MidFusion: middleware for information fusion in sensor network applications

Several applications and services employ networks of embedded sensors to monitor events in the environment. To be effective, such applications should cope with sensor network limitations and characteristics, such as resource constraints, dynamicity and heterogeneity. In sensor network applications, there is a need for some level of fusion of information collected from different sensor sources. Research in this critical area is still in its infancy. We propose MidFusion, a middleware architecture that uses a Bayesian theory paradigm to support sensor network applications performing information fusion. MidFusion discovers and selects the best set of sensors on behalf of applications, depending on the quality of service (QoS) requirements and the QoS that can be provided by the sensor networks in a transparent way. We present the theoretical foundation of the MidFusion architecture and demonstrate its effectiveness through an example scenario.