Scenario-Adaptive Rate Allocation for Multi-Source Video Streaming over Ad Hoc Networks

We propose a scenario-adaptive dynamic rate allocation scheme among multiple video sources over an ad hoc network. The most appropriate rate allocation scheme can be selected with a total rate constraint adopting the media access characteristic of an 802.11 network and the selected route for each sender. The selected scheme optimizes the rate-distortion. The scheme is performed at the receiver of the video streams and the decision is based on a simple rate-distortion model of video and a fast algorithm to classify the activity level of the video content. Experimental results for a multi-source surveillance show that the proposed scheme outperforms the conventional fixed quality scheme and can adapt to different scenarios by adjusting the parameter b dynamically