Optimal video sensing strategy and performance analysis for wireless video sensors under delay constraints

Abstract

[ACCESS RESTRICTED TO THE UNIVERSITY OF MISSOURI AT AUTHOR'S REQUEST.] A wireless sensor network (WSN) is a system of spatially distributed sensors that capture, process and transmit information over a mobile wireless ad-hoc network. In wireless video sensor network (WVSN), each sensor, equipped with video capture and processing capabilities, is tasked to capture digital video information about the target event or situation, and deliver the video data to the remote control unit (RCU) for further information analysis and decision making. Very little research has been done to analyze the application-level performance of the individual sensor node. Unlike the conventional types of sensors, the video sensor needs to perform complex data compression, and the video data has large bandwidth requirement and stringent delay constraint, and is highly error sensitive. Therefore, it is necessary to develop theories and technologies to analyze, control, and optimize the complex behavior of individual sensors in the WVSN. In this work, we will analyze the video sensing behavior of the individual nodes and explore its performance limit. We try to develop an answer for the following research question: given a video sensor with an initial resource configuration, what is the maximum application-level performance it can achieve, and what is the optimum video sensing strategy to achieve this performance? In this work, we will develop a framework to analyze the performance of a mobile wireless video sensor under the delay constraint. We will analyze the R-D behavior of the video encoder and study the impact of packet loss on the decoded picture quality. Based on the concept of effective capacity, we will analyze the queuing behavior of the transmission buffer when the channel service rate is time varying. Based on the R-D model for source coding, the transmission distortion for wireless transmission, and the effective capacity model for queuing analysis, we will formulate the end-to-end distortion minimization problem, and determine the optimum operational bit rate for the video encoder.