Transmission distortion modeling for wireless video communication

Abstract

[ACCESS RESTRICTED TO THE UNIVERSITY OF MISSOURI AT AUTHOR'S REQUEST.] A major challenge in video wireless communication is that the channel is time-varying and error-prone. The lost video packet and the packet received with errors, due to the error-prone wireless network, will cause signal distortion in the decoded video frame. More importantly, the decoding error introduced in one frame will propagate to its subsequent frames along the motion prediction path. This type of video distortion caused by transmission errors is called transmission distortion.In this work, we develop a model for transmission distortion by which encoder is able to estimate the average transmission distortion at the decoder. Our extensive simulation results demonstrate that the model is accurate and robust. In addition, it has very low computational complexity. This type of transmission distortion model has important applications in resource allocation and performance control in wireless video communication.