This thesis proposes adaptive video streaming architectures for a mobile network, compatible with the Scalable Video Coding extension of the H.264 video coding standard, in both timestamp-based and progressive download paradigms. For the timestamp-based architecture, the amount of available radio resources for streaming users is regulated by the congestion situation in the mobile cell. Only minimal cross-layer communication between users’ devices and the adaptation server is required to perform Rate-Distortion optimized adaptation of multiple users. However, no cross-layer operation is required when adaptation is done in an uncoordinated manner. For the progressive download architecture, a statistical model of the best-effort TCP throughput over a mobile channel has been developed and used to make adaptation decisions entirely by the user without any cross-layer operation. Its design also ensures compatibility with the on-going Dynamic Adaptive Streaming over HTTP standard.
«
This thesis proposes adaptive video streaming architectures for a mobile network, compatible with the Scalable Video Coding extension of the H.264 video coding standard, in both timestamp-based and progressive download paradigms. For the timestamp-based architecture, the amount of available radio resources for streaming users is regulated by the congestion situation in the mobile cell. Only minimal cross-layer communication between users’ devices and the adaptation server is required to perform...
»