In this CAREER project, the development of wireless underground sensor networks, which carries information through soil, is investigated. More specifically, the application of underground networking in agriculture is considered, where underground sensor networks promise significant reduction in water usage for irrigation. The objectives of this project are to establish the foundations of underground networking for the realization of underground sensor networks, develop hands-on educational tools, and provide experiences to facilitate the dissemination of these techniques to a larger audience including students and farmers. Communication performance in underground settings is significantly affected by the variations in soil conditions. Hence, the research activities are focused on revisiting the concepts of connectivity and interference under the influence of environmental factors such as soil composition, soil moisture, and depth. Moreover, a theoretical framework is developed to capture the spatial and temporal correlations in soil moisture for underground sensor networks, enabling the development of environment-aware communication protocols. The insights from these analyses are exploited to develop event-based cross-layer communication platforms that adapt to the environment in an energy efficient manner. Finally, an agricultural underground sensor network testbed is developed to evaluate and highlight the outcomes of this research. The educational components include bringing wireless sensor networking to the class, developing an interactive cyber-physical networking lab, and providing hands-on experiences for "little farmers". The realization of the underground networking techniques has the potential to transform the agriculture industry, as well as a broad range of applications including border patrol, perimeter surveillance, and toxic material monitoring.

Cyber-physical systems (CPS) are characterized by extremely tight integration of and coordination between computational and physical resources. CPS integrate computation, communication, and storage capabilities through systems of systems that must interact with the physical world in real-time at multiple time scales and often at multiple spatial scales. The inherent heterogeneity and the non-deterministic operation of different components in these systems pose new challenges to traditional control, communication, real-time scheduling, and robotics disciplines. In conjunction with the IEEE Real-Time Systems Symposium in 2009 (RTSS 2009), this project helps to support a Ph.D. student forum to discuss (i) the set of interdisciplinary research problems that arise in the context of cyber-physical systems, (ii) novel applications that become possible thanks to the integration of computing, communication, and interaction with the physical world at scale, and (iii) initial system architecture that addresses some of these research problems. The primary goal is to help students (and the real-time community) recognize that cyberphysical systems are different from the over-engineered real-time embedded systems of the past, and to provide a forum by which students can discuss their proposal for addressing the complicated aggregate systems issues that arise in this context. As such, we need to encourage constructive debate on emerging research topics. A secondary goal is to encourage student involvement in new research directions and offer a channel to discuss and reward the most innovative student ideas in this exciting emerging research field. Advisors and students will be welcome to attend the forum, but the focus will be on training and motivating the next generation of researchers.