Video streaming over Mobile Ad-hoc Networks
Enabling video streaming over Mobile Ad-hoc Networks (MANETs) is a challenging task. Due to
node mobility and the inherent dynamic nature of wireless
links, video packets are dropped or delayed, and paths are
frequently broken. This severely penalizes the received video
quality, causing frequent playback interruption. We combine Raptor coding with receiver feedback
and path diversity to overcome these obstacles. Raptor
coding allows not only to mitigate the effects of packet loss
but also to efficiently disseminate the video contents over
multiple paths.
Low complexity multiview video coding
Multiview Video Coding (MVC) consists of simultaneously capturing a scene with multiple cameras from different viewpoints. Applications include free-viewpoint
television and three-dimentional television. While it gives users an enriched viewing experience, MVC creates huge amount of data that must be compressed. The
aim of this research is to develop techniques to reduce the time complexity of MVC encoding.
Channel-adaptive video streaming with rateless codes
Video streaming over the Internet and packet-based wireless networks is sensitive to packet loss, which can
severely damage the quality of the received video. One common way to combat packet loss is to use forward error correction at the application layer. However, as network conditions are hard to predict, determining the right amount of error protection is not obvious. We propose a system that efficiently addresses this problem by using rateless coding and receiver feedback.
Real-time video communication over 3G networks
Video communication over wireless networks is a challenging task because these networks typically suffer from fading, additive noise,
and interference, which translate into packet loss and delay. Since modern video encoders deliver video packets with decoding
dependencies, packet loss and delay can significantly degrade the video quality at the receiver. Moreover, real-time video
communication for mobile phones poses extra challenges because of the small end-to-end delay and complexity constraints
on hand-held devices. The goal of the research is to develop appropriate techniques based on forward error correction with
rateless codes, smart combination of error tracking and concealment, and cross layer optimization
Rate-distortion optimized media streaming
The project develops enhanced algorithms for streaming video content over the Internet. The basic idea is to select at the streaming server the transmission strategy
that maximizes the expected fidelity at the end-user, subject to a constraint on the transmission bit rate. In
contrast to the existing commercial solutions, the transmission strategy is selected according to
the importance of the packet and its contribution to receiver quality.
Fast progressive transmission of three dimensional graphics models over the Internet
An increasing number of applications require retrieving 3D models over the Internet. For example, users may want to browse a graphic library consisting of many 3D objects, which are typically modeled as 3D triangular meshes. To enable high-quality renderings, millions of triangles may be required for a single object. Compression techniques must be used to reduce the time needed for the transmission of the 3D meshes. However, since compressed data is very sensitive to the impairments that can occur on the Internet, error control techniques must be used as well. In this project, we optimize a progressive transmission system for 3D models that is based on forward error correction with Reed-Solomon codes.
Flexible modulation scheme for digital terrestrial and handheld television
Over the past 10 years, the UK government has been putting increasing pressure onto the television broadcast industries to switch off the existing analogue television broadcast by 2012. The broadcast industries have adopted a technique known as Orthogonal Frequency Division Multiplex (OFDM). This project investigates the use of Orthogonal Wavelet Division Multiplex (OWDM) as an alternative to ODFM. The expected advantages are more flexibility and better quality of service.