Time: 10:30 am, March 29, 2010

Title: Processing and Analyzing Sensor Data Streams: Watch the Quality of Your Data

Speaker: Kai-Uwe Sattler, Ilmenau University of Technology, Germany

Abstract: Sensor networks have evolved to a powerful infrastructure component for event monitoring in many application scenarios. Among the possible data processing paradigms, online processing of sensor data streams is a challenging task because the theoretical infinite nature of streams and the fast arrival rate of data prohibit storing and multiple reading of data. In addition to simple filter and aggregation operations, an important task in this context is data mining - the identification of relevant information and patterns. However, such analytical operations can typically use only portions or approximations of the stream. Together with resource restrictions in terms of network, CPU, or memory as well as data quality issues raised by the sensor measurements the quality of results depends on many factors. In this talk, we give a survey on the state of the art of quality of data and quality of service in sensor data stream processing and present results from our own work in the AnduIN project.