Development of self-organizing sensor network applications can be simplified with distributed composition services that cooperate with distributed adaptation and lookup services to enable sensor nodes to be self-aware, self-configurable, and responsive to real-time changes. This research focuses on developing novel concepts, architecture and mechanisms for distributed composition services that support continuous operation of self-organizing applications to meet on-demand needs in spite of ad-hoc deployment, frequent node failures, dynamic reconfiguration, and mobility. The utility of the self-organizing services will be demonstrated in collaborative sensor fusion applications for detecting, classifying and tracking moving objects and distributed sensor-actuator control. The testbed is an ad-hoc network of sensor nodes that are equipped with embedded processors, RF communication, multi-mode sensors and actuators. Sensors nodes are augmented as reconfigurable smart components with capabilities for impromptu networking, self-assembly, dynamically adaptation to device failure and degradation, node mobility, and changes in task and network requirements. They use innovative techniques including: (1) Services and mechanisms for self-organizing sensor applications, (2) scalable hierarchical composition services, (3) lightweight reconfigurable communication mechanisms, and (4) implementation of distributed services over event-based data-centric networks. Broader impacts of this research include the development of advanced and introductory course curriculum and innovative sensor network testbeds and laboratories for education and research training of graduate and undergraduate students including minority students.
