Performance of DASH and WebRTC Video Services for Mobile Users

Abstract

Abstract-With the confluence of the growing market for mobile Internet devices, and users' expectations of instant access to high-quality multimedia content, the delivery of video over wireless networks has become the challenge of the decade. Dynamic Adaptive Streaming over HTTP (DASH) and WebRTC are new and evolving standards that have been developed specifically to meet this demand and enable a high-quality experience for mobile users of video on demand and real time communication services, respectively. However, there has been no systematic study of how these services are experienced by users in a realistic mobile setting. In this work, we describe measurements collected from DASH and WebRTC implementations while moving at walking speeds through an 802.16e WiMAX network. Using data from the application, network, and physical layers, in different wireless environments, we identify characteristics of the cellular data network that directly impact the quality of video service, and suggest areas for further improvement.

(Not) yet another policy for scalable video delivery to mobile users

“…For example, a user who is driving will see a fast-varying channel relative to a pedestrian user; a user in an urban environment will see a slow-varying channel relative to one in a suburban environment [6,7]. The channel models are illustrated in Figure 1b. …”

“…For example, few policies are evaluated for mobile users at both walking and driving speeds, or for users in both urban and suburban environments. Empirical measurements show [7], however, that the temporal behavior of a link (and not just its aggregate statistics) significantly impacts the dynamics of the delivery policy and the delivered video quality. The unanswered question is how the optimal adaptive video policy changes depending on the temporal characteristics of the channel.…”

WVSNP-DASH: Name-Based Segmented Video Streaming

“…For instance, the WDP design allows the rendering of images and data that arrive to the client via Common Gateway Interface (CGI) frameworks [79], which are very important for cross-network exchanges, e.g., between Zigbee and Bluetooth networks. The WDP architecture can also readily take advantage of peer-to-peer features of the client and local hardware access features, e.g., via the emerging Web Real Time Communication (WebRTC) interface [80]. WDP's networking flexibility, combined with WVSNP-DASH's independently playable video segments (i.e., no need for manifest file or initialization segment), enables simultaneous streaming and interweaving of video segments from different video acquisition devices/sensors and different networks, e.g., from Bluetooth, Zigbee, and WiFi networks.…”

WebRTC-Based Remote Collaborative Learning Platform

“…[8] 와 함께 웹 서비스 성능의 향상을 위해서 웹에서 각 장치 요소들의 기능에 접근할 수 있는 Device API [9] , 그리고 브라우저간 실시간 커뮤니케이 션을 위한 WebRTC [10] 등의 표준이 개발되었다. 은 원격 협업 학습에 필요한 실시간 비디오 스트리밍 을 가능하게 한다 [11,12] . 이러한 웹 전반의 성능 발전과 함께, 실시간 원격 학습에 사용될 수 있는 다양한 웹 어플리케이션이 개발되고 있다.…”