All-IP network video streaming through interval type-2 fuzzy logic congestion control

All-IP network delivery of video is a promising application of congestion control. This paper compares the advantage of a fuzzy logic controller over existing controllers for this type of network. The acceptance of fuzzy logic for video control has gained ground but its advantages will become even more apparent through interval type-2 (IT2) logic rather than traditional type-1 logic. This paper demonstrates that the new logic provides robustness to uncertainty both in modeling the network and in measuring congestion. The paper establishes that under conditions of heavy congestion, with multiple video sources, an IT2 fuzzy controller consistently outperforms traditional controllers, resulting in an improvement of several dB in video quality when streaming across a bottleneck link.