TÉCNICAS DE CANCELAMIENTO DE ERROR PARA TRANSMISIÓN INALÁMBRICA DE VIDEO

Abstract

This research focuses on restoring or minimizing damage when the video signal has been transmitted through wireless environments. There are many different approaches to protect video transmission system, some of these are applied at the transmitter, others at the receiver and some other approaches use a cooperation between transmitter and receiver signal to ensure the video information. In the pre-compressed video stream, the information traveling through the transmission channel is vulnerable to losses as to receive the video signal there is a high probability that it is degraded even with the best mechanisms to protect the system errors. Therefore, the information obtained at the receiver will result in a video signal with a poor visual quality. If in the transmitter exists the propagation of errors we should apply efficient algorithms and low computational complexity to restore the video signal which is delivered to end users. It is important to note that a video is composed of several images taken in a second. This information is classified by spatial components (within an image) and temporal (within the video sequence) to facilitate the recovery of data lost in transmission. In this research, four error concealment techniques were implemented, two are spatial techniques which are applied on a single image, other two are temporal techniques (using several images from a video sequence) and finally we propose an algorithm which extracts both spatial and temporal information for error concealment. Contribution of this work consists of optimizations made to a space technology, with a visual improvement on the original algorithm was developed for isolated block error. Optimizations were also made for temporal techniques in the search of available motion information, which were obtained improvements in visual reconstruction, although there is further work to develop to improve the objective quality (PSNR) and subjective (visual quality of the image ) of the reconstructed video signal.