4G Wireless Video Communications

7

Video Error Resilience and Error Concealment

7.1 Introduction

Traditionally, the goal of video compression is to represent the video data with as few bits as possible for a given video quality. If the video is to be stored or transmitted via an error-free channel, this is fine. However, if the only goal is compression efficiency, the resulting bitstream is very sensitive to bit errors. A reason for this is the extensive use of Variable Length Codes (VLC) in video compression. Thus, even a single bit error can cause a loss of synchronization between encoder and decoder, which may even cause the rest of the bitstream to be lost. Therefore, if the video is to be transmitted over a lossy channel, further considerations are necessary beyond pure compression efficiency.

There are three main categories of techniques that help to provide reliable video communications: error resilience, channel coding and error concealment. These techniques are complementary to each other. In order to establish reliable video communications, it is necessary to trade off compression efficiency with resistance to channel errors. Error resilience deliberately keeps some redundant information in the video bitstream in order to make it more resilient to channel losses. Thus, with the use of error resilience, losses in the channel will not have a catastrophic effect and video communication will not be interrupted, although video quality may drop. Some error resilience techniques that will be discussed later ...