Browsing by Author "Issa, Sinzobakwira"

The developments and rapid growth of smartphones and other electronic gadgets have opened a new era of data traffic. According to Sisco, video traffic itself now is estimated to occupy more than 75 percents of data traffic. Researchers have contributed a lot in this area of video streaming. There are many challenges when it comes to deliver quality videos especially for mobile users. The major problem is that, the bandwidth which works well now may not be working sometimes later. This therefore complicates the distribution of the bandwidths for users by ISPs. So many researches focus on system delivery. However users have their own demands for the videos to be perceived. Those users’ demands are usually volatile. They can dynamically change their demands by requesting to view the video with high resolutions. As such, it is extremely challenging to mitigate system demands and users with heterogeneous needs for perceiving satisfactory video quality. This dissertation deals with the design of an effective video streaming for users with demand heterogeneity in terms of quality. To reduce part of those challenges, this dissertation, through the algorithm proposed, it firstly tackles the scalability issue as it is one of the constraints in video streaming. It gives the overview architecture of scalable video coding for adaptive quality. It underlines the selection of different layers of adaptive video and the type of block to be selected in order to stream the videos with satisfactory quality to viewers. Secondly, the design also highlights two main parts of the system bandwidth rescaling. The bandwidth resizing is basically subdivided into two subsections; one part consider resizing system based on adaptive streaming system and the other side peer cooperation as well as the support offered by users. Linear programing is also design as an optimization strategy to resolve the problem of users’ demands of video streaming quality in heterogeneous system. It is basically a trade-off between bandwidth and users’ demand in terms of quality. High Fanout Protocol (HFP) can serve 17 % to 50% more subscribers than Greedy, Centralized, and Snow-ball approaches for large numbers subscribers with delay-bounded video. The algorithms developed show that the perceptual users’ quality is improved, playback delay is reduced and bandwidth consumption is minimized. The system uses mesh overlay and future researches could also considered other forms as well such tree or cloud, or mixture of two.

The development of multimedia propagations and applications has led to a greater expansion in the field of video transmission over heterogeneous media as well as iterative delivery platforms with dedicated content requirements. It is known that conventional video coding systems encode video content with given bitrates adapted to a specific function or application. As a result, conventional video coding does not meet the fundamental requirements of the state-of-the-art flexible digital media application. Meeting the required bandwidth for video transmission is extremely complex resulting in poor video quality. The new technology H.264/AVC, based on scalable video coding appears as a new modus operandi that has the ability to satisfy the underlying requirements. In this thesis a rate distortion optimized video frame dropping strategy which can be applied on active network nodes during high traffic intensity for H.264/AVC was developed. To improve the perceptual quality of coded video in a computation-constrained scenario, through controlling per-frame complexity, a rate-distortion optimised (RDO) rate control algorithm for encoding low bit rate video helped to achieve the target bitrates and PSNR using Lagrangian multiplier function. However simulation was conducted using JM and JSVM software. The choice of a scalable video coding framework in this context brings technical and economical advantages. The result shows that the managed complexity algorithm based on Lagrangian Multiplier function outperforms the normal encoder. The comparison between different video streams proves that the scalable video coding leads to good constructed video quality.