High-Level System Support for Automatic-Identification Applications Matthias Lampe and Christian Floerkemeier Institute for Pervasive Computing, ETH Zurich, CH-8092 Zurich, Switzerland {lampe, floerkem}@inf.ethz.ch Abstract. RFID systems have begun to find greater use in tagging consumer goods, in industrial automation, and in mobile asset and supply chain management as an enabling technology for smart objects. This introduces the need for software systems that manage RFID readers, filter and aggregate captured RFID data, combine and enrich the RFID data with application logic, and generate appropriate business events. We concentrate on the higher levels of a smart object system and address several challenges that application developers face when implementing Auto-ID applications. Derived from a requirements analysis, we propose an object model based on a symbolic location model. We also present a prototypical implementation of the model, the Object Monitoring System (OMS) that includes object persistence, query capabilities, and business event generation and present a case study as a proof-of-concept. We argue that our approach offers appropriate object and programming models that abstract from RFID specific details and provide the necessary services and the adequate level of reuse to facilitate the development of Auto-ID applications.