Implementation of a Peer-based RFID Event Processing Framework

Mr. Xin Tang

Overview

RFID (Radio Frequency Identification) systems are emerging as one of the most pervasive computing technologies today due to their low cost and their broad applicability. RFID systems consist of tiny integrated circuits equipped with antennas (RFID tags) that communicate with their reading devices (RFID readers) using radio-frequency waves without line of sight. This creates tremendous opportunities for linking various objects from real world.

Today, RFID is playing a more and more important role in tracing and tracking objects of interests.
Managing RFID data in a large scale environment presents many challenges. The basic problem is that RFID data is in very large volume and it is not feasible to publish RFID data in a single database. Companies also might be hesitated to do so because of sensitivity of their data. So an acceptable way is that companies maintain their RFID data locally and share the data when necessary with their partners. In particular, peer-based solution poses an opportunity to solve this problem with a self-governed P2P network where organizations can decide the level of data sharing themselves.

This project aims at developing a peer-based framework for RFID event processing. An authorization/authentication module is included to ensure the privacy and security. This project presents an exciting opportunity for the student to obtain sound and fundamental knowledge of this hot area and gain research skills and experience.

Intelligent Returnable Asset Management System

Miss. Yvonne Teo

Overview

According to the Global Retail Theft Barometer, retailers experienced total shrinkage of $105 billion in 2008. Of that, more than half, $62 billion, was reportedly the result of employee theft, vendor discrepancies and various internal errors. With today's economy, these numbers are likely to grow. Curbing supply chain and in-store shrink should be a major initiative for retailers and identifying its source is the first step.

This project aims at developing a real-time returnable asset visualiser tool based on an RFID enabled traceability network. This project presents an exciting opportunity for the student to obtain sound and fundamental knowledge of this hot area and gain research skills and experience.

Available Projects

Lightweight Random Number Generator

Overview

Passive RFID tags require low cost and lightweight random number generators to implement a widely used communication protocol more commonly called the C1G2 Spec. The quality of the random numbers is an important consideration in the collision avoidance strategy used by the protocol and also for encrypting the backward link when sensitive information such as passwords and access numbers needs to be transmitted back to the RFID reader.

You will have the opportunity to develop and test a pseudorandom number generator and assess its suitability for use in passive RFID tags based on the quality of the random numbers and the cost of implementation.

Development of a Product Information Portal Using Web Services

Overview

There are a number of RFID standards, one of the more prominent standards are those advocated by the EPCglobal. These standards cover a complete network stack from the physical layer to the application layer. Communication between peers is achieved by using a centralised architecture.

This project aims at extending the existing framework by developing a lightweight EPC Information Service (EPCIS) based on P2P networking.

A Reconfigurable Supply Chain Model for Traceability Applications

Overview

The ARC funded PeerTrack project (http://www.cs.adelaide.edu.au/peertrack) aims at supporting applications such as anti-counterfeiting, product safety or service supply chain optimisations by solving many of the issues related to using RFID technology for tracking objects or people.

This project aims at developing a tool for building supply chain networks. Using this tool, a user is able to create a model of the supply chain. The tool can automatically generate a graph depicting the product-specific movement inside the supply chain based on observations from the underlying RFID track and trace framework.

Supply Chain Logistics Planning Under Uncertainty

Overview

In dynamic environment of a logistical operation, making fast and effective decisions is key to remaining competitive and reducing costs. One problem that needs to be solved is the dynamic monitoring of logistical operations (e.g., estimated product arrival, asset returns) and alerting exceptional events when operations do not perform according to predicted patterns.

This project aims at developing a probabilistic reasoning model for estimating and predicting product movement in supply chain (e.g., where the product will be next?) using machine learning algorithms. This project presents an exciting opportunity for the student to obtain sound and fundamental knowledge of this hot area and gain research skills and experience.

Past Projects

TnT : A Query Language for Tracing and Tracking Movable Objects

Overview

Summer Scholarship Project 2008-2009.

Implement a query language on top of the existing standard SQL to enable track and trace queries on specific tables with location information. Stored procedures can be used for implementation, but the best way is to extend the current DBMS, like MySQL to achieve this.