Summary
How does network centric organization of sensors affect time critical fusion of dynamic spatial-temporal events in urban environments? This question has become more pertinent today because of the proliferation of multi-source sensor data due to the DoD's paradigm shift to network centric warfare, the urban area monitoring demands of the Global War on Terror, and the collaboration needs of Future Combat System platforms. It demands extensions to network science and engineering that enable the design of sensor networks that can support critical operations in real-time.
We develop theoretical foundations to answer this question with the intent of designing and operating flexible network structures that utilize constrained resources effectively to support dynamic fusion needs. Our approach is based on driving change in the network design space in response to emerging change in the statistical characteristics of the information space. We introduce the fundamental concepts of space-time neighborhoods for dynamic sensor grid formation in the vicinity of an event, and symbolization and nonlinear filtering to formulate rigorous mathematical methods that capture the causal dynamics of distributed fusion processes. We use these concepts to construct a probabilistic theory for measuring the effect of a sensor network structure on hierarchical fusion under uncertainty, partial information, and resource limitations. Building on this analytical framework, we seek to design and operate adaptive sensor network organizations that dynamically cluster sensing, processing and communications resources in space-time neighborhoods of emergent hotspots for progressively fine grained sampling and prediction, and collaborate with other dynamic clusters for event tracking. The sensor network can, thus, support high predictability of microscopic and macroscopic dynamic events.
In sensor network operations, basic tradeoffs exist between energy, information, and their time critical effect on operations. Tradeoffs of architectural design parameters like number of nodes, node placement, routing, clustering density, and resource constraints have been studied in recent years. These influence the in-situ fusion of spatialtemporal information. However, structural properties like hierarchical consistency or resilience are generally treated as emergent behaviors- a consequence of the design. Consequently, network designs based on brute force optimality of architectural parameters may be brittle. Our aim is fundamentally different-we aim to engineer structure for operational dependability and performance under known or unknown perturbations. We thus envision a fundamentally new approach to sensor network operations. Instead of specifying parameters for worst-case design, we postulate designing these systems by organizing a scalable set of diverse resources- sensing, communications and computational, that interact to best support fusion needs in operational environments.
Prominent researchers from Harvard, Duke, Ohio State, and Penn State have jointly formulated this vision building on their multidisciplinary research insights and experiments. Dr. Shashi Phoha from Penn State is the Principal Investigator. Extensive indoor and outdoor experiments are planned for guiding and validating theoretical hypotheses. Our existing test bed, built with previous MURI and DURIP funds, will be enhanced for experimentation, training and technology transition to Government and industry.
This research has the broader implication of altering the computational paradigm to support the DoD's paradigm shift to network centric warfare. Pushing computation outside the box of abstract Turing machines and into the real world, it develops an interactive computational model of the evolutionary physics of the distributed system for adaptive computation. By fusing atomic patterns of causal knowledge embedded in the sensory observations, it incorporates in-situ learning, concurrency, and adaptation. It provides scientific insight and predictability in designing complex network centric warfare operations and evaluating their resilience to un/known operational perturbations. In urban operations particularly, it will provide tactical advantages to war fighters obstructed by urban strucures to overcome fading and obfuscation effects on transient and evasive targets.