Research On RFID Complex Event Processing Technology In Internet Of Things

Internet of Things (IoT) aims to make active interaction between virtual network and real world, which automatically transforms all of physical information of objects into the virtual-network information in real-time, and thus realizing the visit and exchange of information. Radio Frequency Identification (RFID) becomes one of the core techniques. Advance of RFID technology provides significant new power for humans to sense, understand and manage the world. However, many challenges such as RFID data massive, data correlation and semantic implicit need to be conquered for RFID data processing, so it is necessary to introduce the complex event processing (CEP) technology to the RFID data processing in order to extract valuable information from RFID data flow and process data efficiently.First of all, the RFID event means the abstract modeling of RFID data in RFID data processing and is also the theoretical basis of RFID complex event processing technology. The new features of RFID data and RFID events themselves need to be studied in RFID data processing. There are generally two types of RFID technology applications and they are history-oriented object tracking and object real-time monitoring. The new features of RFID data are summarized from the two types of applications. RFID event is also defined and divided into RFID primitive event and RFID complex event in this paper. Therefore, the RFID events are processed by layers.Furthermore, in order to effectively track and monitor RFID objects, the paper proposes a rules-based RFID data processing framework for RFID data flow processing, including real-time monitoring, data filtering, data transformation and aggregation. On the basis of the processing framework, the paper puts forward a RFID complex event description rule and a rule-matching RFID complex event detecting framework, wherein, the rule can solve the problem that the existing tree-based RFID complex event detection algorithm does not support constraint of attribute levels of event instances and the detecting framework can detect the corresponding complex events in accordance with the given rules so as to efficiently process RFID data streams.Thirdly, the paper puts forward a complex event detection algorithm BT-RCEDA in order to solve the low efficiency problem of detecting non-spontaneous events under RFID event temporal constraints in real-time detection RFID complex events. And Active-Event and Inactive-Event are introduced to the algorithm. The complex events are constructed into event matching tree and the algorithm can detect events by automatically generating consuming events and executive events.Finally, the proposed rule and method are inspected with experiments in performance and function. The results show that the BT-RCEDA increase the detection efficiency of non-spontaneous events and the complex event description rule meets the requirement of CEP.