Rate Adaptation for Conversational 3G Video

Abstract

Abstract-Wireless cellular environments, such as UMTS, are often affected by congestion and errors, which are inherent to wireless transmission channels due to fading, interference, resource scarcity, mobility, etc. For a conversational video application to be successful i.e., to provide good viewing quality to the receiver at all times, the sender must be able to quickly adapt its sending/encoding rate (and other related parameters) to that offered by the link. Moreover, for a rate adaptation scheme to be successful, the receiver must provide timely feedback in order to mitigate further losses due to congestion. In this paper, we investigate different rate adaptation mechanisms and redefine them for 3GPP networks, reusing existing RTCP extensions standardized in the IETF and in 3GPP where possible.

Empirical Evaluation of HTTP Adaptive Streaming under Vehicular Mobility

“…Prior works [16,18,20] have proposed mechanisms for implementing adaptive streaming by extending traditional streaming protocols. The basic concept involves the server and/or client relying on some rate control algorithms/protocols to infer the actual bandwidth conditions.…”

“…[11]. The performance of the non-HTTP adaptive streaming techniques have been evaluated and studied in previous works [16,18,20]. However, evaluations were conducted using simulations with synthetic bandwidth data.…”

“…To mitigate the effect of the bandwidth dynamics and achieve the smooth streaming experience, adaptive streaming is emerging as the promising solution [2-4, 9-12, 16,18,20]. In adaptive streaming, the streaming servers and/or clients constantly monitor real-time end-to-end network conditions.…”

“…As a result, adaptive streaming delivers a better quality video stream as compared to conventional non-adaptive approaches. The principle of adaptive streaming has been incorporated with traditional streaming protocols, such as RTSP [11], and evaluated in various prior works [16,18,20]. However, due to the popularity of using HTTP in video streaming, the common trend in the industry is to use adaptive HTTP progressive download [2-4, 9-12].…”

Improving QoS in High-Speed Mobility Using Bandwidth Maps

“…Several such rate adaptation algorithms have been proposed for streaming applications [17], [24], [25], [26]. These algorithms are reactive in nature, in that, they monitor the video playout buffers or certain end-to-end network parameters (i.e., delay and packet loss) and adapt the streaming rate depending on the changes in the observed values.…”