Research And Achievement Of RFID Middleware Event Processing Model

Radio Frequency Identification is a non-contact automatic identification technology that identifies objects automatically and acquires relevant information through radio frequency signals. Compared with traditional identification techniques, it has advantages of both multi-target recognition identification and quick identification, so it shows widely application prospect in retail, medical and other areas and been known as one of the most promising technologies in the 21st century. RFID middleware is the key software between hardware and application, it provides with a set of application programming interfaces, shielding of the underlying hardware details, the upper application can transparently acquire tag information. RFID middleware provide support for the promotion of RFID technology.In the field of RFID, RFID middleware has not formed unified standards, but event processing function has been recognized one of the core functions,so far event processing research has been launched in lots of IT companies and laboratories, but RFID event processing is a new field and still has the following problems: on the one hand event processing layer are too simple to reflect the relevant application business information; on the other hand event processing was too slow to meet the real-time demanding RFID applications; at the same time, large amount of tag information transmission heavily burden upper application. These problems have seriously restricted the RFID application and its popularity.With above problems,this paper analyzes the existing typical RFID middleware solutions and describes the principle of complex event processing technology and the syntax structure of event processing language, then it designs a three layers RFID event processing model which combined with RFID technology and complex event processing technology with the implementation of data cleaning,patternl-based and business-based event processing,and message transfer key modules.It has proved that this model can customize complex event rules based on user's different needs and transform raw event to business event effectively, especially it reduce number of tags which report to application layer and event processing speed has been increased. So this model has good practice value.