The emerging heterogeneity of data management, programming models and operating platforms in sensor nodes are making the seamless interoperation of devices a challenge. We consider this problem and leverage on the concepts of semantic web and service-oriented design to propose a model that adds a semantics and service-oriented layer to existing network models. This added layer hides the hardware as well as software implementation details from the application layer and thus allows seamless data management between nodes whose capabilities may even be unknown. The proposed model builds on top of well established standards OWL and OWL-S to encapsulate sensor data and services as hierarchical ontology and provides interfaces for service advertisement and discovery. A case study has been discussed whereby the proposed model is used for service-oriented data management (acquisition, archival, processing).
In this paper, we study the problem of sensor relocation for emergent data acquisition (initiated by a base station) in sparse mobile sensor networks. We propose a distributed scheme called BRIDGE that relocates mobile sensors to fulfill an emergent data acquisition task with the objective to minimize the task completion time. BRIDGE gradually finds a sensor that is close to the task location and relocates that sensor to the task location, and at the same time relocates some other sensors to connect that sensor to the base station. BRIDGE exploits the encountered sensors during relocation, and handles the challenges caused by intermittent connections. Our extensive performance study shows the effectiveness of our proposed scheme.
Global climate change can be mitigated with innovative solutions in the ambient intelligence domain that utilize distributed and heterogeneous sensor networks for high resolution sensing of resource utilization (e.g. energy, water, etc.) via intelligent coordination among sensor nodes. We propose the EcoSense framework as a holistic solution that enables dynamic control and reconfiguration of the sensor nodes in an ambient environment based on our resource consumption patterns. EcoSense introduces service and context layers to existing sensor network stacks to enable interoperability between heterogeneous sensor node platforms. A prototype of the framework for an eco-aware smart space has been developed as a pervasive infrastructure for monitoring and minimizing resource consumption at individual levels and also for building eco-aware homes and workspaces.