Joint source-channel coding and decoding: Towards viable mobile video

点击翻译按钮获取中文摘要

This project is concerned with developing transmission and detection methods and algorithms for improvementof the reliability of video communication in wireless communication networks.The structuring objective of this project is to build a strong research group with a unique focus on theend-to-end design and analysis of mobile video systems. The scientific and technical objectives of this projectare in the improvement of the performance of mobile video systems. We want to develop transceivers that areaware of the communicated video contents. The enhancement is expected to come through the interactionbetween the video encoder and the data transmitter, as well as through the collaborative communicationbetween the data receiver and the video decoder.Namely, at the transmitter, we propose to adjust the protection available in various modes of the datatransmission to varying sensitivity of the compressed video. To this end, we will analyze the degrees offreedom offered in the data transmission due to multi-dimensional transmission. We will also develop thesensitivity models for the outputs of the Scalable Video Compression (SVC) compliant with H.264 standard.At the receiver side, we propose to improve the video quality combining the spatial/temporal visual errorconcealment with the iterative decoding of the channel codes. We will also study how such jointsource-channel decoding impacts the design principles of the video encoder and data transmitter.During this project, four Ph.D. and three undergraduate students will receive multidisciplinary training incutting edge areas of data transmission and video (de)compression.