Adaptive architecture for radio frequency identification middleware

Abstract

Middleware is a software that separates the hardware and application layer. In Radio Frequency Identification (RFID) environment, middleware faces many challenges in adaptation to multi readers, multi applications, and changes in readers or applications environment. Different readers may cause RFID middleware not to work properly during runtime if previous configuration does not match with the current one. Some changes such as adding a new reader or applications could also disturb the operation of RFID middleware. Hence adaptive components in RFID middleware could provide a solution for the problem. This research proposes an RFID middleware architecture that provides two adaptive components; reader detector and message adapter. Reader detector allows the middleware to communicate with multi readers. One unique feature of the reader detector is its ability to autodetect the presence of multi readers and the changes occurred in the reader environment. Reader detector gives response as a reaction to the changes to ensure RFID middleware performs correctly in runtime. Message adapter detects new application, resolves each application name to its IP address, and checks its activity in distributed network. To validate the proposed adaptive middleware, the architecture and adaptive components are tested using various RFID readers with short, medium and long range signal readers. Furthermore, the functionality and performance of the message adapter are tested on several applications developed such as the Operation Theater Tracking (OTT), Integrated Facilities and Asset Maintenance Management System (IFAMMS), and E-Point of Sale (ePOS) that utilize different manufacturer and specification of RFID readers. The results show that during runtime, the proposed architecture has succesfully detected different readers and various applications automatically and are able to adapt to changing situation in the reader or application environment.