Implementation and evaluation of a reaction-diffusion based coding rate control mechanism for camera sensor networks

Due to the limitation of wireless communication capacity, a camera sensor network will be easily congested and the perceived video quality considerably deteriorates, when all nodes generate and transmit high-quality video data. To tackle this problem, taking into account that not all video data are equally important from a viewpoint of surveillance or observation applications, our research group proposes an autonomous control mechanism, where each node appropriately determines its video coding rate according to the location and velocity of targets without any centralized control. In this paper, we implemented the mechanism and conducted practical experiments. We verified that the video coding rate was appropriately adjusted and the loss of packets was suppressed. Consequently, the perceived video quality was higher than the cases where the lowest coding rate was always used to avoid congestion and the highest coding rate was always used intending the high quality video.