Adaptive unequal loss protection for scalable video streaming over IP networks

Video streaming over IP networks faces many challenges due to the varying network conditions such as time-varying packet loss and fluctuating bandwidth. In this paper, we focus on streaming scalable video over IP networks in order to enhance the received video quality. We present an adaptive video transmission scheme, which combines unequal loss protection and layer selecting. In this scheme, we estimate the packet loss effects on video quality using model-based distortion estimation approach, both for the base layer and the enhancement layers. Based the estimated distortion, the scheme adaptively assigns the unequal amount of forward error correction to each layer and adjusts the sending rate by dynamically selecting the appropriate number of layers to transmit, according to the packet loss rate and the available network bandwidth. The scheme can provide a graceful degradation of received video quality as packet loss increasing and available bandwidth decreasing. Simulations results show that our scheme can provide reasonable video quality under varying network conditions.