Skip to main content
eScholarship
Open Access Publications from the University of California

UCLA

UCLA Electronic Theses and Dissertations bannerUCLA

RFID Asset Management Solution with Cloud Computation Service

Abstract

This work proposes the deployment of a web-based architecture of an RFID tracking solution. An extensive literature survey has been conducted to find existing RFID web based architectures and the current requirements of supply chain based organizations. These findings were used to design a web based RFID tracking system with a goal of reducing set up time and infrastructure for the client. The architecture attempts to depart from the typical localized approach of most RFID middlewares. The middleware has been replaced by a web service residing in a cloud. The traditional roles of the middleware such as tag data filtering, sorting, storage and event processing have been redistributed amongst an RFID reader-based application and the web service. This architecture provides more flexibility and control to large scale RFID-based supply chain operations which are global in nature and unsuitable to the traditional localized

iii

middleware based architecture of RFID tracking solutions. The feasibility of the web based architecture is further augmented by validating performance parameters such as tag processing and primitive and complex event processing delay. Different kinds of RFID readers are also tested to ensure the suitability of the architecture. Different sub-architectures are proposed based on performance data for different scales of operations. Event processing capability on a reader device is demonstrated and tested, where occurrence of certain specified sequence of events results in automated notifications sent to the clients by means of e-mail, text messaging etc. It is concluded that for small number of tag operations (<400) or small scale businesses, performing complex event processing at the reader level is faster and more economical. For large scale tag operations (>400), performing only partial event processing at the reader level leads to lowest event trigger delays.

Finally, the event processing algorithm is demonstrated as a proof of concept for an RFID enabled parking lot to control and schedule charging and discharging of EV's and inform their users of the charge status. The RFID triggered events are used to automatically set charge/discharge preferences of the EV owner for the duration for which the EV is parked. This was supplemented by designing a charge/discharge algorithm to optimize the consumption of energy by parking lots.

Main Content
For improved accessibility of PDF content, download the file to your device.
Current View