A rate-distortion empirical model for rate adaptive wireless scalable video

Resent studies (L. Larzon, et al., June 1999)-(S. A. Khayam, et al., Aug, 2003) have indicated that a significant improvement in wireless video throughput can be achieved by cross layer design with side-information (CLDS) protocols. In (Y. Cho, et al.), for real-time video streaming over wireless LANs the optimal rate prediction architecture under CLDS ( ORPA_CLDS ) was developed. In (Y. Cho, et al.), we showed that an accurate source and channel coding rate prediction can be achieved by utilizing the video quality models and the distribution of channel prediction error process under CLDS. To optimally utilize ORPA_CLDS , the video quality function for rates exceeding the channel capacity (i.e., video rate-distortion function) should be taken into consideration. To this end, in this paper we propose an empirical model for video rate-distortion. To deduce the generic video rate- distortion model, actual scalable video coding (SVC) and low density parity check (LDPC) code were used. The outstanding performance of ORPA_CLDS in accordance with the video rate-distortion model is exhibited by using a comprehensive set of IEEE 802.11b wireless traces.