bit rate, probability of interruptions, instability and fairness. In addition, this work presents the study of the impact of the strategy on such metrics.