ATT Labs

An archive of

AT&T Laboratories

hosted in

Cambridge University Computer Laboratory

The Digital Technology Group

Home Page

RFID Tagging System


Many of the applications of the Bat System involve computing systems reacting to the proximity of one object with another, or the proximity of an object with an area of space. The continuous stream of readings from Bats allow constant monitoring of these interactions, with the result that devices can automatically personalise themselves to the user standing in front of them, and users can trigger a personalised service simply by moving their Bat into a pre-defined area of space. However, might it be possible to develop a system that allowed users to trivially (though not automatically) personalise devices and services, but which had lower infrastructure requirements than the Bat system?

Proximity using RFID tagging

We've built a low-cost, low-infrastructure, proximity-based user interface device that uses RadioFrequency IDentification (RFID) technology.

Small passive RFID tags, each with a unique 64-bit identifier, are attached to equipment and are placed behind posters. Users carry a keyfob-sized personal RFID reader, which also has a unique identifier. Both an RFID tag and reader unit are shown in the photo on the right.

To personalise a device, or activate a service denoted by a poster, a user scans the RFID tag with their reader, by moving the reader to within a few centimetres of the tag and activating the unit by pressing a button. This generates a pair of unique identifiers that describe what action should be performed by the system for which user.

Each reader also has a bidirectional 868MHz radio module, which allows it to communicate with a (software) monitoring service via a set of LAN-connected base stations (a base station unit is shown in the photograph on the right). The reader unit transmits the (tag ID, reader ID) pairs via this radio interface to the monitoring service, which then looks up the appropriate action to be taken based on the RFID tag ID. It then performs that action over the network, personalising it (if necessary) to the user, based on the reader ID. The monitoring service can send feedback to the reader (which has a buzzer and tricolour LED) via the base stations, to indicate the status of the action (e.g. success or failure).

RFID Tagging System features

Many types of RFID tags are suitable for use in the system. Tags cost around 50 cents in medium-scale volumes, are very robust, and come in a range of sizes. We use flat, circular plastic-encapsulated tags with radii of 1-1.5cm, and thicknesses of around 1.5mm.

The reader units measure 6cm x 2.8cm x 2cm, and have been designed to fit on a user's keyring. They have a six-month battery lifetime.

The base station units connect directly to an Ethernet network. They register automatically with the monitoring service via DHCP, and use anticollision protocols to ensure that base station coverage areas can safely overlap. Each base station has a range of around 10m in a typical building. Extending system coverage is as simple as plugging a new base station into the Ethernet LAN.

RFID Tagging examples

This image shows a paper icon that is attached to a workstation. An RFID tag has been placed behind the paper icon. When a user clicks on the icon using their reader unit, their personal VNC desktop is moved to that screen.

This image shows a control panel for a distributed audio system, placed against a bright light. The shadows of the RFID tags that are scanned to control the service can be seen behind the icons.

RFID Tagging People

Andy Ward,Mike Addlesee

Please note that the techniques disclosed here have been the subject of patent applications.

Page last modified on 1/7/05

Contact information
Copyright © 2002 AT&T Laboratories Cambridge