Use this URL to cite or link to this record in EThOS:
Title: Adaptive wireless video streaming for in-home environments
Author: Stapenhurst, Robert Karl
Awarding Body: University of Bristol
Current Institution: University of Bristol
Date of Award: 2013
Availability of Full Text:
Access from EThOS:
The challenge of wirelessly streaming high quality video around the home is of considerable commercial interest. Consumers have access to ever-increasing quantities of audio-visual content and a growing choice of options for consuming that content, from large screen televisions to tablet-format mobile devices. Advances in consumer wireless technology (802.11n WiFi) and video compression (H.264/ AVe) make it possible to deliver High Definition (HD) video of good quality around the home over consumer-grade networks. However, multiple clients consuming content over the same network - a scenario becoming ever more typical in home environments - make the above streaming task more challenging. The trend of increasing spatial, temporal and dynamic range resolutions will most likely keep this problem relevant in the near future. In this thesis we propose methods for improving the quality of wirelessly streamed video in a 'home-theatre' scenario. This application assumes HD video resolutions and a viewer who is watching for entertainment purposes on a large display and expects little visible distortion to the content. We target a consumer electronics application where cost - and thus complexity - are important design features. Specifically, we examine two techniques which can be employed in this scenario: rate-adaptive streaming through Hy-pothetical Reference Decoder (HRD) parameter selection and spatiotemporal resolution scaling. The first of these techniques is motivated by the fact that designers of consumer electronics products which incorporate off-the-shelf components often have limited access to the internal parameters of those components. Adaptive streaming techniques which call for fine-grained control over compressed bit stream generation, for example, may not be usable. Instead, we propose a technique for meeting varying channel bandwidth constraints by adaptively specifying HRD parameters to an encoder. We implement the technique in a simulation which operates on a windowed basis, monitoring buffer conditions and specifying parameters for use in the upcoming window, and show that it is able to offer reduced video distortion by adapting to changes in the radio environment. The second technique involves the scaling of spatial and/or temporal resolution prior to compression at a target bit rate. With certain types of content and for certain bit rates, this results in improved perceptual video quality. We create a test set of content impaired id this way, then conduct a large subjective quality assessment of that content. We then use these results to evaluate the performance of a number of common objective video quality metrics. Finally, we present a classification-based method for selecting the appropriate level of spatiotemporal resolution scaling based on low-complexity features of the video.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (D.Eng.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available