VBR Video Traffic Models

There has been a phenomenal growth in video applications over the past few years. An accurate traffic model of Variable Bit Rate (VBR) video is necessary for performance evaluation of a network design and for generating synthetic traffic that can be used for benchmarking a network. A large number of models for VBR video traffic have been proposed in the literature for different types of video in the past 20 years. Here, the authors have classified and surveyed these models and have also evaluated the models for H.264 AVC and MVC encoded video and discussed their findings.

There has been a phenomenal growth in video applications over the past few years. An accurate traffic model of Variable Bit Rate (VBR) video is necessary for performance evaluation of a network design and for generating synthetic traffic that can be used for benchmarking a network. A large number of models for VBR video traffic have been proposed in the literature for different types of video in the past 20 years. Here, the authors have classified and surveyed these models and have also evaluated the models for H.264 AVC and MVC encoded video and discussed their findings.

INTRODUCTION viiCHAPTER 1. VIDEO CODING 11.1. Video coding 11.2. Video coding standards 31.2.1. The MPEG video coding standard 41.2.2. H.264/MPEG-4 AVC 81.2.3. H.264 SVC 101.2.4. H.264 MVC 121.3. Rate control 151.4. Summary 16CHAPTER 2. VIDEO TRAFFIC MODELING 192.1. The AR models 192.1.1. Review of the AR process 202.1.2. Survey of AR video traffic models 222.2. Models based on Markov processes 322.2.1. Review of Markov process models 332.2.2. Survey of Markov process models 352.2.3. Summary 422.3. Self-similar models 432.3.1. A survey of self-similar models for video traffic 442.3.2. Summary 462.4. Wavelet-based models 472.4.1. Survey of wavelet-based video traffic models 472.4.2. Summary 492.5. Other approaches 502.5.1. The M/G/ infinity process 502.5.2. The SARIMA model 512.5.3. TES-based models 532.5.4. Summary 542.6. Video traffic models for layered scalable video 542.6.1. Summary 572.7. Video traffic models for three-dimensional video 582.7.1. A video traffic model for MVC video 602.8. Conclusion 62CHAPTER 3. EVALUATION OF VIDEO TRAFFIC MODELS FOR H.264 AVC VIDEO 653.1. Model implementation 673.1.1. The DAR(1) model 673.1.2. A frame-based AR(2) model 693.1.3. A Markov-modulated gamma mode 703.1.4. A wavelet model 723.2. Experimental setup 733.3. Frame size distribution and ACF comparisons 743.4. QoS evaluation 813.4.1. End-to-end delay 813.4.2. Jitter 823.4.3. Packet loss 833.4.4. The simulation model 843.4.5. Results 863.5. Conclusion 94CHAPTER 4. EVALUATION OF VIDEO TRAFFIC MODEL FOR H.264 MVC VIDEO 974.1. A video traffic model for MVC video 974.2. Experimental setup 994.3. Results 1004.3.1. Q-Q plots and ACF comparisons 1004.3.2. QoS evaluation 1004.4. Conclusion 113CONCLUSION 115APPENDIX 119GLOSSARY 131BIBLIOGRAPHY 135INDEX 145