Wireless video sensor networks have important applications. In addition to their traditional surveillance and monitoring applications, these networks have recently found many new applications for enhancing national security, reducing the impact of security breaches on the nation's critical infrastructure, and improving the government's ability to prevent, detect, respond to, and recover from both man-made and catastrophic events. This project will examine a wireless video sensor network that is designed for surveillance over an extended period of time. Consequently, this project will explore several definitions for network lifetime, taking into account the percentage of nodes required to remain alive (corresponding to network coverage) and the different priorities of nodes based on their locations.
The PIs will consider several important factors that affect network lifetime, including base-station placement and network connectivity topology. An analysis of power dissipation behavior at video sensor nodes suggests that communication consumes significantly more energy than any other node activity. This power consumption can be controlled effectively by monitoring and adjusting the power level of the transmitter and the network traffic routing topology. By adjusting the topology of the network through intelligent base station placement and network connectivity among the video sensor nodes, we can optimize the transmitter power of video sensor nodes, thus extending network lifetime. We have found that techniques and algorithms developed in the field of computational geometry can help us obtain a good understanding of the nature of the problem and identify where difficulties lie. More important, such approaches enable us to develop very useful performance bounds to characterize properties of wireless video sensor networks, which is the basis to further development of customized solutions. In this research, we plan to further explore algorithms and techniques from computational geometry to study the impact of topology control (base-station placement, temporal topology, traffic routing) on network performance and characterize basic properties for wireless video sensor networks. Proposed research components include, among others, joint problems of base-station placement and multihop routing, considerations of constraints on base-station placement, and dynamic varying network topology.
In addition to theoretical study, a software toolkit will also be developed to implement the intelligent topology control techniques that are developed in this research. The proposed research will also foster the integration of research and education at the investigators' institutions and will have a significant impact on education. Students working on this project will have a unique opportunity to explore creative approaches and innovative methods for solving IT problems. In addition, some results from this research will be incorporated into an advanced topics class on wireless and sensor networks at the PIs' institutions.
