Macroprogramming Spatio-temporal Event Detection and Data Collection in Wireless Sensor Networks: An Implementation and Evaluation Study

This paper proposes and evaluates a spatio-temporal macroprogramming paradigm for wireless sensor networks (WSNs). The proposed paradigm, called SpaceTime oriented programming (STOP), is designed to reduce the complexity to program event detection and data collection by specifying them from a global viewpoint as a whole rather than a viewpoint of sensor nodes as individuals. STOP treats space and time as first-class citizens and combines them as space- time continuum. A spacetime is a three dimensional object that consists of a two spatial dimensions and a time playing the role of the third dimension. STOP allows application developers to program event detection and data collection to spacetime, and abstracts away the details in WSNs. The notion of spacetime provides an integrated abstraction to seamlessly express event detection and data collection for both the past and future in arbitrary spatio-temporal resolutions. This paper describes the implementation of STOP, and evaluates the performance of applications built with STOP.