RFID Explained PDF

P1: IML/FFX P2: IML/FFX QC: IML/FFX T1: IML
MOBK032-FM MOBK032-Want.cls December 8, 2006 17:36

RFIDExplained: APrimer on
Radio Frequency Identication
Technologies
i
Copyright 2006 by Morgan & Claypool
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or
transmitted in any form or by any meanselectronic, mechanical, photocopy, recording, or any
other except for brief quotations in printed reviews, without the prior permission of the publisher.
RFID Explained: A Primer on Radio Frequency Identification Technologies
Roy Want
www.morganclaypool.com
ISBN-10: 1598291084 paperback
ISBN-13: 9781598291087 paperback
ISBN-10: 1598291092 ebook
ISBN-13: 9781598291094 ebook
DOI 10.2200/S00040ED1V01200607MPC001
A Publication in the Morgan & Claypool Publishers series
SYNTHESIS LECTURES ON MOBILE AND PERVASIVE COMPUTING #1
Lecture #1
Series Editor: Mahadev Satyanarayanan, Carnegie Mellon University
Series ISSN: 1933-9011 print
Series ISSN: 1933-902X electronic
First Edition
10 9 8 7 6 5 4 3 2 1
Printed in the United States of America
RFIDExplained: APrimer on
Radio Frequency Identication
Technologies
Roy Want
Intel Research
M
&C
Morgan
&
Claypool Publishers
iii
iv
ABSTRACT
This lecture provides an introduction to Radio Frequency Identication (RFID), a technology
enabling automatic identication of objects at a distance without requiring line-of-sight. Elec-
tronic tagging can be divided into technologies that have a power source (active tags), and those
that are powered by the tag interrogation signal (passive tags); the focus here is on passive tags.
An overview of the principles of the technology divides passive tags into devices that use either
near eld or far eld coupling to communicate with a tag reader. The strengths and weaknesses
of the approaches are considered, along with the standards that have been put in place by ISO
and EPCGlobal to promote interoperability and the ubiquitous adoption of the technology. A
section of the lecture has been dedicated to the principles of reading co-located tags, as this
represents a signicant challenge for a technology that may one day be able to automatically
identify all of the items in your shopping cart in a just few seconds. In fact, RFID applications
are already quite extensive and this lecture classies the primary uses. Some variants of modern
RFID can also be integrated with sensors enabling the technology to be extended to measure
parameters in the local environment, such as temperature & pressure. The uses and applica-
tions of RFID sensors are further described and classied. Later we examine important lessons
surrounding the deployment of RFID for the Wal-Mart and the Metro AG store experiences,
along with deployments in some more exploratory settings. Extensions of RFID that make
use of read/write memory integrated with the tag are also discussed, in particular looking at
novel near term opportunities. Privacy and social implications surrounding the use of RFID
inspire recurring debates whenever there is discussion of large scale deployment; we examine the
pros and cons of the issues and approaches for mitigating the problems. Finally, the remaining
challenges of RFID are considered and we look to the future possibilities for the technology.
KEYWORDS
Automatic identication, distributed memory, electronic tagging, passive tagging, privacy
debate, radio frequency Identication (RFID), remote sensing
v
Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Lecture Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2. Principles of Radio Frequency Identication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.1 Near-Field-Based RFID Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.2 Properties of Magnetic Fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.3 Far-Field-based RFID Design. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.4 Properties of Backscatter RF Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.5 Modulation Techniques Used with RFID. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.6 Comparison of the Properties of RFID Based on Frequency . . . . . . . . . . . . . . . . . 15
3. RFIDIndustry Standards. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.1 EPCglobal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.1.1 Generation-1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.1.2 Generation-2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.1.3 EPC Packet Formats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.2 ISO 15693 Vicinity Cards and RFID . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.3 ISO 14443 Proximity Cards and RFID. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.4 The NFC Forum. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
4. Reading Collocated RFIDTags . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.1 Query Tree Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
4.2 Query Slot Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
4.3 Summary Tag Read Rate Timing for EPC Generation 1 and 2 . . . . . . . . . . . . . . 28
5. Applications of RFIDTagging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
5.1 Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
5.1.1 Access Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
5.1.2 Verication. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
5.1.3 Antitheft . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
5.2 Tracking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
5.2.1 Supply Chain and Inventory Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
5.2.2 People . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
vi CONTENTS
5.2.3 Hospital Patients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
5.2.4 Runners . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
5.2.5 Cattle. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
5.2.6 Pets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
5.2.7 Airline Luggage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
5.3 Authenticity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
5.3.1 Money. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
5.3.2 Drugs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
5.4 Electronic Payments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
5.4.1 Auto Tolls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
5.4.2 Electronic Tickets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
5.4.3 Electronic Credit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
5.5 Entertainment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
5.5.1 Smart Toys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
6. RFIDIncorporating Sensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
6.1 Extending RFID to Sensing Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
6.2 Monitoring Physical Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
6.3 Tamper Detection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
6.4 Detecting Harmful Agents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
6.5 Non-invasive Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
6.6 Logging Sensor Activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
6.7 Longer Range Sensing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
6.8 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
7. Deployment and Experience with RFIDSystems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
7.1 Store of the FutureMetro AG . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
7.2 Wal-Mart RFID Trials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
7.3 RFID Support for Maintenance Operations at Frankfurt Airport . . . . . . . . . . . . 52
7.4 Intel Research: Ibracelet and Detecting the Use of Objects . . . . . . . . . . . . . . . . . . 53
7.5 University of Washingtons RFID Ecosystem Project . . . . . . . . . . . . . . . . . . . . . . . 55
7.6 Future Deployments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
8. Privacy, Kill Switches, and Blocker Tags . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
8.1 Kill Switches. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
8.2 Blocker Tags. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
8.3 Tagging is Already an Integral Part of Modern Living . . . . . . . . . . . . . . . . . . . . . . 59
8.4 Future Impact on Society . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
CONTENTS vii
9. Opportunities for RFIDIntegrated with Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
9.1 Read-only Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
9.1.1 Enhancing Objects with RFID Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
9.2 Read/Write Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
9.2.1 Location and Directions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
9.2.2 Memory and Time. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
9.2.3 Another Use of Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
9.2.4 Facilitating Wireless Connections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
9.3 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
10. Challenges, Future Technology, and Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
10.1 Core Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
10.1.1 Design. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
10.1.2 Cost . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
10.1.3 Acceptance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
10.2 Additional Challenges for Short-range RFID. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
10.3 Future Technologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
10.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
viii
List of Figures
1.1 Examples of (a) 1D (39 Code) and (b) 2D barcodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 RFID tagsvarious shapes and sizes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.3 RFID reader and tag . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.4 Comparing barcodes and RFID. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.5 Forecast volume RFID tags shipped 20052010 (1000s). Source: Instat 12/05 . . . . . . . 5
2.1 Logical components of an RFID tag. Note that the antenna can take many
forms including a coil and a dipole depending on the tag type . . . . . . . . . . . . . . . . . . . . . . 8
2.2 Near-eld power/communication mechanism for RFID tags operating at less
than 100 MHz . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.3 RFID tags based on near-eld coupling. (a) Trovan tag (128 kHz), size:
1 cm [17]. (b) Tiris (13.56 MHz), size: 5 cm 5 cm [18] . . . . . . . . . . . . . . . . . . . . . . . . 10
2.4 Magnetic eld calculation at the center of a coil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.5 Magnetic ux density (B) at a distance x from the center of an N-turn coil,
with radius r , and current owing, I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.6 RFID tags based on far-eld coupling. (a) Alien (900 MHz), size: 16 cm 1 cm.
(b) Alien (2.45 GHz), size: 8 cm 5 cm. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.7 Far-eld power/communication mechanism for RFID tags operating at greater
than 100 MHz . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.8 Backscatter signal strength at a distance x . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.9 Modulation coding options for RFID signaling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.10 Comparing the properties of RFID operating in different frequency bands . . . . . . . . . 15
3.1 Format of a 96-bit EPCglobal tag . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.2 The Texas Instruments Generation-2 tag. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.3 The Nokia 3200 cell phone features an NFC reader: Front sideIt looks like an
ordinary cell phone. Back sideyou can see the reader coil molded into the
housing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
4.1 Arbitration mechanism used by EPCglobal Generation-1 Class-0 . . . . . . . . . . . . . . . . . 24
4.2 Example of the Query Slot Protocol used in EPCglobal Generation-2 . . . . . . . . . . . . . 27
5.1 The iClass RFID identity card from HID. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
5.2 Automobile ignition key with additional RFID activation. . . . . . . . . . . . . . . . . . . . . . . . . 31
5.3 RFID tags and handheld readers assure security patrols are carried out
consistently and according to a predened schedule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
LISTOF FIGURES ix
5.4 Subdermal RFID tag in the arm Kevin Warwick (University of Reading) . . . . . . . . . . 33
5.5 (a) An RFID tag on a shoelace. (b) This system is used in the Boston Marathon . . . . 34
5.6 (a) RFID tag mounted in the ear of a cow. (b) Dog with subdermal tag being
identied with a handheld reader . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
5.7 An airline luggage label that provides RFID, barcodes, and printed information
about the owner, ight, and destination of the tagged bag. . . . . . . . . . . . . . . . . . . . . . . . . 36
5.8 Hitachis Mu-chipso small it can be sandwiched between the paper layers
of a banknote. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
5.9 FasTrak Toll Pass System: (a) Transponder tag and (b) Booth . . . . . . . . . . . . . . . . . . . . . 38
5.10 Ski pass with embedded RFID serves as a ticket to enter the chair lift . . . . . . . . . . . . . . 39
5.11 A Star-wars character from Hasbro. Placing different characters on the podium
plays the sound and voice of that character . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
6.1 An RFID tag that can detect a critical temperature threshold . . . . . . . . . . . . . . . . . . . . . 42
6.2 A packaged chicken incorporating an RFID temperature sensor . . . . . . . . . . . . . . . . . . . 43
6.3 Car tire incorporating RFID pressure sensor readable from the car . . . . . . . . . . . . . . . . . 44
6.4 Auburn University RFID bacterial sensor chip (www.auburn.edu/audfs). . . . . . . . . . . . 45
6.5 Showing a prototype power scavenging WISP tag . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
7.1 Metroa future store used to test new RFID concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
7.2 One of Wal-Marts early RFID trials stores . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
7.3 RFID maintenance tagalso includes a barcode for redundancy . . . . . . . . . . . . . . . . . . 54
7.4 The iBracelet created by Intel Research Seattle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
7.5 Paul Allen CS Building at University of Washington . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
8.1 Scene from the movie Minority Report in which billboards customize
themselves to the shoppers in the vicinity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
8.2 (a) Tesco trial store in Cambridge, UK. (b) Protester outside Tesco in the UK
voicing concerns about RFID trials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
9.1 A movie poster with additional information provided by embedded RFID tags . . . . . 64
9.2 (a) RFID thumb tags. (b) A control in the forest with RFID writer station and
position to insert the thumb tag . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
10.1 Self-assembly techniques being pioneered by Alien Technology Inc . . . . . . . . . . . . . . . . 73
10.2 (a) Philips experimental set-up for testing a plastic RFID Tag. (b) A prototype
plastic RFID tag up-close . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
x
Acknowledgements
The author would like to thank Satya for providing the opportunity to write about RFID for
the Morgan Claypool Synthesis Lecture Series, and Waylon Brunette, Adam Rea, Gaetano
Borriello, Trevor Pering, Josh Smith, and his family Susan Want and daughters Hannah and
Becky for their constant support.
P1: JYS
MOBK032-01 MOBK032-Want.cls December 8, 2006 17:31
1
C H A P T E R 1
Introduction
Ever since the advent of large-scale manufacturing, rapid identication techniques have been
needed to speed the handling of goods and materials. Historically, printed labels, which are
a simple cost-effective technology, have been the staple of the manufacturing industry. In the
1970s, labeling made a giant leap forward with the introduction of UPC barcodes [1] making it
possible to bothautomate and standardize the identicationprocess. Barcodes [2], althoughvery
inexpensive to produce, have many limitations: a clear line of sight is needed between the reader
and the tag, they can be obscured by grease and nearby objects, are hard to read in sunlight or
when printed on some substrates (Fig. 1.1). There are many types of barcode in use for specialty
applications, includingblock-basedoptical codes. There are evenminiture plastic barcodes called
taggants [3] incorporated into explosives that are design to withstand an explosion and identify
the supplier in case used for criminal purposes. However, for the scope of this article, barcodes
labels and optical codes are considered as a single group. An alternative labeling technology
is Radio Frequency Identication (RFID) [46], which enables identication at a distance
without a line of sight. To provide some context Fig. 1.2 shows a variety of RFID tags designed
for diverse applications in comparison to the size of a dime. Figure 1.3 shows a typical tag reader
and remote antenna that can be installed in the area the tags are expected to pass through. It
should be noted that as RFID is a radio technology, the tags do not need to be visible at all,
and can be concealed behind an aesthetically designed label, or even molded into the product
housing itself. For these reasons it is possible that many people may have encountered RFID
tags but were unaware of themdue to their invisible placement. Electronic tagging is superior to
barcodes in many ways as it can reliably support a much larger set of unique IDs, and incorporate
additional data, such as the manufacturer, and product serial number (Fig. 1.4). Furthermore,
RFID systems can discern many different tags that are located in the same general area without
human assistance. In contrast, consider the individual care needed at a supermarket checkout
where each item is carefully orientated toward the reader before it can be scanned.
RFIDis not a newtechnology. For example, the principles of RFIDwere employed by the
British in World War II to identify their aircraft using the IFF system (Identity: Friend or Foe).
Later, work on access-control that is more closely related to modern RFID, was carried out at
Los Alamos National laboratories during the 1960s. In this application, RFIDtags incorporated
P1: JYS
2 RFIDEXPLAINED
FIGURE1.1: Examples of (a) 1D (39 Code) and (b) 2D barcodes
FIGURE1.2: RFID tagsvarious shapes and sizes
FIGURE1.3: RFID reader and tag
P1: JYS
INTRODUCTION 3
FIGURE1.4: Comparing barcodes and RFID
in employee badges enabled automatic identication of people to limit access to secure areas,
and had the additional advantage that it made the badges hard to forge. For many years this
technology has been relatively obscure, although it has been adopted in various niche domains,
such as to identify animals, make toys interactive, improve car-key designs, label airline luggage,
time marathon runners, prevent theft, enable automatic toll-way billing, and many forms of ID
badge for access control. Today, it is even being applied to validate money and passports, and
as a tamper safeguard for product packaging.
In recent years, RFID has been widely written about, and even appeared in a primetime
television advertisement as a promotion of IBMs business solutions. The technology has moved
from obscurity to applications that are now rmly in the public eye, and the ethics of its use
are regularly debated by journalists, technologists, and privacy advocates. You may ask why
it took over 50 years for the technology to become mainstream? The primary reason is one
of cost. When electronic identication technologies compete with the rock bottom pricing of
printed symbols on paper, it either needs to be equally low-cost, or provide enough added value
to an organization that the cost is recovered elsewhere. RFID is now at a critical price-point
that could enable its large-scale adoption for the management of consumer retail goods. At the
time of writing this article, Alien Technologies [7] are able to supply a modern RFID tag at a
P1: JYS
4 RFIDEXPLAINED
unit price of 12.9 cents in quantities of 1 million, notably still quite a bit more expensive than
printed sticky labels. When adoption begins to take hold, it will rapidly accelerate as volume
production drives prices down, making it more attractive to deploy the technology to support
a wider range of markets. Modern semiconductor manufacturing has also played a role in the
progress of RFID design, driving up the functionality of the tags, at progressively lower power,
using a smaller area of silicon, which in turn lowers cost. The sensitivity of the receiver in the
reader has also increased and these improvements in the reader and tag can now be achieved at
low manufacturing costs (see Section 2).
Today, applications of RFID are being extended to new domains. The European Union
is considering putting RFID into ECUs; a small chip that is sandwiched between the layered
papers of the European paper denomination. The objective is to make forgeries more difcult
and provide automatic tracing of its usealthough in the latter case, some people feel this
undermines the benets afforded by paper money.
On a similar theme, the US government is also planning to incorporate RFIDinto the US
passports to reduce counterfeiting and enable efcient automatic checks at the national border.
RFID is also being used to tag family pets, and in some states a dog registration process
involves injecting a sub-dermal tag under the skin of the dog. Unlike a neck collar the tag cannot
be accidentally removed, or fall off, and even intentional theft of the animal requires surgical
skills if its identity is to be obscured. RFID is already responsible for reuniting many runaway
dogs with their owners.
There are three primary organizations that are pioneering the adoption of RFID on a
large scale: Walmart, Tesco, and the US DoD. Each is driven by the potential to lower their
operational costs in order to have the most competitive product pricing. RFID, with its fast-read
times and high-reliability can do this by streamlining the tracking of stock, sales, and orders.
When used in combination with computerized databases and inventory control, linked together
by digital communication networks across the globe, it is possible to pinpoint the progress of
individual items between factories, warehouses, transportation vehicles, and stores.
The potential benets of automatic RFIDtracking yielding improvements inefciency are
alluring to large companies that are trying to squeeze the cost out of manufacturing, distribution,
and retail within their organization. The economics of this attraction will be a major force in
the adoption of the technology and will also drive improvements in its own evolution through
the resulting investment. It is interesting to note that the announcements of plans to roll-out
electronic tagging have also stirred up concerns that personal privacy may be eroded. RFID
tagging opens up the possibility for item level identication, and that means products that we
buy and carry with us, which contain RFID, can uniquely identify us, and further more this
can be done covertly at a distance. The resulting privacy implications are discussed in a later
section.
P1: JYS
INTRODUCTION 5
0
5000000
10000000
15000000
20000000
25000000
30000000
35000000
40000000
2005 2006 2007 2008 2009 2010
Entertainment
Authenticity
Payment
Access
Tracking
RFD shipments by use and year
Year
V
o
l
u
m
e

i
n

1
0
0
0
s
FIGURE1.5: Forecast volume RFID tags shipped 20052010 (1000s). Source: Instat 12/05
To provide an overall picture of how RFID is expected to be adopted in the coming years
see Fig. 1.5, and In-stats forecast for RFID adoption [8].
The expected growth curve is the classic hockey-stick shape with tag sales estimated at
1.3 billion in 2005, and expected to rise to 33 billion by 2010, a 2500% increase over in 5 years.
This is a phenomenal growth by any standard. The 2010 forecast expects tag applications to
be dominated by tracking at 94%, which can be broken down as supply chain (83%) and food
(11%). Access control (3%) is the next highest volume use and all other uses are expected to total
3%. From a nancial perspective, revenues are expected to increase from $683M to $2480M in
the same time frame.
1.1 LECTUREOVERVIEW
In this section we have provided an overviewof the broad topic of RFID. The following sections
examine the technology and its implications in greater depth, and are organized as follows:
Section2 describes the principles of RFIDoperation, including the various types of tagging
system. We consider the physics behind the design of RFID and the physical constraints that
limit how tags can be used.
Section 3 looks at the most inuential standards that are shaping the adoption of electronic
tagging technologies across the industry.
Section 4 examines the issues associated with reading multiple collocated tags and the
algorithms that can be used to resolve them.
P1: JYS
6 RFIDEXPLAINED
Section 5 provides a more in-depth overview of commercial applications by classifying
them and providing detailed examples in each category.
Section 6 describes the capabilities of tags that include sensing, some of the existing
products in this space and how these devices will be used in the future.
Section 7 presents deployments of the latest RFID systems and summarizes the key expe-
riences found along the way.
Section 8 considers the hot topic of privacy, and attempts to provide a broad consideration
of how RFID might be used by commercial, government, and criminal groups, and how we can
benet from the advantages while mitigating the disadvantages of the technology.
Section 9 introduces the opportunity for RFID tags that contain read/write memory, their
memory capacity and the scope of applications that can exploit them.
Section 10 summarizes the challenges facing RFID, from several perspectives: technology,
manufacturing cost, deployment, and social acceptance; and concludes by looking at the future
of electronic tagging.
P1: JYS
7
C H A P T E R 2
Principles of Radio Frequency
Identication
There are many types of RFID [9], and at the highest level of classication these can be divided
into two classes: active and passive devices. Active tags require a power source [10] and either
need a connection to powered infrastructure or have a limited lifetime dened by the energy
stored in an integrated battery, balanced against the number of read operations that will be
performed on the tag. Examples of active tags are, transponders attached to aircraft to identify
their national origin, and LoJack devices attached to cars that incorporate cellular technology
along with a Global Positioning System (GPS), communicating the location of a car if stolen.
Olivetti Research Ltds Active Badge, used to determine the location of people and objects in a
building is an example of a small wearable active tag with a lifetime of about 1 year [11]. There
are also some types of active tag that scavenge power from their enviornment. MIT Media
labs push-button powered doorbell controller [12] is another; the mechanical energy scavenged
from pushing the switch is used to power the electronics.
However, it is the passive RFID tag that is of interest to retailers, requiring no mainte-
nance and exhibiting an indenite operational life [1316]. They have no battery, and can be
made small enough to be incorporated into a practical adhesive label. A passive tag consists
of three parts: an antenna, a semiconductor chip attached to the antenna, and some form of
encapsulation, which could be a small glass vial or a laminar plastic substrate with adhesive
on one side to enable easy attachment to goods, see Fig. 2.1. The encapsulation is necessary
to maintain the integrity of the tag and protect the antenna and the chip from environmental
conditions or reagents that would cause damage.
The purpose of the tag antenna is to receive power from the reader, and shortly after to
transmit its ID in response. The tag chip is powered by the energy in the signal received at the
tag antenna, which activates an electronic circuit and encodes an ID onto the return signal that,
in turn, is communicated back to the reader by the antenna. In the history of RFIDdesign there
have been two fundamentally different design approaches for delivering power from reader to
tag: magnetic induction and electromagnetic wave capture. These two designs take advantage
of the electromagnetic properties associated with an RF antenna; the near eld and the far eld.
P1: JYS
8 RFIDEXPLAINED
FIGURE2.1: Logical components of anRFIDtag. Note that the antenna cantake many forms including
a coil and a dipole depending on the tag type
If an alternating current is passed through a coil it will create an alternating magnetic eld in
the locality of the coil, and this is referred to as the near eld. This circuit will also give rise
to propagating electromagnetic waves that breakaway from the coil/antenna and radiate into
space, this is termed the far eld and is the principle of radio transmission. Both radio properties
can be used to transfer enough power to a remote tag to sustain its operation, typically between
10 W and 1 mW depending on the tag type, and through modulation can also transmit and
receive data [4]. To showhowsmall these power budgets are, by comparison, the nominal power
consumed by an Intel XScale processor is approximately 500 mW and an Intel Pentium-4 is
50 W.
2.1 NEAR-FIELD-BASEDRFIDDESIGN
The use of near-eld coupling between reader and tag can be described in terms of Faradays
principle of magnetic induction. A reader passes a large alternating current through a reading
coil, resulting in an alternating magnetic eld in its locality. If a tag that incorporates a smaller
coil (Fig. 2.2) is placed in this eld, an alternating voltage will appear across it, and if rectied
and coupled to a capacitor, a reservoir of charge will accumulate that can be used to power a
tag chip. Tags that use near-eld coupling send data back to the reader using load modulation.
Since any current drawn from the tag coil will give rise to its own small magnetic eld which
will oppose the readers eld, this can be detected at the reader coil as a small increase in current
owing through it. This current is proportional to the load applied to the tags coil (hence
load modulation), and is the same principle used in power transformers found in most homes
todayalthough usually the primary and secondary coil of a transformer are wound closely
together to ensure efcient power transfer. Thus, if the tags electronics applies a load to its own
antenna coil and varies it over time, a signal can be created that encodes the tags ID, and the
P1: JYS
PRINCIPLES OF RADIOFREQUENCY IDENTIFICATION 9
Using induction for power coupling from reader to tag
and load modulation to transfer data from tag to reader
Magnetic
field affected
by tag data
Data via
changes in
field strength
Power
Data
Glass or plastic
encapsulation
Note: Relative size of tag
in comparison to reader is
exaggerated for clarity
Far-field region
Near-field region
Coil
Propagating EM waves Alternating magnetic field in the near field region
Binary Tag-ID
C/2f
(If tag supports write)
RFID tag
RFID
reader
FIGURE 2.2: Near-eld power/communication mechanism for RFID tags operating at less than
100 MHz
reader can recover this signal by monitoring the change in current through the reader coil. A
variety of modulation encodings are possible depending on the number of bits of ID required,
the rate of data transfer, and additional redundancy bits placed in the code to remove errors
resulting from noise in the communication channel.
Near-eld coupling is the most straight forward approach for implementing a passive
RFID system and as a result it was the rst, and has led to many standards such as ISO
15693 and 14443 (see Section 3), and a variety of proprietary solutions. However, near-eld
communication has some physical limitations. It turns out that the range for which it is possible
to use magnetic induction approximates to c /2 f . Thus, as the frequency ( f ) of operation
increases, the distance that near-eld coupling can operate over decreases (c being a constant,
the speed of light). A further limitation is the energy available for induction as a function of
distance from the reader coil. It can be shown (Section 2.2 below) that the magnetic eld drops
off at a 1/x
3
factor along a center line perpendicular to the plane of the coil. As applications
require more ID bits, and have the requirement to discriminate between multiple tags in the
same locality during a xed read time, it is necessary to increase the data rate used by the tag
and thus the operating frequency. These design pressures have led to new passive RFID designs
based on far-eld communication (Fig. 2.3).
P1: JYS
10 RFIDEXPLAINED
(a)
(b)
FIGURE 2.3: RFID tags based on near-eld coupling. (a) Trovan tag (128 kHz), size: 1 cm [17].
(b) Tiris (13.56 MHz), size: 5 cm 5 cm [18]
2.2 PROPERTIES OF MAGNETICFIELDS
Since RFID readers and near-eld tags couple to each other using Faradays principle of in-
duction [19], we can analyze the basic interaction between RFID reader and tag by considering
the two devices as two circular coils parallel to each other and aligned along a perpendicular
line that runs through their centers. This is a similar conguration to the more familiar AC
power transformer. If current is passed through the reader coil, the magnetic eld H (in Webers)
will be dened at its center by (I current in Amperes, N number of turns, r radiues of coil):
H = I N/2r
And the magnetic ux density B (for free space) is given by:
B =
0
H
The resulting characteristic magnetic eld pattern around the coil is shown below in
Fig. 2.4. As you can see in the center of the coil the magnetic eld is perpendicular to the plane
of the coil and extends outward along that axis becoming weaker at a distance. In comparison,
Current l
N turns
Radius r
Magnetic field strength (H)
H = I N / 2 r
Magnetic flux density (B)
B =
0
I N / 2 r
FIGURE2.4: Magnetic eld calculation at the center of a coil
P1: JYS
FIGURE2.5: Magnetic ux density (B) at a distance x from the center of an N-turn coil, with radius
r , and current owing, I
a power transformer has its primary and secondary coil tightly coupled and therefore the eld
strength at a distance is not normally of interest.
However, in the case of an RFIDtag, it is the projection of the readers magnetic eld that
allows a reader to interact with a tag at a distance even with optically opaque materials in the
path. To understand how the eld strength varies along this line we can consider the equations
in Fig. 2.5.
Since the distance x from the center of the coil will be large compared to the radius of
the coil, it will dominate the magnitude of B, which can be approximated to 1/x
3
, which is a
rapidly decreasing function and one of the main reasons why near-eld coupling, using practical
eld strengths, can typically only be used to read tags at up to 1 m away, see Fig. 2.5.
2.3 FAR-FIELD-BASEDRFIDDESIGN
RFID tags based on far-eld coupling (Fig. 2.6) capture electromagnetic waves propagating
from a dipole antenna attached to the reader. A smaller dipole antenna in the tag will receive
(a) (b)
FIGURE 2.6: RFID tags based on far-eld coupling. (a) Alien (900 MHz), size: 16 cm 1 cm.
(b) Alien (2.45 GHz), size: 8 cm 5 cm
P1: JYS
12 RFIDEXPLAINED
Using electromagnetic (EM) wave capture to transfer power from reader to tag
and EM backscatter to transfer data from tag to reader
Data modulated on
signal reflected
by tag
Power
Antenna
dipole
Propagating
electromagnetic waves
(Typ. UHF)
Far- field region
Near- field region
Binary Tag-ID
Glas or plastic
encapsulation
Note: Relative size of tag
in comparison to reader is
exaggerated for clarity
Data (If tag supports write)
RFID
reader
RFID tag
FIGURE 2.7: Far-eld power/communication mechanism for RFID tags operating at greater than
100 MHz
this energy as an alternating potential difference that appears across the components of the
dipole. This signal can also be rectied and used to accumulate energy in a capacitor reservoir to
power its electronics. However, unlike the inductive designs, these tags will be beyond the range
of the readers near eld, and information cannot be transmitted back to the reader using load
modulation. The technique used by commercial far-eld RFID tag designs is back-scattering
(Fig. 2.7). If an antenna is designed with precise dimensions, it can be tuned to a particular
frequency band and absorb most of the energy that reaches it in that band. However, if there
is an impedance mismatch at this frequency, some of this energy will be reected back as tiny
waves from the antenna toward the reader, where it can be detected using a sensitive radio
receiver. By changing the antennas impedance over time, the tag can reect back more or less of
the incoming signal in a pattern that encodes the ID of the tag. In practice the antenna can be
detuned for this purpose simply by placing a transistor across the dipole and turning it partially
on and off. As a rough design guide the tags that use far-eld principles operate at greater than
100 MHz typically in the UHF band (e.g., 2.45 GHz); below this frequency is the domain of
RFID based on near-eld coupling.
2.4 PROPERTIES OF BACKSCATTERRF SYSTEMS
The range of a far-eld system is limited by the amount of energy that reaches the tag from
the reader, and the sensitivity of the readers radio receiver to the reected signal. The actual
P1: JYS
Transmitted power density at distance r in far-field
S = P
TX
.G
TX
/4.X
2
Reflected power back to reader from a tag
P
RX
= P
TX
.G
TX
. G
RX
/(4)
4
.X
4
Doubling separation of tag and reader requires x16 more
transmit power to maintain the power level in the response
Tag
Reader
Backscatter
Reader signal

KKK KEEE EYYY Y
P
TX
= Transmit power
P
RX
= Receive power
G
TX
= TX Antenna Gain
G
RX
= RX Antenna Gain
= Wavelength
x = Separation distance
P
RX
signal
S
P
TX
radius x
FIGURE2.8: Backscatter signal strength at a distance x
return signal is very small because it is the result of two inverse square laws, the rst as EM
waves radiate from the reader to the tag, the second when reected waves travel back from the
tag to the reader (see Fig. 2.8). Thus, the returning energy is proportional to 1/x
4
where x is
the separation of the tag and reader.
As can be seen from the power density equations above, the absolute receive power is
also proportional to the product of the transmitter and receivers gain (G), and the wavelength
of the carrier signal. Fortunately, thanks to Moores Law [20], and the shrinking feature size
of semiconductor manufacturing, the energy required to power a tag at a given frequency of
operation continues to decrease (as lowas a fewmicrowatts). Moores Lawis more usually used to
explain the increasing speed of computers for each new generation of semiconductor. However,
if a CMOS transistor is built with reduced dimensions, at a given switching frequency it will
consume less power. Thus, RFIDtags build fromCMOS transistors can operate at lower power
when manufactured using smaller lithographies, and this has the consequence of increasing the
operational range. In step with this trend, the sensitivity of inexpensive radio receivers has
also been improving, and can now detect signals with power levels on the order of 100 dBm
in the 2.4 GHz band. A typical far-eld reader can successfully interrogate tags 3 m away,
and some RFID companies claim their products have read ranges of up to 6 m. This is the
result of many factors that include improved component tolerances, better antenna design,
low-noise transistors, improved tag signal coding along with signal processing at the receiver
to decode data on the return signal. Modern coding techniques support this trend by allowing
P1: JYS
14 RFIDEXPLAINED
more bits to be coded per cycle of the carrier. Moores law plays an indirect role in this broad
evolution by enabling inexpensive high-performance processors in the reader to run complex
signal-processing algorithms in real time. Putting all of these factors together, RFID tag and
reader designs can now be built that are more effective than ever before.
The work of EPCglobal Inc. [13], (originally the MIT Auto-ID Center, a nonprot
organization set up by the MIT Media Lab., and later divided into Auto-ID laboratories, still
part of MIT, and EPCglobal Inc., a commercial company), was key to promoting the design
of UHF tags which has been the basis of the RFID trials by Walmart and Tesco. Although an
extensible range of tags has been dened by this group, it is the Class-1 Generation-1 96-bit
tag that has been the focus of recent attention. It has the exibility for an EPC manufacturer to
create over 12 10
17
codes, making it possible to uniquely label every manufactured item for
the foreseeable future, and not just using generic product codes. While this is not necessary for
basic inventory control, it does have implications for tracing manufacturing faults, stolen goods,
detecting forgery, and for the more controversial postsale marketing opportunities, enabling
directed-marketing based on prior purchases (Section 8).
2.5 MODULATIONTECHNIQUES USEDWITHRFID
RFID tags that are designed to use backscatter have limited options for modulating data sent
back to the reader. Amplitude Shift Keying (ASK) is the most basic and easiest to implement,
but like all amplitude modulation techniques, this approach is prone to the affects of channel
noise. When load modulation is used to transmit information, there are a greater variety of
modulation options. Phase Shift Keying (PSK) is more robust than ASK, and in some designs
FIGURE2.9: Modulation coding options for RFID signaling
P1: JYS
FIGURE2.10: Comparing the properties of RFID operating in different frequency bands
a Binary PSK approach has been used successfully. Figure 2.9 shows how a tag transforms the
received reader signal in order to send data back to the reader.
2.6 COMPARISONOF THEPROPERTIES OF RFID
BASEDONFREQUENCY
Given the highly diverse properties of RFIDtags described in this section it is useful to summa-
rize by comparing the typical characteristics of the tags based on frequency. Figure 2.10 groups
tags into Low Frequency (LF), High Frequency (HF), and Ultra High Frequency (UHF)
both 900 MHz and 2.45 GHzand catalogs their typical characteristics. However, it should
be emphasized that within each category the properties can vary considerably by manufacturer
and application. For example, the antenna size used in the reader and the tag design, respec-
tively, will have considerable bearing on the read range. For this reason it is unwise to project
some trends from the table, for instance it appears that larger antennas are required for the
high-frequency tags. However, this is not the case as the larger antenna is needed to achieve the
long read-range that is possible with backscatter modulation. A UHF antenna could be made
very small if it were only expected to be read from a few centimeters away. But as a general
guide to the practical implementations of RFID technologies, the table serves its purpose.
P1: JYS
16
P1: JYS
17
C H A P T E R 3
RFIDIndustry Standards
The International Standards Organization (ISO) and European Telecommunication Standards
Institute (ETSI) have been set up to establish industry-wide standards across many disciplines.
RFID can directly benet from standardization to ensure widespread interoperability, and in-
dustry wide adoption has already been enabled by standards in common use; examples include
IS015693 and IS014443.
There have been many RFID standards created over the years, each designed to solve
a particular set of application requirements. The technology available to implement various
standards has also improved over time and hence older standards dene capabilities that are no
longer state-of-the-art. Furthermore, the standards are very different from each other and the
tags used by one standard are in no way compatible with another due to differences in operating
frequency, power harvesting techniques, modulation, and data coding. It is beyond the scope
of this lecture to provide a broad description of all the standards, but instead we focus on a few
that are currently dominating the industry.
3.1 EPCGLOBAL
EPCglobal is a consortium that created a new de facto standard for UHF-based RFID tags.
It was originally a grass roots initiative created by MITs Media Lab. However, due to the
success of their Generation-1 tags, ISO are in the process of working with EPCglobal to create
a joint Generation-2 standard [21] which contains modications to enable it to be adopted
on a global scale. One of the key additions was the use of a bit in the EPC tag header that
differentiates it from the ISO Application Family Identier, thus enabling RFID readers to
distinguish them. At this time it is expected the new ISO 18000-6 standard, which covers the
EPCglobal specication will be available by the Fall of 2006.
3.1.1 Generation-1
Within this standard there were two tag classes dened: Class 0 and Class 1.
Class 0 : is a read-only identity tag that is programmed during the manufacturing process.
Class 1: is a write-once read many (worm) tag that may be eld programmed.
P1: JYS
18 RFIDEXPLAINED
Despite the success of Generation-1 this standard had several limitations. The two tag
class denitions did not interoperate with each other and used different wireless protocols. Both
classes could coexist, but required two different reader implementations to interrogate collocated
tags. This was compounded as RFID manufactures created their own proprietary extensions,
such as Matrix (now Symbol) with their class 0++product. Furthermore, the standard was not
ratied as a world standard, and there were many countries that could not use the tag products.
In a world of interconnected economies, manufacturers are not likely to adopt such a standard
unless they are able to make use of it to sell and ship products within a global economy.
3.1.2 Generation-2
The primary goal for the Generation-2 standard was to create a global standard that would
mitigate many of the issues that limited the success of Generation-1. The RF specication is
nowmore exible and can be used across national boundaries operating in the region of 860960
MHz and there is a broad support from the majority of technology providers. Furthermore,
robustness and read throughput for co-located high-density tag environments has increased.
Also to address various privacy concerns the standard nowhas greater emphasis on secure access
control [22]. Fig. 3.2 Shows an early implementation of a Generation-2 tag by Texas Instru-
ments. Impinj Inc. is another important player driving the Generation-2 spec, providing silicon
designs ready to be manufactured by the various tag vendors. Four classes of tags are dened that
progressively build on the properties of the lower classes. The class properties are listed below:
Class 1 Passive Tags (backscatter)
write-once read-many to establish a standard EPC identity;
tag identier (TID)information about the manufacturer of the tag (read-only);
password protected access control;
kill switchto disable the tag at POS;
user memory (optional).
Class 2 Passive Tags (backscatter) Extended Functionality
re-writable memory;
extended TID;
extended user memory;
authenticated access control;
additional featureswork in progress.
P1: JYS
RFIDINDUSTRY STANDARDS 19
FIGURE3.1: Format of a 96-bit EPCglobal tag
Class 3 SemiPassive Tags
an integral power source to supplement captured energy;
integrated sensing circuitry.
Class-4 Active Tags
tag-to-tag communications;
complex protocols;
ad-hoc networking.
3.1.3 EPCPacket Formats
The EPC tags were dened with the following four elds: header, EPC manager, object class,
and serial number (Fig. 3.1).
FIGURE3.2: The Texas Instruments Generation-2 tag (courtesy of Texas Instruments)
P1: JYS
20 RFIDEXPLAINED
Header is an 8-bit eld allowing for expansion of the EPC tag format including 64-, 96-, and
256-bit versions. The most popular of these standards is the 96-bit Universal Identier
format. The remaining list describes the elds and sizes for the EPC-96 format.
EPC Manager is a 28-bit eld that denes the domain manager for the remaining elds.
Object Class describes the generic type of the product tagged and is 24 bits wide.
Serial Number an individual itemnumber that has 36 bits available to it. The large number space
provides the rst opportunity in the industry for item-level tagging.
Removing the 8-bit header and 28-bit manager bits, there are 60 bits remaining allowing
approximately 12 10
17
items to be tagged cataloged by each domain manager.
3.2 ISO15693 VICINITY CARDS ANDRFID
ISO 15693 is a standard for both vicinity cards and RFID. Devices can typically operate at
distances of 11.5 m and use inductive coupling to provide power and load-modulation to
transmit data. The standard operates in the 13.56-MHz band, and a typical tag must be able
to operate with a magnetic eld strength between 0.15 and 5 A/m. For comparison 5 A/m is
about a tenth of the earths magnetic eld strength at the surface.
Tags based on this standard have been widely produced and used for a variety of applica-
tions. See Section 5 that describes sports and security applications based on this standard; and
in particular products made by Texas Instruments (Tiris) that support these markets.
A full specication of this standard (four parts) can be obtained from ISO, see http://
www.iso.org/iso/en/prods-services/popstds/identicationcards.html.
3.3 ISO14443 PROXIMITY CARDS ANDRFID
Similar to the ISO156803 standard, 14443 was created for proximity cards that operate at short
distances. A typical application is fare collection on public transport, typically at a turn-style
requiring a passenger to place a card near to a reader in order to make a payment. As a result the
standard is more complex than ISO15693, providing more than a simple identity and supports
two-way data exchanges. These cards are dened to work with a magnetic eld strength of
1.57.5 A/m, and thus for a similar reader are designed to be used closer to the reader coil than
with ISO15693.
A full specication of this standard can be obtained from ISO, see: http://www.iso.
org/iso/en/prods-services/popstds/identicationcards.html
3.4 THENFCFORUM
An important recent development opens up new possibilities for more widespread applications
of RFID. Since 2002, Philips has pioneered an open standard through ECMA International,
resulting inthe Near-FieldCommunicationForum[23] that sets out to integrate active signaling
P1: JYS
RFIDINDUSTRY STANDARDS 21
FIGURE3.3: The Nokia 3200 cell phone features an NFC reader: Front sideIt looks like an ordinary
cell phone. Back sideyou can see the reader coil molded into the housing (courtesy of Nokia, Inc)
between mobile devices using near-eld coupling, and be compatible with existing passive RFID
products. The new standard aims to provide a mechanism by which wireless mobile devices can
communicate with peer devices in the immediate locality (up to 20 cm), rather than rely on
the discovery mechanisms of popular short-range radio standards, such as Bluetooth [24] and
WiFi [25], whichhave unpredictable propagationcharacteristics andmay formassociations with
devices that are not local at all. NFCaims to streamline the process of discovery by passing MAC
address and channel encryption keys between radios through an NFCside-channel, which when
limited to 20 cm allows a user to enforce their own physical security. NFC has been deliberately
designed to be compatible with ISO15693 RFID tags operating in the 13.56 MHz spectrum,
and allow mobile devices to read this already popular tag standard. It is further compatible with
the FeliCa and Mifare smart card standards that are already widely used in Japan.
In 2004, Nokia announced the 3200 GSM cell phone (Fig. 3.3) that would incorporate
an NFC reader. Although the company has not published an extensive list of the potential
applications, it can be used to make electronic payments (similar to a Smart Card) and place
P1: JYS
22 RFIDEXPLAINED
phone calls based on the RFIDtags that it encounters. For example, at a taxi-stand a prospective
client might bring their phone near to an RFID tag attached to a sign at the front of the waiting
area. The result would be a phone call to the taxi companys coordinator and a request for a car
to be sent to that location [26]. This model allows a close link between the virtual components
of our computer infrastructure and the physical world, such as signs and taxis, and is a key
enabling technology that contributes to the implementation of the Ubiquitous and Pervasive
Computing vision as proposed by Mark Weiser [27].
A complication for the wide-scale adoption of the NFC standard is that state-of-the-
art EPCglobal RFID tags are based on far-eld communication techniques, working at UHF
frequencies. Unfortunately, NFC and EPCglobal standards are fundamentally incompatible.
P1: JYS
23
C H A P T E R 4
Reading Collocated RFIDTags
One of the ultimate commercial objectives of RFID systems is the ability to read, and charge
for, all of the tagged goods in a standard supermarket shopping cart by simply pushing the cart
through an instrumented aisle. Such a system would speed the progress of customers through
checkout areas and reduce operational costs. The solution to this problem can be thought of as
the holy grail of RFID technology. It has many engineering issues that make it difcult. First,
the RF environment inside a shopping cart is particularly challenging. The product packaging
in the cart is made of a wide variety of materials that include metal cans and aluminized plastics
that reect and shield the interrogation signals. Furthermore, some of the products contain
water, and plastics, that may absorb RF signals in the microwave band. To complicate matters
further, all of the products are in close proximity to each other and in random congurations.
RFID tags attached to these products will sometimes be poorly orientated with respect to the
readers antenna, thereby making RF communication difcult. In addition, tag antennas are
typically at to enable them to be embedded in labels, but if orientated edge-on to the reader
the tag will likely not receive enough energy to power up. These specic problems are discussed
in more detail in Section 10; however, even if the RF reading environment for a group of RFID
tags is ideal, it is still an engineering challenge to design readers that can successfully query
multiple collocated tags, and accurately determine all of their IDs in a short period of time.
Consider two tags that are situated next to each other and equidistant fromthe reader. On
hearing the readers signal they will both acquire enough energy to turn on and then transmit
their response back to the reader at roughly the same instance in time. The result will be a colli-
sion between the two signals, and the data from both tags will be superimposed and garbled as a
result. Collisions can be detected at the readers receiver by augmenting its demodulation circuit
to look for signal encodings that contain an anomolous format. For multinode communications
networks, such as Ethernet, this is a well-understood problem addressed in protocols such as
CSMA/CD [28], or 802.11 that uses a variant of MACA (Multiple Access/Collision Avoid-
ance) [29]. The solutions employed are based on arbitration protocols providing the colliding
nodes with a new opportunity to successfully deliver their data while minimising the resulting
wasted channel bandwidth.
P1: JYS
24 RFIDEXPLAINED
Arbitration using a statistical approach has been implemented in some RFID systems by
inserting a random delay between the start of the interrogation signal and the response from
the tag. But even if each tag randomizes its response time there is still a nite probability that
a collision will occur, and the reader must carry out several rounds of interrogation until all of
the tags in that area have been heard with high probability. This algorithm can be enhanced
further by using a protocol that prevents tags that have already been heard from responding
again until the current interrogation cycle has ended. At each interrogation request there will
now be a progressively smaller population of tags that will respond, reducing the likelihood the
remaining tag responses will collide.
Unlike a general purpose wireless communication network in which an ad-hoc collection
of nodes have equal status, an RFIDreader has a privileged position and can centrally orchestrate
an arbitration protocol. This allows for a more deterministic arbitration protocol to be used.
The process of uniquely determining a tags ID from the surrounding population of tags is
sometimes called singulation.
4.1 QUERY TREEPROTOCOL
EPCglobal Generation-1 class-0 tags use a Query Tree Protocol to singulate tags. Figure 4.1
shows an example of how three unique, but collocated tags (001, 100, and 110), can be success-
fully read using this protocol. The reader R starts an interrogation (level-0) by asking which
FIGURE4.1: Arbitration mechanism used by EPCglobal Generation-1 Class-0
P1: JYS
READINGCOLLOCATEDRFIDTAGS 25
of the top branches of the tag identity-space (modeled as a binary tree) contain tags. It does
this by broadcasting a prex that initially selects the left branch in the tree; in our example the
bit-mask would be [0] selecting all tags with codes 0XX, where X is either a 0 or 1. A mask
operation is achieved by only considering the number of bits in the prex and logically XORing
the mask value with the corresponding top-bits in the Tags ID, and then looking for a zero
response to determine a match. If a tag matches the condition it will respond with the value of
the bit that follows the prex (either 0 or 1).
If only one tag exists in the left sub-tree, as in our example 001, its next bit 0 will be
returned to the reader R, and it can make an additional query with prex [00-] to nd the nal
bit of the sequence, thus singulating tag 001.
If multiple tags are present matching this condition, as in the right branch example
detected by prex [1], they will all respond and a collision will result. A collision situation is
recorded at the reader as aberrations in the received waveform. As a result the reader R will
respond by separately querying each of the sub-trees below this point in the tree by separately
transmitting the query prex [10] and [11]. Each will result in a tag response without a
collision, and thus singulate the ID for 100 and 110.
Although the process above may seem complicated, the same steps are applied repeatedly
at each of the sub-trees in depth-rst-search order, and by using recursion to implement the
Query Tree Algorithm; the process can be dened concisely, even for an arbitrary number of
bits in the tags. An important aspect of the algorithm is that when there is no response from
one of the sub-trees, it is removed from the tag search-space. Thus, the queries need only be
applied to parts of the tree that contain tags, and after a short time all tags present will be able
to respond to the reader in depth-rst-search order. The cost of the algorithm is bounded by
the number of bits in a tag ID n times the number of tags being read. In practice, tags are likely
to be allocated to organizations in sequential batches that will tend to localize the tag identities
in particular sub-branches of the code space. This will accelerate the search process, because
the amount of recursion needed to complete the algorithm will be reduced.
Manufacturers of EPCglobal Generation-1 based their arbitration mechanism on this al-
gorithm, and claim it is possible to accurately read up to 500 collocated tags per second. One of
the advantages of this arbitration algorithmis that it does not require any state to be held in a tag
itself, instead a reader has the responsibility of probing the tag identity space, managing the re-
cursive queries, and keeping track of the branches of the tree that contain tags. However, this ap-
proach does raise a privacy concern. As the reader homes in on the tags it identies, it broadcasts
their IDs using the full power of the readers transmitter. This means that a distant eavesdropper
with a suitable receiver may be able to record the IDs being scanned. And more of a concern,
the products being purchased can also be identied. Many of the undesirable consequences of
unwittingly disseminating this type of information are discussed in Section 8 on Privacy.
P1: JYS
26 RFIDEXPLAINED
In order to overcome this limitation, the EPC Generation-2 arbitration protocol was
designed to avoid the reader transmitting the tag IDs at high power. The essence of this
algorithm is described in Section 4.2 below.
4.2 QUERY SLOTPROTOCOL
Query Slot Protocol is an arbitration mechanism used as the basis for singulation in the EPC-
global Generation-2 tags. Unlike the Query Tree Protocol Algorithm described in 4.1, it has an
advantage that the reader does not need to transmit the IDs of the tags in order to determine the
inventory. In this algorithm only the tags backscatter their ID thus limiting the range at which
the IDs can be detected as radio signals. However, it requires additional state registers available
in each tag, and increases the complexity of the design. But yearly progress in the capabilities
of CMOS integration, as predicted by Moores law, allow for considerably more transistors to
be fabricated in the same area of silicon. Today, the additional complexity of adding a few data
registers, and the associated protocol state-machine, does not add a signicant cost burden to
the manufacture of an RFID tag.
An example of the Query Slot Protocol is shown in Fig. 4.2. In the specication there are
other aspects to the protocol, but this simplied description helps us understand the core mech-
anism. To make a comparison with the Query Tree Protocol shown in Fig. 4.1, the same three
tags with IDs 001, 100, and 110, are used as an example collocated tag inventory that must be
successfully read.
The algorithm requires tags to provide the following capabilities. Each tag contains a
counter count, initially set to zero; an inventoried ag; initially cleared, and a random number
generator that can produces 16-bit values rn16.
The reader R transmits a QueryRequest command to the tags with a parameter Q. All
tags that hear the command start an inventory round and clear their inventory ag, they also
enter an arbitrate state. They then generate a Q-bit random (0 to 2
Q
1) and load this into
their slot counter count. If a tags count is zero, as it the case of tag 001 in the example, it will be
in a reply state and will generate a 16-bit random number rn16 and backscatter it to the reader.
The reader will then respond with an acknowledgement including the rn16, and if a match
is found, will backscatter its EPC code, which can be recorded by the reader and stored in an
inventory database. This tag will now set its inventoried ag and go to sleep until it hears a new
QueryRequest.
In order to nd all of the tags in the inventory, the reader will now initiate a QueryRep
(repeat) command which will cause all remaining tags to decrement their slot counter, count =
count-1 and once again, any tag with a value of zero will respond with an rn16 value and the
P1: JYS
READINGCOLLOCATEDRFIDTAGS 27
FIGURE4.2: Example of the Query Slot Protocol used in EPCglobal Generation-2
process will continue as before. If by chance each tag had found a unique slot, after 2
Q
1
QueryReps had been issued, all of the tags would have been inventoried.
However, in our example there are two tags that have the same count value, 100 and 110.
When their count is zero both tags will be in the reply state, and respond at the same time
with different rn16 numbers and a collision will result. This is detected by aberrations in the
signals detected at the reader and it will instead respond with a nack and a QueryAdjust with a
new Q-parameter without changing the state of any of the inventoried ags in the tags. The
remaining tags will again randomize their slot counter value count with a 2
Q
1 number and
the process will continue until all tags have been singulated.
An interesting aspect of this protocol is that the optimal value of Q is dependent on the
number of tags being inventoried. Ideally Q should be chosen so that there will be as many
slots as there are tags. If Q is too large there will be too many slots that are empty for each
QueryRep, and if too small there will be too many collisions. However, the number of tags
is initially not known, and therefore the value of Q must be found experimentally by testing
the environment and increasing or decreasing Q until an acceptable arbitration behavior is
found.
P1: JYS
28 RFIDEXPLAINED
4.3 SUMMARY TAGREADRATETIMINGFOR
EPCGENERATION1 AND2
To provide some context for the tag read rates used in the algorithms described above, it is
useful to consider timings used by the EPCglobal Standard. In the EPC Generation-1 tag the
reader-to-tag command timing is approximately 1 ms, and the tag reply 0.9 ms. The total round
trip of 2 ms leads to the claim of a maximum read rate of 500 tags-per-second. However, a
Query Tree Protocol as described in 4.1 will reduce the tag read-rate as the tree walk procedure
takes time to singulate each tag and this will depend on the distribution of the IDs. The
binary tree algorithm minimizes the overhead by a logarithmic factor rapidly descending into
the parts of the tree populated by tags. Comparing the Generation-1 standard with the future
Generation-2 standard, Generation-1 systems read tag-data at 140 kbps, whereas Generation-2
has an adaptive scheme from 40 to 640 kbps. Generation-2 EPC-96 bit IDs can potentially be
read over four times the speed of Generation-1 tags. The Generation-2 Query Slot Algorithm
described in 4.2 can thus have arbitration slot times as small as 0.5 mS.
P1: JYS
29
C H A P T E R 5
Applications of RFIDTagging
In this section we examine the wide variety of applications that can take advantage of RFID.
The capability of identication at a distance can be extended to include sensing applications and
read/write memory, however, these topics will be covered later in Sections 6 and 9 respectively
as they warrant special discussion. Below we provide a categorization of mainstream RFID
applications:
Security:
access control: keys and immobilizers;
patrol verication: process management;
antitheft: merchandize.
Tracking:
supply chain: warehousing and inventory control;
people and animals: personel, children, patients, runners, cattle, and pets;
assets: airline luggage, equipment, and cargo.

Authenticity:
money: banknotes;
pharmaceuticals: packaged drugs;

Electronic payments:
transportation: auto-tolls: FasTrak, EZ-pass;
ticketing: ski passes;
credit/debit cards: PayPass by MasterCard.

Entertainment:
smart toys: interactive characters.

5.1 SECURITY
It is possible to divide RFID-based security applications into three subcategories: Access Con-
trol, Verication, and Antitheft.
P1: JYS
30 RFIDEXPLAINED
5.1.1 Access Control
One of the rst applications that motivated the design of modern passive RFID was Access
Control, enabling mechanical keys to be replaced by an electronic card. The primary advantage is
that card keys are harder to forge and it is much easier to revoke a key that has been compromised
or lost, simply by deleting it from the access database, or to issue a security alert if a revoked
key is used nefariously.
This kind of RF tag has been improved since its early beginnings in the 1960s and is now
widely adopted as the basis for corporate identity badges by numerous organizations around the
world. In addition to verication of an employees identity, they often serve as proximity cards
providing access to corporate campuses, buildings, and laboratories; and reduce the need for
security guards at all of the entrances. The read range is usually limited to a foot or less to avoid
unintentionally opening a door, but still has the advantage over a key that the badge can be left
in a bag or a wallet, and provide convenient access for the owner without needing to physically
remove it.
In comparison to inventory control, RFID-based access control is not a cost-sensitive
application. ID cards have a long life and may incorporate other prerequisite features necessary
even without RFID, such as a photograph and a robust laminated plastic form factor. One of
the manufacturers of these cards is Hughes Identication Devices (HID) that provide a wide
range of RFID solutions. Their contactless ID cards operate either at 125 kHz or 13.56 MHz
(depending on local spectrumlegislation) and can also store between 2 and 16 kbits of read/write
data (Fig. 5.1).
Another, now common, form of RFID key is used in automobiles to make it harder for
vehicular theft. An example is the Chrysler Jeep that incorporates an RFID tag into the body
of the ignition key. The lack of the correct RFID tag serves as an immobilizer. The car will only
start if both the unique mechanical key and the unique RFID tag are present. Thus, even if a
would-be thief were able to make an impression of the key and covertly reproduce it, the forgery
would be of no use without the embedded RFID tag. And similarly a smart RFID reader that
can determine the tags identity and later masquerade as the tag, would also fall short as a key.
The requirement that both a unique RF interaction and a physical key are present considerably
raises the technical skill necessary for a successful forgery (Fig. 5.2).
5.1.2 Verication
Most companies and government institutions employ security personnel to guard their en-
trances, and make periodic checks that their campuses are secure. Despite modern electronic
measures to help with this task, such as the use of security cameras, the most versatile line-of-
defense is to employ a security patrol to periodically make the rounds and check for suspicious
activity. Employers may nd themselves legally bound to ensure that security measures are being
undertaken, and an insurance company might also require verication of these measures before
P1: JYS
APPLICATIONS OF RFIDTAGGING 31
FIGURE5.1: The iClass RFID identity card from HID
providing coverage. A traditional method for verication of a security patrol has been the use
of log stations; a guard is required to punch a card at a log station validating the location and
time of the patrol. A modern alternative requires that guards carry a handheld RFID scanner
which electronically logs the time strategically placed tags in the building are read. The scanner
can be interrogated at the end of a work shift to ensure the guard was present at all the critical
FIGURE5.2: Automobile ignition key with additional RFID activation
P1: JYS
32 RFIDEXPLAINED
FIGURE 5.3: RFID tags and handheld readers assure security patrols are carried out consistently and
according to a predened schedule (courtesy of Proxiguard)
locations and at the required times. This idea can be extended beyond security to any task that
needs periodic verication for location-based activities (Fig. 5.3).
5.1.3 Antitheft
Automatic mechanisms to protect a stores merchandize from shop-lifting have been used for
many years. The commonantitheft tagis a simple device that is attachedtomerchandise ina store
and disabled at the check-out desk at the time of purchase. However, in the case of theft, the tag
will not be disabled and trigger an alarm at the exit of the store. This kind of tag is usually based
on an inexpensive resonate circuit that can communicate its presence through load modulation
(see Section 2Principles of RFID). However, these tags can be thought of as binary RFID
tags, indicating their presence when interrogated, but do not provide any additional information
beyond this. However, if RFID tags become common for item level inventory control, they can
also serve a dual purpose providing an integrated antitheft capability, and may eventually replace
the simple binary tags.
5.2 TRACKING
There are numerous examples of large organizations that need to track the location of equipment
or people in order to operate efciently. This is a logistics problem that on a small scale is
easily handled by well-trained people, but on a large scale can only be achieved effectively by
automation. For example, consider the problem of tracking goods that might be waiting for
shipment from a factory, or in transit, or have just arrived at a distributor, or on the shelf of
a retail store in one of many possible locations. These problems can be mitigated by the use
of automatic identication, computer networks, and computer databases, which can be rapidly
P1: JYS
queried and searched to provide answers. Large retailers, health organizations, and military
operations, can all benet from automatic tracking enabled by RFID.
5.2.1 Supply Chain and Inventory Control
A driving force behind the widespread adoption of UHF based RFID is supply chain man-
agement. The potential for lowering the operational cost of the supply chain is what motivates
Wal-Mart, Tesco, Target, and other major retail stores to adopt UHF RFID into their work
practices. Even in a warehouse, inventory can be lost or misplaced, and RFID systems are well
suited to nding its location. This is due to a tags long read range, resilience, and the property
it can be read without requiring line of sight. The latter property has the corollary that it is
possible to continuously scan for RFID tags, and the supporting systems can therefore contin-
uously track the comings and goings of inventory without human intervention. RFID systems
are well suited to mitigating human error in warehouse environments which are dynamic and
sometimes hectic environments, resulting in a more efcient operation. More details about the
Wal-Mart RFID trials are presented in Section 7.
5.2.2 People
Sub-dermal tagging of animals is more socially acceptable than sub-dermal tagging of people.
The suggestion of tagged people immediately brings up images of George Orwells novel 1984.
However, there are already examples of injectable RFID tags that have been applied to people.
Kevin Warwick, and professor of Cybernetics, at the University of Reading experimented with
RFID in 2000. Placing a tag under the skin of his own arm and using it as a unique key to gain
access to his house (Fig. 5.4). He described his experiences in Wired Magazine [30]. In the
late 90s, Applied Digital Solutions produced the Veritag design specically for tagging people
FIGURE5.4: Subdermal RFIDtag in the armKevin Warwick (University of Reading) (Wired, February
2000)
P1: JYS
34 RFIDEXPLAINED
using a sub-dermal glass capsule; other human injectable tags are also made by Trovan Ltd., in
the UK. The technology has found a number of niche applications. In some parts of the world
such as Mexico, kidnapping is more prevalent than in the US, and tagging children may help
parents identify them years later, and possibly also serve as a deterrent.
5.2.3 Hospital Patients
RFID has application in the health care industry for tagging patients to ensure that medical
records are correctly associated with the people they describe, and that the correct medications
are administered. These records can also provide information about a patients allergies, and is
therefore critical for this association to be made correctly. Printed labels, or even typing in a
name on a computer keyboard in plain text, can lead to simple mistakes, e.g., Mrs. I. Smith
versus Mrs. L. Smith. However, many hospitals currently use barcoded wrist bands which also
solves the problem, and it is unclear if RFID will improve this work practice.
5.2.4 Runners
Since the late 90s the organizers of major marathon races in the US have provided runners with
an RFID tag that can be incorporated into the laces of their running shoes (see Fig. 5.5). This
became necessary to handle the logistics of timing 1000s of runners in a major metropolitan
race, and a solution to the problem of managing so many staggered start times because it may
take at least an hour for so many runners to cross the start line and begin the race. The system
operates by employing RFID readers at the start and nish, and other key checkpoints along
the race course. As the runners pass the tag reader stations, a time is recorded for each ID and
(a) (b)
FIGURE 5.5: (a) An RFID tag on a shoelace (courtesy of Texas Instruments). (b) This system is used
in the Boston Marathon
P1: JYS
(a)
(b)
FIGURE 5.6: (a) RFID tag mounted in the ear of a cow. (b) Dog with subdermal tag being identied
with a handheld reader
thus at the nish, the running time for each contestant can be calculated automatically as they
complete the course.
Furthermore, with several checkpoints on route it is possible for the organizers to pro-
vide breaking news of how the top runners are performing while the race is in progress. This
information is also of great interest to the athletes after a race, because the split times help them
understand how they performed through the event, and at what stage in the race they might be
able to push themselves in the future.
5.2.5 Cattle
Managing a modern dairy farm requires detailed accounting for the entire herd. This includes
monitoring how much they eat and a list of all the medications that have been administered.
Keeping track of the identity of each animal is therefore important, but conventional labels
such as barcodes cannot be used in the dirty environment of a farm. RFID is better suited as
it is not affected by soiling. Figure 5.6a shows an ear tag that has been designed for use under
these conditions. In recent years there has been cause for concern that livestock may contract
Mad-cow disease, and as a result there is more interest than ever in tracking the ownership and
medical history of a cow throughout its lifetime.
5.2.6 Pets
Many States in the US provide a service for owners to have their dogs and cats electronically
tagged using aninjected RFIDtag behind the ear or neck. If lost and thenlater found, authorities
recovering the animal can scan for a tag to identify the owners name and address. Stray animals
often lose their collars, and thus sub-dermal RFID is more likely to survive the ordeal. Figure
5.6b shows a commercial reader designed for this purpose. Unfortunately, there is no US-wide
standard for the type of tag used for this application, and States use tags that are not compatible
with the result that some tags are not detected. This is an example of why adopting uniform
standards is so important for successful RFID applications.
P1: JYS
36 RFIDEXPLAINED
FIGURE 5.7: An airline luggage label that provides RFID, barcodes, and printed information about
the owner, ight, and destination of the tagged bag (courtesy of Texas Instruments)
5.2.7 Airline Luggage
In an era of heightened terrorism concerns it is important that airline baggage be tracked and
ideally travel on the same plane as its owner. If bags can be automatically identied while
being moved to a plane we can have greater condence the process will be error free. Also, if a
passenger has checked in but later does not board the plane, RFID can help locate the bags in
the cargo hold for removal before the plane takes off. Figure 5.7 shows how traditional labels
can be integrated with RFID allowing backward compatibility with existing barcode systems.
A further benet is to help mitigate the costs associated with recovering a passengers
lost luggage and delivering it to the correct address. The US Bureau of Transportation claimed
that over a billion items of checked luggage were transported in 2004 and if only 0.5% were
misrouted, this represents 5 million lost bags. The accumulated costs of these mistakes can be
a signicant burden for an airlines operating costs, and automated RFID tracking can help
remove the human errors that lead to this problem.
5.3 AUTHENTICITY
In order to have condence that any manufactured itemof value comes froman authentic source,
it is necessary to have a means of validating its origin. Examples of authentication mechanisms
from other industries include hallmarks for items made from precious metals, watermarks on
banknotes, and artist signatures on paintings. However, all of these can be forged if investment is
P1: JYS
made in the appropriate equipment, but provide a high enough nancial hurdle that in practice
a forgery is a relatively rare event.
5.3.1 Money
The Mu-chip from Hitachi is one of the smallest RFID implementations at 0.4 mm 0.4 mm
and designed to be read at a very close range. The EEC have been considering embedding the
Mu-chip in future ECU banknotes, primarily to provide an automatic means of validating their
authenticity, and for rapidly counting a pile of notes. Such automation removes the chance of
humanerror whenreading denominations, or whenseparating notes that may have become stuck
together. However, adding a unique number that is automatically readable detracts from paper
moneys most valuable assetit is not easily traceable. Tagged banknotes may be automatically
tracked between transactions and thus provide information about when and where you spend
your money. In contrast, credit cards have long since given away this information, but the use
of cash in transactions has preserved our privacy. Adding RFID tags to banknotes is potentially
another area that may erode our fundamental right to privacy (Fig. 5.8).
5.3.2 Drugs
Pharmaceuticals often have a high market-value and are therefore a target for forgery. However,
unlike many consumer products, it is difcult for the lay person to know if the pills purchased in
a bottle are really the drugs they claim to be. We usually rely on the reputation of the dispensing
pharmacist to validate the purchase, but in the era of the Internet there are often attractive drug
purchases to be made online. Off-shore companies may well have lower operating costs and can
legitimately provide bargain drugs, but a consumer can no longer be certain they are getting
FIGURE5.8: Hitachis Mu-chipso small it can be sandwiched between the paper layers of a banknote
(courtesy of Hitachi, Ltd)
P1: JYS
38 RFIDEXPLAINED
(a) (b)
FIGURE5.9: FasTrak Toll Pass System: (a) Transponder tag and (b) Booth
what they paid for. Sealed packaging that includes hard-to-forge RFIDtags with batch numbers
allocated from the pharmaceutical manufacturer are a solution. The batch number could then
be read at home and validated online through the Internet. See Section 7 and antitamper proof
packaging.
5.4 ELECTRONICPAYMENTS
5.4.1 Auto Tolls
For many new road and bridge projects, the use of tolls has been the only way to raise the capital
investment to pay for them. However, tolls are inconvenient for drivers, who need to carry
the appropriate change. Trafc is also slowed, often leading to trafc jams at peak commuting
times. RFID technology can reduce these problems.
By placing a suitably designed tag in the windshield of a car (see Fig. 5.9a) a tag reader
at the tollbooth can automatically scan its ID as it passes by. The systems are designed so that
customers establish a prepaid account and a booth can then deduct the appropriate fee from the
account each time the ID is detected. The toll-reader technology has been developed so that
it can operate at freeway speeds and thus cars, in theory, do not need to slow down. However,
in practice most tollbooths are narrow (see Fig. 5.9b), and cars are required to slow down for
safety reasons, but even at speeds of 20 mph there is a signicant increase in trafc throughput.
In recent years more lanes at toll plazas have been converted to provide automatic toll
charges. On the west coast of the US, FastTrak is the predominant system, whereas on the east
coast, EZ-pass is more common.
5.4.2 Electronic Tickets
Ticketing is another domain where RFID can provide a unique contribution. A ticket is a
prepaid token that provides personal access to a resource, for example, a movie, exhibition, or
P1: JYS
FIGURE5.10: Ski pass with embedded RFID serves as a ticket to enter the chair lift
museum. Usually a ticket must be purchased ahead of time, but when it is used, no additional
money is required. Furthermore, a ticket can only be used once (although multiple tickets can
be held in one physical token) and they usually have a limited lifetime. The main purpose of
a ticket is to provide rapid access to an event when large numbers of people are converging on
the same location. Any delays in providing payment, giving change, or validating that credit is
available, are decoupled from the process of gaining access. Access to the event is achieved by
simply surrendering a valid ticket.
An RFID-based ticket has greater advantage over a traditional paper counterpart as it
can be left in a pocket while being validated, and electronically stamped as used, when the
patron passes through a turn-style. It can also store several virtual tickets in the same device and
surrender each one as required; or a single token can enable access for a limited period of time.
An RFID-based ticket can also be renewed electronically and thus can be used on multiple
occasions, which can offset the cost of the technology in comparison to printing tickets. RFID
has already been used to implement high-value tickets such as those found at a ski resort to
gain access to chair lifts (Fig. 5.10), and to provide daily access to the subway in Tokyo using
the Milfare system.
5.4.3 Electronic Credit
PayPass is a payment token being pioneered by MasterCard to provide a fast and convenient
method of buying low-value items. The PayPass token can be in the form of a card or a key fob.
To buy merchandise, a customer can use the token based on contactless RFID technology
(ISO 14443) to make a payment, simply by moving the token in front of a reader. No signature
is required for items below$25, making the transactions fast and convenient for small purchases,
but for items of greater value a PIN or signature is required. PayPass is expected to extend and
P1: JYS
40 RFIDEXPLAINED
FIGURE 5.11: A Star-wars character from Hasbro. Placing different characters on the podium plays
the sound and voice of that character
augment the current network of magnetic strip readers available for MasterCard purchases and
has been in trials since 2003.
5.5 ENTERTAINMENT
5.5.1 Smart Toys
The invisible nature of RFID communication has been used by some toy manufacturers to
create toys that appear to magically take on a personality when brought near other objects. The
Hasbro Star Wars character (Fig. 5.11) contains an RFID tag and when placed on a podium,
generates sound effects and speech associated with that character. Although the toy could have
been designed so that its tag had a simple unique ID that triggered the podium to play a
corresponding audio le, the designers felt is was more exible if the RFID tag stored the entire
audio clip and the reader simply had to play the le that it was able to read. Using this approach,
new characters could be sold without updating the software in the podium, providing greater
exibility for the manufacturer, and at the same time, keeping the consumer experience very
simple.
P1: JYS
41
C H A P T E R 6
RFIDIncorporating Sensing
One of the most intriguing aspects of modern RFID tags is that they are able to convey
information that extends beyond an ID stored in an internal memory, and dynamically read
data from an on-board sensor [31]. Today, there are commercial implementations of RFID
technology that can verify that critical environmental parameters remain within a safe range,
and as a result can be used to ensure the integrity of perishable goods, and protect the interests
of retailers and customers alike.
In the following sections we look at the various categories of RFID sensor in detail.
Specically, we examine the characteristics of sensing applications that are suitable for integra-
tion with RFID technologies, and provide examples of monitoring physical parameters such as
temperature, pressure, and acceleration; tamper detection; chemical or bio-agent detection; non-
invasive medical monitoring; the use of memory in combination with sensing, and techniques
for building longer range sensors.
6.1 EXTENDINGRFIDTOSENSINGAPPLICATIONS
The same mechanisms that enable an ID to be read from an internal register in an electronic tag
can also be applied to collecting data derived from a sensor. Extending the capabilities of the
silicon chip to interface with a sensor is straightforward, but the design of a suitable sensor is
usually an engineering challenge. First, the sensor will not be able to scavenge energy from the
reader while the tag is out of range; and this is likely to be the predomiant state during the tags
lifetime. Second, even when it is in read range, the available energy is very small. As a result, this
limits the capabilities of the electronics that can be used to process and record a sensor reading.
6.2 MONITORINGPHYSICAL PARAMETERS
An important application of RFID sensing is in the realm of monitoring perishable goods.
Typically items such as meat, fruit, and dairy products, should not exceed a critical tempera-
ture during transportation, or they may not be safe for consumption at their destination. An
RFID temperature sensor can serve to both identify and track crates of perishable goods, and
ensure their critical temperature has remained within recommended parameters [32,33]. An
example is the KSW TempSens RFID tag [34] (Fig. 6.1) which has been designed explicitly
P1: JYS
42 RFIDEXPLAINED
FIGURE 6.1: An RFID tag that can detect a critical temperature threshold (courtesy KSW Microtec
AG)
for this purpose, can be integrated with a standard product label. An example application is
illustrated by the challenges related to the transport of frozen chicken, which has a high risk
of salmonella contamination if allowed to thaw (Fig. 6.2). Furthermore, if later frozen again, it
may not be apparent from a visual inspection that a problem had occurred. A temperature mon-
itoring tag operates by incorporating a material in the tags substrate that makes a permanent
electrical change when the critical temperature has been exceeded. This can be represented as a
single binary digit appended to the ID, or if more bits are available, a measure of the maximum
temperature exposure. When the tag is read, it will not only respond with an ID, but also provide
a warning if the temperature variation has been an issue.
Monitoring the pressure of an automobiles tires from inside the vehicle is another appli-
cation that is well suited to the unique capabilities of remote RFIDsensing. This is a feature that
P1: JYS
RFIDINCORPORATINGSENSING 43
can be of benet to drivers, as a slow leak often goes by unnoticed before the tire becomes com-
pletely at. By the time the fault is discovered, tire damage and considerable inconvenience may
result. This is a challenging problem because the tires are spinning and sensors can only be con-
nected by wires if concentric connection rings are engineered around each wheel axial, making
the engineering complicatedandcostly. Furthermore, evena wiredconnectiontoanexternal sen-
sor placedonthe opening of a tire valve is likely tobe unreliable, andsubject todamage depending
on how the car is driven. Instead, some companies such as as Royal Philips Electronics have
been developing an RFID chip that can be bonded inside a tire, and read remotely by antennas
installed in the wheel hubs of the car. As there is no physical link, reliable interaction with the tag
is possible even while in motion. The tags can also provide additional information about the tires
maximum ination pressure and identify the tire for record keeping, e.g., frontback rotation
history. Currently, Michelin and other major tire vendors are trialing RFIDpressure monitoring
systems, but they have not as yet made it to the mainstream automobile market (Fig. 6.3).
Another physical parameter that can be monitored to useful effect is acceleration. If
a fragile package has been dropped during transport, it is likely a critical acceleration thresh-
old would have been exceeded. Today, some shipping companies employ a nonelectronic tag
solution to solve this problem, which utilizes a thin plastic membrane to hold a colored dye.
If the membrane breaks after an impact, the dye ows into a visible chamber and the tag
changes color. This kind of tag is used to detect poor handling in a warehouse, or loading and
FIGURE6.2: Apackaged chicken incorporating an RFIDtemperature sensor (courtesy KSWMicrotec
AG)
P1: JYS
44 RFIDEXPLAINED
FIGURE6.3: Car tire incorporating RFID pressure sensor readable from the car
unloading mishaps during transportation. RFID adds more utility to this solution by enabling
the automatic detection of a damaged item without having to inspect each package by hand.
To incorporate this capability into a tag, the dye in the example above can be replaced with a
conductive liquid, and electrodes in the rupture chamber would be linked to a circuit that in turn
changes the state of a bit when the RFID tag is read; the damage bit. This type of information
could have considerable application in retail stores as a checkout cashier monitoring the tag
reader can be made aware of damage before passing the product on to a customer.
6.3 TAMPERDETECTION
RFID sensing can also be used to support antitamper product packaging. Most modern con-
sumable products are protected by packaging technology that clearly indicates to a customer
if the product has been tampered with. Devices that detect tampering are relatively straight-
forward and generally require a simple, single-bit interface to detect the alarm state. A simple
binary switch-based sensor can be incorporated into an RFID tag, such as a thin loop of wire
extending from the tag through the packaging and back to the tag. If tampering occurs, the
wire is broken and will show-up as a tamper bit when the tag is read during checkout. In this
way, a store can ensure that it only purveys items that are tamper free. Moreover, at each point
in the supply chain, from factory to retail, it is possible to check individual products for tamper
activity making it easier to nd where a crime has been carried out.
P1: JYS
6.4 DETECTINGHARMFUL AGENTS
In todays political climate people and governments worry that chemical, biological, or radioac-
tive agents might be used by terrorists to threaten populated areas. Many of these agents are
themselves invisible, which adds to the danger because people might be unknowingly exposed
for long periods of time. The risks can be minimized if it is possible to build detectors that can
rapidly alert a population to the presence of the contaminant. Furthermore, if these detectors
could be deployed at a countrys ports of entry, it might help to identify terrorist plots before
they can be put into effect.
A problem with many conventional contaminant detectors is that they are relatively
expensive and cannot be deployed on the scale necessary to effectively protect a metropolis. An
RFID sensor based on passive detector technologies can be deployed more ubiquitously. The
reader part of the system, which is more expensive, can then be installed on vehicles or carried
by security personnel. The readers, congured to automatically interrogate nearby tags, would
provide a warning about the contaminant as they passed by. At present, sensors that detect
biological agents are very limited in scope. A great deal of work needs to be done in this area to
build passive detectors that are both effective and inexpensive. However, RFID can be used as
the reporting mechanism to make this kind of sensor practical.
A further example is detecting bacterial contamination of food products through routine
handing. Although some problems can be detected indirectly using a temperature sensor, a more
direct indication is given by testing a sample for bacteria growth. Auburn Universitys Detection
and Food Safety department is carrying out research that will allow them to build an RFID tag
providing a measurement of the growth of a particular organism (Fig. 6.4).
6.5 NON-INVASIVEMONITORING
Advancedmedical monitoring canalsobe supportedby RFID. Some diagnoses canonly be made
when there is direct access to the internal organs of the bodyeven advanced MRI scanning has
its limitations. Advances in biopsy techniques and keyhole surgery provide a partial solution, but
FIGURE6.4: AuburnUniversity RFIDbacterial sensor chip(www.auburn.edu/audfs) (courtesy Auburn
University)
P1: JYS
46 RFIDEXPLAINED
some conditions are progressive and call for continuous monitoring without repeated surgery.
This is where an RFID tag can play a valuable role. RFID sensors can be designed to be
placed in the human body during surgery. An external reader can then be used to periodically
communicate with the device either during routine visits to the doctor, or as a result of being
carried by the patient. A device such as this can provide an on-going and progressive evaluation
of the condition being monitored. Such designs are only in their infancy, but implantable
temperature sensing devices are currently being designed by Silicon Craft Technology [33],
allowing an accurate body temperature reading to be obtained for livestock implanted with the
device. As a result it is easier to detect infection and take the appropriate action early on. Such
devices may also be used with poultry to detect the onset of the deadly Avian Flu.
6.6 LOGGINGSENSORACTIVITY
Detection of an unusual sensor reading helps us know if a physical trigger has occurred, but
it would be even more useful to know where and when it happened. Unfortunately, without a
conventional battery, RFID sensors cannot continuously monitor the state of a sensor or utilize
an electronic clock and automatically record the time of a sensor event. However, readers with
accurate clocks (see Section 9) can help in this process by utilizing a tags on-board writable
memory, and recording read-time and sensor-state in a circular buffer. As a result, the time of
an aberrant sensor reading can be bounded by the prior and subsequent readings in the buffer.
The value of sensor data can be increase futher by adding location information. A reader
equipped with a Global Positioning System (GPS) can write the reader location into the tag,
along with the current time and sensor state. If GPS is unavailable, a low cost alternative is
a location reference tag near by that can be read to initialize the location of the reader before
it scans other nearby tags. The electronic memory in a population of tags can thus serve as a
distributed database of the sensing history, including time and location, without requiring that
all the readers coordinate their scanning activity through an external network. Although such
systems are not available today, the growing use of RFID tags, and the availability of on-board
writable memory, is likely to enable extended logging capability in the near future.
6.7 LONGERRANGESENSING
Most of the applications for sensing describedinthis sectionhave beenbasedaroundshort-range
RFID tags that incorporate sensing, typically using inductive coupling to derive their energy,
and load modulation for data return. There are many other categories of sensing application that
need to communicate over much greater distances. In recent years some of these applications
have been a research topic addressed by the ad-hoc sensor-network community [35]. Remote
sensing is achieved by building a wireless multi-hop ad-hoc network in which active sensor-
nodes transfer their information to a collection point, or network gateway, by hopping the data
P1: JYS
through a sequence of nodes. The research has addressed issues such as the optimum routing
algorithm to collect the data, and how to make best use of the energy available at each node.
These topics are often related because a network route might need to change as the result of a
failing battery at an intermediate node. Some researchers have proposed that sensor networks
may extend the reach of the Internet by allowing web-based clients to make queries about the
physical world. Wireless ad-hoc networks deployed on a large scale have the potential to provide
that service by bridging their data to the Internet at a gateway node. However, sensor networks
require power if they are to operate for an extended period, or else batteries need to be used and
replaced periodically, something that is too costly and labor intensive to consider on a large scale.
Long-range RFID sensing can provide a partial solution. The same principles used by
far-eld RFID can be used to build sensors that are powered typically up to 20 feet, and
communicate data back to the reader using backscatter modulation. At Intel Research Seattle
the Wireless Identication and Sensing Platform(WISP) initiative sets out to explore the limits
of long-range RFID sensing. A WISP is an augmented RFID tag designed so that it can reside
in the interrogation eld of a reader and accumulate more energy than is needed to perform
a simple ID function. The additional energy can be used to power a microcontroller with an
A-to-D converter, allowing a sensor to communicate its state. The energy reaching a tag from a
reader will approximate to a 1/(distance)
2
function, and therefore at 20 feet the energy available
is small. The availability of the sensor will be determined by the relative time spent reading the
sensors state in relation to the time it spends in the eld between interrogations.
Some of the rst motion sensing WISP designs [36] required no more energy than a
conventional RFID tag. Simple passive motion-sensors, provided by two miniature mercury
switches were used to select between two unique ID-chips, independently connecting each
one to a common antenna. The mercury switches can be physically mounted 180-degrees with
respect to each other so that as a WISP is inverted it will switch its ID, and as a result, the WISP
tag can sense tilting or rotational motion. If it were attached to a vertically mounted spinning
wheel, the rate of alternation between the two IDs, as determined by a reader continuously
interrogating the tag, would measure the rotational speed of the wheel; as long as the centrifugal
acceleration of the wheel is less than the acceleration due to gravity.
Zero-power sensing is limited to a small set of sensor technologies (such as mercury
switches), but most of the more diverse sensors require power. The later WISP designs [37,21]
use an energy scavenging circuit that can efciently store enough energy to power a sensor,
plus the associated conditioning circuits, and a low-power microcontroller (Fig. 6.5) which can
also provide digital ltering, a coded representation of the measured parameter over time, and
error-control coding.
Using this approach, long-range RFID sensors (20 feet) can be used to extend ad-hoc
sensor-networks another 20 feet into areas where it may only be feasible to provide power to
P1: JYS
48 RFIDEXPLAINED
FIGURE6.5: Showing a prototype power scavenging WISP tag
one central node (the tag/sensor reader). For example, if RFID strain gauges were placed in
an engineering structure such as a bridge to determine its response to load, using conventional
designs, cables providing power and sensing would be needed to connect to each sensor. Using
WISPs, only a few tag reader hubs would be necessary to interrogate the outlying sensors in
the extended locality.
Other applications of long-range RFID sensors include: warehousing (measuring condi-
tions inside crates high up on storage shelves), monitoring long runs of pipes, and radiation/
chemical detectors inside a transport container.
6.8 SUMMARY
As RFIDbecomes more prevalent, growing economies of scale will allowenvironmental sensors
to be integrated with tags allowing RFID readers to report back on a wide variety of real-world
conditions. Tag readers that are typically powered by the mains, or conventional batteries, will
in many cases be able to access wireless networks connected to the global Internet. They will be
able to relay the state of the physical world, and make it available to users and servers through
web-services, allowing real-time data mining to be implemented on a scale larger than ever
before. RFID sensing is part of the classic story of Ubiquitous Computing [27], a technology
that can improve our lives while remaining invisible, as users do not have to be directly aware
of it to reap the benets.
P1: JYS
49
C H A P T E R 7
Deployment and Experience with
RFIDSystems
This section describes trials of RFIDsystems that are breaking newground by utilizing RFIDto
support commercial ventures that have not previously adopted this technology. We summarize
three high-prole systems that provide insights into the benets of modern RFIDapplications:
Store of the FutureMetro AG;
Wal-Marts trials of RFID;
Frankfurt airports maintenance operations.
Looking further into the future, we describe two additional deployments from the realm
of research that provide a glimpse of the variety and scope of future tagging systems:
iBracelet: supporting work practices;
University of Washington plans for Ubiquitous RFID reader deployment.
7.1 STOREOF THEFUTUREMETROAG
One of the best known, and most progressive, trial sites for testing RFID concepts in the
retail trade is the Store of the future established in April 2003 by the Metro AG group in
Rheinburg, Germany (the worlds fth largest retailer) [38] (Fig. 7.1). Financial support has
also been provided by major IT companies such as Intel, Philips, SAP, and IBM who strive to
learn howRFIDmight change processes in the retail trade, and howto integrate this technology
with existing supply chain management systems, and the computer systems that support them.
Metros mainobjective is to improve the management of its goods at all stages of the supply chain,
from a suppliers warehouse to the shelves of retail stores, utilizing the efciency of automation
wherever possible. In addition, the store provides an opportunity to explore applications beyond
the supply chain, providing a richer and more engaging experience for the customers, enabled
by the use of modern RFID tags and other state-of-the-art technologies.
P1: JYS
50 RFIDEXPLAINED
FIGURE7.1: Metroa future store used to test new RFID concepts
Customers experience the technology as soon as they enter the store pushing a smart cart
that has a tag reader built in. Soon they pass smart shelves that also contain RFID readers, and
smart checkout aisles that speed the process of payment. The smart carts also contain a tag and
their identity recorded as they enter and leave the store. The number of carts in the store can
be used to gauge how many checkouts need to be open for optimal efciency. As the customers
remove items fromsmart shelves they automatically informthe store management systemabout
the goods that need to be replaced and where they need to go. A customer can also scan items
as they are placed in their cart to keep a running tally of the purchase cost (initially only some
items were tagged with RFID and barcode scanners were also used). At the checkout, the carts
RFID tag is further identied and the grand total displayed for payment.
Specic suppliers such as Gillette and Proctor & Gamble were early collaborators, but
commercial interest has been growing, and by 2006 over 300 suppliers for the Metro Group
are expected to incorporate RFID in their delivery pallets as part of this trial. There continues
to be some debate over the benet of pallet versus item tagging. The greatest benets initially
come from pallet tagging, but item tagging is likely to follow once the cost of the tags drop
further. RFID labels on individual items can also serve as an antitheft mechanism when used in
combination with readers at store exits. By comparison, conventional UPCcodes do not provide
any form of antitheft detection, and if products need to be protected in this way, they require
an additional antitheft tag. RFID can serve as an identity, an antitheft tag, and even provide a
tamper detection mechanism (see Section 6).
At the Metro Store of the Future, RFID has also been used to test new services that
help customers choose merchandise they may wish to buy. For example, tagged CDs can be
passed over a reader station attached to a computer, and a snippet of the music played back
without opening the CD package. Food that may have an expiry date can be interrogated as it is
P1: JYS
DEPLOYMENTANDEXPERIENCEWITHRFIDSYSTEMS 51
purchased to ensure it is still fresh enough for consumption. And tagged clothes can be scanned
in front of a monitor, showing the customer howthey look on a model. Several items scanned in
this way allow combinations of clothes to be seen together in order to quickly understand how
they blend, and all without trying anything on. These systems can also make recommendations
for alternative garments that might replace, or be added to, existing choices. Even in a dressing
room this system can have added value by alerting staff to bring new garments, or additional
sizes, to help a customer nd the items they are looking for. The jury is still out on the value
of these services, and many others in the planning stage, but the Metro trial does allow these
ideas to be tested and move beyond idle speculation.
Reports from Metro in early 2006 indicate that efciency and cost saving are already
apparent, and 250 of its 2300 stores in Europe and Asia are in the process of installing RFID-
based supply chain management systems.
7.2 WAL-MARTRFIDTRIALS
Wal-Mart, the largest retailer in the world, has been one of the driving forces spurring the RFID
industry to provide effective solutions to improve the efciency of supply chain management
(Fig. 7.2). Initially, the deployment of RFID technologies was limited to tagging pallets and
cases, and not individual items, as it was believed that the cost savings in this area alone
would provide enough justication for the project. In 2004, at an early stage of deployment of
EPC-Global tags in their supply chain, the measured benets were described as promising.
FIGURE7.2: One of Wal-Marts early RFID trials stores
P1: JYS
52 RFIDEXPLAINED
By October 2005, the results reported were more quantiable, with data captured during a
29-week trial period in 2005, using 500 of its Wal-Mart and Sams Club stores, supported by
140 suppliers. The results can be summarized as:
16% decrease in out-of-stock items for EPC tagged products; and
300% improvement in restocking time for EPC tagged items in the store.
It is further reported that at the trial sites excess inventory was also lower, although not
quantied at this time. Riding on this success, by Q1 2006 the number of Wal-Mart suppliers
incorporating RFID is expected to increase to 300 [39]. The RFID industry has been fortunate
that Wal-Mart has chosen to lead the charge. If it were not for its premier market position, a
lesser company would probably not have the economic power to persuade suppliers to adopt
RFID, which at rst is only going to benet the retailer. Supporting Wal-Marts position, Tesco,
the UKs largest retailer (the worlds third largest) is also pioneering the adoption of RFID. In
the US, the DoD is another major consumer of goods and materials with its own unique set
of supply chain requirements, and it too is carrying out trials based on EPC RFID tags. In
summary, there is considerable momentum behind RFID standardization and adoption by the
major retailers of the world, and is likely to have a knock-on effect for smaller companies.
Having the rst quantiable numbers that demonstrate real value in the supply chain
will also provide a foundation for the RFID-based services being investigated in the Metro
AG stores. From a customer perspective these are likely to be more exciting than improving
the supply chain, but would not have been a large enough economic driver to motivate RFID
adoption on their own.
7.3 RFIDSUPPORTFORMAINTENANCEOPERATIONS AT
FRANKFURTAIRPORT
The second largest airport in Europe is Frankfurt, typically handling over 50 million passengers
per year, and in some ways can be compared to a small town, relying on numerous utilities
and support services for daily operation. Although there are many well known applications of
RFID in an airport, such as tracking airline-baggage as described in Section 5, there are many
other lesser known procedures that can benet from it. For example, electrical systems require
regular maintenance to provide uninterrupted operation and ensure the safety of passengers.
The University of St. Gallen have studied how RFID has been successfully trialed at Frankfurt
airport [40] to improve the efciency of maintenance operations, and their observations are
summarized here:
The airport maintenance operation group, Fraport, is responsible for ensuring smoke and
re control systems are always fully operational. The danger of not doing so is highlighted in the
P1: JYS
1996 Dusseldorf airport re that resulted in loss of life. Following this tragic event, to improve
on safety the German government created new legislation that mandated the use of improved
and more extensive maintenance logging to ensure essential work would be carried out in
the future. Given the cost constraints of maintenance organizations, some of their activities
are carried out by external rms and it is necessary to set up procedures that can efciently
validate and record the maintenance process, which is made more difcult as the work force is
rapidly changing. The airport authorities have a responsibility to ensure this work is performed
according to government regulations, and in the event of an actual re the authorities would be
negligent if proof of maintenance was not available.
In 2003, Fraport, decided to use RFID (Fig. 7.3) to support the inspection and mainte-
nance processes required for the re shutters and ventilation system installed throughout the
airport. The system made use of RFID by deploying a tag next to each of the re shutters. A
maintenance technician, supplied with a handheld computer incorporating an integrated RFID
reader, checks each shutter and uses the computer to log each associated tag, thus providing
evidence each re shutter location had been visited during the inspection process. For each scan
of a tag the computer engages the technician in an electronic dialog to ensure that every aspect
of the inspection/maintenance has been considered and the answer recorded. At the end of the
dialog, the technician is asked to scan the tag once more, this time writing a reference to the
computer log into the tags memory, along with the date and time of the inspection. This process
locks the log le so that it cannot be changed at a later time, and provides two independent
points of reference for the inspection, the RFID tag and the computer.
Since the electronic log replaced a previously handwritten report that was often incom-
plete, and resulted in 88,000 pages of logs per year (for 22,000 shutters) which also needed to
be archived for 10 years, the system was a considerable improvement and provided tangible cost
saving. Furthermore, it is reported that the technicians are now more motivated, preferring this
process over handwritten documentation. Based on this positive experience, Fraport is likely
to extend RFID supported maintenance to other aspects of the re prevention system, and to
tracking mobile equipment that is frequently misplaced within the airport campus.
7.4 INTEL RESEARCH: IBRACELETANDDETECTING
THEUSEOF OBJECTS
Intels Research Laboratory in Seattle (IRS) has been experimenting with systems that improve
work practices, and enable the elderly to live independent lives in their own homes for longer
than would have been previously possible [41]. The thesis behind this work is that if you can
understand what somebody is doing through their actions, you can automatically provide help
when a problem arises. In the case of the elderly, it may also be possible to use a log of their
actions to determine if a persons mental ability is stable or in decline.
P1: JYS
54 RFIDEXPLAINED
FIGURE7.3: RFID maintenance tagalso includes a barcode for redundancy
In order to support this exploratory work, a system must be built that can accurately
and automatically capture events in our daily routine both at home and at work. Although
this could be a task for computer vision, placing RFID tags on all the objects that need to be
monitored, and employing wearable tag-readers to log the objects used by our test subjects,
is an alternate approach to the problem. The tagging solution also has some advantages in
that the processing requirement for the system is reduced, and the accuracy of interpretation is
much greater. Furthermore, cameras raise many privacy concerns, whereas reading tags in the
environment is less intrusive with arguably better delity.
Since we interact with most of the things in our lives by touching them with our hands,
an RFID reader was designed to be worn on the hand, and record nearby tags. During the
evolution of this concept the initial implementation consisted of a modied glove, the iGlove,
which incorporated a short-range reader on the top side of the garment. However, although
successful as a reader, the glove was uncomfortable to wear indoors and impractical in many
domestic situations and so an improved follow-on design was created based on a plastic molded
braceletthe iBracelet [42, 43] (Fig. 7.4). This device was more convenient to wear, but
mounted further away from the ngers compared to the iGlove solution. As a result a major
consideration of the design was howto increase the range of the reader fromthe previous 2.5 cm
to 30 cm (the reference says 10 cm but the RF design has improved further since then), without
signicantly increasing the power consumption of the device, and hence without decreasing its
battery life. This problem was overcome by improving the design characteristics of the antenna
and associated load demodulation.
P1: JYS
FIGURE7.4: The iBracelet created by Intel Research Seattle
The iBracelet is currently being consideredfor trials withinIntels siliconfabricationplants
in order to track the progress and management of silicon wafer carriers at the planta valuable
commodity that when misplaced, or more seriously process steps are left out, can result in a
considerable loss of revenue. Other projects at fabrication plants, such as the LotTrack project at
Inneon, have already found value in tracking wafer cassettes using RFID [44]. iBracelet is also
expected to provide support for community Aging in Place projects, but at this stage it is being
used to verify that statistical models can be constructed that produce accurate recommendations
in a test environment before being trialed in the real world.
7.5 UNIVERSITY OF WASHINGTONS RFID
ECOSYSTEMPROJECT
The University of Washingtons Department of Computer Science & Engineering is in the
progress of deploying a large experimental RFID system in their new building, the Paul G.
Allen Center on the UW campus (Fig. 7.5). By deploying ubiquitous RFID readers mounted
at doorways, ends of hallways, and other pedestrian funnel points in the building, it should
be possible to monitor the comings and goings of many different types of tagged objects.
The objective of the project is to investigate a host of consumer (as opposed to supply chain)
applications of RFID such as reminding, nding lost objects [45], inventory control, gaming,
activity inference, etc. At the same time, the focus is on building a data architecture that will
respect user privacy and allow users to retain control over how their RFID tags are utilized. To
that end, the research is investigating approaches for encryption, selectively enabling tags for
specic applications, and using readers that act as local RFID sensors and rebroadcast their tag
P1: JYS
56 RFIDEXPLAINED
FIGURE7.5: Paul Allen CS Building at University of Washington
reads rather than sending themto a centralized database. The project is currently funded by NSF
and the UWs College of Engineering with support from Impinj, Inc., a leading manufacturer
of EPCglobal Generation-2 tags and readers that enjoy the benets of long read-ranges, fast
multitag singulation, and on-tag memory.
7.6 FUTUREDEPLOYMENTS
Looking beyond the current commercial applications, and the new trials in progress, RFID
technologies and large-scale business applications are still in their infancy. There are some
problems to overcome, but the technology is versatile and can be adapted along many dimensions
to provide effective solutions. If this text were to be revised in the future, the scope of this section
with respect to deployment and learning, will be likely evolve considerably.
P1: IML
57
C H A P T E R 8
Privacy, Kill Switches, and
Blocker Tags
One of the reasons RFID has been written about so much in recent years is that some people
believe the introduction of RFIDtechnology will erode their right to privacy [46]. Privacy advo-
cate groups are concerned that even though many of the corporations considering using RFID
as part of their inventory tracking mechanism have honorable intentions, without due care the
technology might be unwittingly used to create undesirable outcomes for many customers. The
inherent problem is that radio-based technologies interact through invisible communication
channels and we are not aware when communication is taking place. Consider a situation in
which RFID tags are used to label garments in a clothing store. From the stores perspective
a conventional inventory stock check is difcult because customers frequently mix-up the gar-
ments, and theft can take place making the sales records incomplete. On the upside, with RFID
tagging the various racks and bins of clothes can be checked very quickly, even when muddled,
improving the efciency of the store. On the downside, if a tag is not removed when a customer
buys an item of clothing and later wears it, the tag can be used to track them wherever they
go. This capability might be used by other vendors to learn about the shops they frequent, and
then target them with direct marketing based on this information. This scenario was presented
graphically in the science ction movie Minority Report in which the hero, played by Tom
Cruise, was identied in department stores not by RFIDs in his clothes, but by his eyes and
the use of ubiquitous retina scanners. As a result he was subjected to a multitude of multimedia
marketing materials chosen to appeal to his lifestyle (see Fig. 8.1). In this story in which he
was trying to avoid arrest, the solution was more dramatic than removing tags from his clothes,
he had to nd a surgeon that could perform an ocular transplant. While removing RFID tags
from clothing purchases would not be so dramatic, it would be very frustrating to have to take
this kind of action in order to maintain personal privacy.
In an even scarier scenario, criminal elements could judge your personal wealth depending
on the purchases you have made, and then target you for theft. It was because of a growing cloud
of public and media concern that Benetton, a well-known clothing store, had to make a hasty
P1: IML
58 RFIDEXPLAINED
FIGURE 8.1: Scene from the movie Minority Report in which billboards customize themselves to
the shoppers in the vicinity
retreat after it announced its plans for using RFID tags in its stores [47]. A similar response
resulted fromthe US governments plans to put RFIDtags into passports in an attempt to make
them easier to check at borders, and harder to forge. However, privacy advocates would argue
that covert readers might steal information that can be used to enable identity theft [48]. In
contrast to Benetton, the passport scheme is still going forward, although its implementation
is being modied to address some of the public concerns.
This potential for misuse of RFID is high, but like many modern technology debates the
story often has two sides. The undesirable scenario relating to tagged garments described above
can be turned into a potentially useful one. Washing machine manufacturers could integrate
RFID readers into the door of their machines, making the machines aware of all items that
have been selected for washing. As a result, they could choose the appropriate washing cycle,
and possibly warn you about incompatible garments that might result in color runs.
8.1 KILL SWITCHES
In order to overcome many of the concerns, EPCglobal designed a feature into its RFID tags
called a Kill Switch. This allows vendors to permanently disable an RFID tag at the point it is
sold, without necessarily having to remove the tag itself, which might be woven into a garment
deliberately making the tag difcult to remove, and serving as an antitheft device. Kill switches
can certainly help, but there are additional concerns that retailers may become complacent, and
that not all stores will be vigilant about disabling the tags. It is still possible that an insidious
number of operational tags could hitch a ride in our clothing and later on, criminal elements
could take advantage of this situation.
P1: IML
PRIVACY, KILL SWITCHES, ANDBLOCKERTAGS 59
8.2 BLOCKERTAGS
RSA corporation has proposed a solution that individuals can take into their own hands, the
concept of a Blocker Tag [49]. This is a modied RFID tag that takes advantage of the anti-
collision protocol used by EPCglobal Generation-1 tags by responding to each interrogation
in such a way that it appears that all possible tags are present. As a result the tag reader has no
idea what tags are actually near by. Perhaps having simple countermeasures to prevent misuse
of these tags is exactly what is needed to overcome privacy concerns.
8.3 TAGGINGIS ALREADY ANINTEGRAL PART
OF MODERNLIVING
Taking a more general view of electronic tagging; cell phones, credit-cards, and networked
computers, similar to RFID, are all technologies that make use of a unique identity which is
a fundamental part of their operation. Because of its uniqueness, the number can be used to
identify and locate each instance of the technology, and furthermore, because these items are also
related to chargeable services, the service providers are able to track the locations and activities
of their customers. So, how are RFID tags different and should we be any more concerned
about them than computers, cell phones, and credit cards that are already part of our lives?
One difference is that all these devices provide considerable utility in a modern world,
and, so far, we have been prepared to give up some amount of privacy to enjoy their benets.
As with most technologies there are advantages and disadvantages, and we must individually
evaluate whether the benets outweigh the cost. On the other hand, RFID tags embedded
in the merchandize we buy have no direct value to us. In fact, on the downside, undesirable
applications such as personalized marketing campaigns are the most likely result. Perhaps it is
the imbalance of the technologys pros and cons for the consumer, in favor of the cons, that has
contributed to antagonism toward RFID tagging.
When using cell phones, credit cards, and computers, we inadvertently give up informa-
tion about who we speak to, our location, what we like to buy, where we buy it, and when we
surf the world-wide-web we also give away our personal reading preferences. We continue to
do this because mostly this information is kept condential, and no detrimental consequences
result. We should ask ourselves if this will be true for RFID. For some technology we can take
control of the privacy issues by carefully choosing how we use it. For example, credit cards give
away our location each time we make a purchase, but as the cards are personal and we control
the account, we can explicitly decide when and where to apply them. The use of cash is always
an option, and just knowing there is an alternative makes credit cards more palatable.
Cell phones are more problematic in this regard because they do a poor job of protecting
location privacy even when we are not making a phone call. Simply put, a wireless service
P1: IML
60 RFIDEXPLAINED
provider needs to know your location at least to the closest cell tower, in order to route your
phone calls. When carrying a cell phone that is turned on, we are continuously giving away
our location without explicitly intending to do so. We can choose to turn the phone off, and
only turn it on to make a call, but then we lose all of the advantages of being able to receive
calls from friends and colleagues. Further, the short-range radio technology Bluetooth, which
is also integrated with many cell phone products, contains a unique MAC address to support
its protocol. And when turned on, it is possible to track your location using nearby computers
that can automatically discover the device. The Bluetooth radio can be turned off manually, but
many people are not familiar with this cell phone capability and simply may not realize it is a
privacy concern.
Identity theft is also a danger when our identity can be dened by a single unique number.
Even if we are personally unconcerned about technologies that reveal our identity and location,
perhaps because we trust the service providers and feel we are doing nothing wrong, there is the
potential for nefarious individuals to steal our identity. For example, if our cell phone or credit
cards are stolen, the thief will appear to take on our identity and journey to places and initiate
transactions that are beyond our control. In the case of a phone or credit card, which is used on
a daily basis, it is likely we will discover the loss early on and report it to the service provider
or bank, before too many charges accumulate. However, we might still need to dispute a bad
credit report or a large phone bill, if the theft is not reported promptly, and therefore must take
special care to keep track of these items.
In the case of RFID tags, identities may be stolen inadvertently, not because anybody
wanted to steal a tag, but because the tag was embedded in a high-value item that had been
associated with a person at the time of purchase. In a world in which tags and readers become
more ubiquitous, a thief carrying your purchased items may inadvertently link you to a crime
scene. In one vision of the future, if such inferences can be easily made, it may be necessary to
keep track of all personal items that contain RFID, and report them when stolen. This has an
upside and a downside. The upside is that stolen items may be more easily recovered when you
report the theft and the thief will be caught. Or the downside, you do not report an item stolen,
and your identity is falsely associated nefarious events. However, although inference based on
RFID, or other forms of electronic identity, may cast suspicion, without additional proof it is
likely we will always be able to claim plausible deniability, and further proof will be required.
It should be noted this concern is not without precedent, as in the past car license plates have
been used to infer the identity of people attending illegal events, simply by noting the license
numbers of cars parked nearby. However, innocent people who happen to park nearby also
become suspects. False inference of this kind provides us with reasonable cause for concern
about the potential secondary uses of RFID tags.
P1: IML
PRIVACY, KILL SWITCHES, ANDBLOCKERTAGS 61
8.4 FUTUREIMPACTONSOCIETY
In this section we have considered the pros and cons of using RFID and why concerns arise.
Looking forward, how will this technology be deployed and will the concerns be taken into
account to create the appropriate legislation? Is it also possible that our current expectation of
person privacy will soon to be lost?
RFID is just one of many new technologies that are able to track our location and
automatically identify us. Consider how many trafc-cams installed along roads can record
your cars license plate as you drive to work. Given the application of technology in all aspects
of life is continuing to increase, it is worth considering how society can inuence the use of
emerging tagging technologies, and how it might adapt to their adoption.
Opposition to RFID tagging is most likely to be strong in the labor force that supports
industrial manufacturing. Due to competition that arises from globalization, this industry is
forced to remain competitive at every level, a problem undermined by the labor costs in the
industrialized countries. Tagging provides a tool to monitor work practices and improve ef-
ciency. If deployed in moderation, the results are likely to positive but if too aggressive, may
become onerous for workers. The latter is clearly a future we should try to avoid. On the other
hand, for example, if one person is able to rapidly record the inventory of an entire warehouse,
using a motorized cart, a computer and an RFID reader, something that previously would have
only been possible with a team of workers, this is a level of progress that industry cannot afford
to ignore. Society, on the other hand, must then come to terms with more job losses, and the
retraining of workers to create skills that are in demand.
As a further illustration of the reaction to electronic tagging, in 2002 when Tesco in the
UK started RFID trials at one of its grocery retails stores in Cambridge, the result was open
protest outside the store. Figure 8.2 shows one of the protesters holding a poster with the slogan
Say No to Spy chips. The concerns are real and should be taken seriously if the technology
is going to be deployed effectively without a consumer or worker backlash.
In the US, currently, there are pressures on the government to gather internal intelligence
in order to combat terrorism. After the tragic attacks on the World Trade Center in New
York, and the Pentagon in Washington on 9/11 2001, there is well-founded concern that terror
groups waiting for similar opportunities are already within the country. The ability to keep
track of electronic identities could play a role in helping the homeland security ofce do their
job. By gathering information about events involving mobile devices, people, and places, and
continuously feeding this data into computational inference engines, it may be possible to look
for suspicious circumstances that need further investigation. RFID may also play a role in these
investigations. Because repeating the circumstances of 911 is something that we all wish to
avoid, it could be argued that the government has the right to track electronic identities in this
P1: IML
62 RFIDEXPLAINED
FIGURE 8.2: (a) Tesco trial store in Cambridge, UK. (b) Protester outside Tesco in the UK voicing
concerns about RFID trials
way, and to some degree should be a counter-balance to the concerns described earlier. Just as
airport security checks are an invasion of privacy and often inconvenient, it is something we
are all prepared to endure, because the consequences of poor airport security are too terrible to
leave to chance.
In a future world in which RFID has been deployed on a grand scale, it is possible there
will be databases that record our identity along with the time and places we go from day to
day. This information can be protected and used for the benet of society, or be made available
indiscriminately, and it will be up to government policy to protect us as these policies evolve. In
Europe, The Data Protection Act already limits access to all computer records that contain
private information requiring written consent for its disclosure, but similar legislation does not
exist in the US at present. As RFID technologies mature, and the EPCglobal standard becomes
adopted around the world, the issues surrounding the use of electronic tagging will become
better known, and it is likely there will be more international agreement on when and where
this type of information can be disclosed.
P1: JYS
63
C H A P T E R 9
Opportunities for RFIDIntegrated
with Memory
Adistinguishing feature of modern RFIDis that electronic tags can nowcontain far more infor-
mation than a simple identity. Today, it is possible to integrate additional read-only or read/write
memory into a tag, which can be queried or updated by a tag reader/writer. If, in the future,
RFID becomes an established technology that is used to label common products, distributed
read/write memory in these devices could become a resource that application developers will use
to their advantage [50]. Here we discuss some of the possibilities for this distributed memory
revolution.
9.1 READ-ONLY MEMORY
When using RFID tags to identify consumer products, additional read-only memory in the
tags can store product details. This information does not need to be read every time a tag is
interrogated, but is available if required. For example, the tag memory might contain a batch
code, and if some products are found to be faulty, the batch code can be used to nd other items
that potentially have the same defects.
An alternative approach to tag-based memory is to use a tags unique ID as a key into an
online database to recover the batch code, and other product specic data. However, there are
many situations in which communication with a database may not be possible. For example,
the store selling a product may not have access rights to the computer systems used by the
manufacturer. By writing the batch number directly into the tag, it is accessible at all stages
of the supply chain. Consider another example involving a tagged parcel that is misdirected
during transportation; the receiving organization may not be able to determine its intended
destination. Additional information in the tag can be self-describing, and include the name of
the destination stored as a human readable text string within its memory, thus obviating the
need and cost of a fully networked tracking system.
Although todays passive writable RFID tags can only store up to approximately 8000
bits, in the future, tags may have much larger memories with megabits of data. Taking advantage
P1: JYS
64 RFIDEXPLAINED
of this opportunity, manuals, or other documentation associated with a product could be stored
in the same tag that also identies it for sale. Further, by embedding the tag into the products
case, it cannot be easily separated; with the result that the documentation will remain readily
available. This is an advantage because paper manuals are frequently lost, leaving the owner in
some difculty when trying to nd out how to use an unfamiliar product feature.
Similar to the batch-code example given above, nding a products manual also has a
network-based solution. The manufacturer can provide a consumer with an online manual
through the world wide web. However, this approach may not stand the test of time as the
consumer must now rely on the manufacturer to maintain the website. In practice, modern
products have short lifecycles, and the companies involved can fail nancially, thus the web-
based information might disappear even though the products are still in use. RFID-based
memory embedded in the product does not have this shortcoming. Furthermore, it can also
help conserve natural resources by lessening the environmental impact associated with creating
extensive paper documentation.
9.1.1 Enhancing Objects with RFIDMemory
There are many examples of howRFIDcanusefully augment interactions withour environment.
Consider a poster advertising a movie at a nearby theater (see Fig. 9.1). It will likely contain a
title, a graphic that depicts the story and the characters, a list of actors, and below that a date
FIGURE 9.1: A movie poster with additional information provided by embedded RFID tags (poster
Newline Cinema)
P1: JYS
OPPORTUNITIES FORRFIDINTEGRATEDWITHMEMORY 65
indicating the opening night. RFID tags can be attached to the poster behind each of these
regions in order to provide additional information to a curious passer by. For example, the poster
may have intrigued you, but you may want to learn more about the movie before deciding to
see it. A pocket computer, or smart phone, with an RFID reader can be used to interrogate
the tagged title region and obtain a detailed summary of the story. Moving the reader over the
pictures of the actors would provide information about other movies they have starred in. And
moving the smart phone over the text of the movies opening date and time, could instruct your
smart phone to open your electronic calendar and enter the movie title into your schedule at the
corresponding date and time. This example illustrates two points that have not been discussed
earlier. First, an object may have multiple RFID tags attached to it, the purpose of each being
indicated by its position on the host object. Second, additional instructions stored with the data
can suggest how to process it. In the last example it was processed as calendar information.
This technique can also be used to initiate nancial transactions. Extending the poster example,
there might be an additional area that advertises Buy Ticket Here. Scanning this area with a
smart phone supporting both an RFID reader and GPRS capability, could automatically and
wirelessly connect the phone to a ticket ofce provided by an internet service, and purchase the
ticket electronically. To mitigate accidental purchases, when the phone is unwittingly brought
too close to the poster, the system should provide a conrmation dialog before committing to
the payment.
9.2 READ/WRITEMEMORY
More intriguing applications of RFID take advantage of read/write memory available in some
types of tags. Writable memory sizes are likely to follow similar trends to that of read-only
memories. However, as data can be stored in arbitrary formats by an interrogator, its use is only
limited by the creativity of the application developers.
For example, secondhand consumer goods which contain write-once embedded RFID
tags may tell you something about the prior list of owners and when and where ownership
changed hands. This is similar to the provenance documentation that usually accompanies
valuable antiques. RFID tagging may extend this kind of tracking to everyday items, allowing
consumers to have greater condence that they are making good purchases and the price of
the item will be reected by its history. As a result there is the potential to have a higher resell
value when the provenance is favorable, which may motivate buyers and sellers to ensure these
records are correctly maintained.
One of the consequences of using RFIDfor most forms of automatic identication is that
in time it may lead to an extensive deployment of electronic tags in our surroundings. If these
tags contain digital read/write memory, our homes, cars, ofces, and cities could soon have the
memory resources to store sizable amounts of data. For example, if public places are tagged
P1: JYS
66 RFIDEXPLAINED
with RFID [51], this can be a resource available to the city planners to store location-based
information accessible by the public. Such memory might be used for recording historical data,
or information about community services available in the locality, or provide data about the
presence of utilities, such as electrical cables, gas, and water pipes. An advantage of RFID
read/write memory is that its use can be decided after its deployment, and thus city planners
only need to ensure memory-based RFID is put in place during construction or renovation, and
decide on its content and application at a later time. Furthermore, information can be written to
these tags using a simple handheld interrogator, and thus maintaining the information is no less
burdensome than any other city maintenance task, such as checking conventional signs are up-
to-date. RFID-based information in this context can also save the city money when compared
with the cost of creating physical signs that must be large and made of materials robust enough to
endure severe weather. Furthermore, updating physical signs is a costly endeavor, but updating
information stored in an RFID memory is inexpensive.
RFID memories can also be made available as a medium for public messaging. In the
same way that message boards are available at the entrance to some campuses, you might be
able to leave messages for friends stored in RFID tags mounted on walls for the same purpose.
These messages could be accessed by a handheld-reader based on a smart-phone using the NFC
standard. One can imagine a creative younger generation having fun with this concept: It could
even lead to a new type of grafti with a more socially acceptable outcome than the property
damage resulting from the conventional form.
9.2.1 Location and Directions
In a city that in the future may deploy RFID tags on signs, street corners, and building placards;
tag IDs can be used to help determine a location or the direction of travel, and are an inexpensive
alternative to a portable Global Positioning System(GPS). Smart phones which already support
enough memory to store digital maps can augment this information with databases of tag IDs
tabulated against location coordinates and lists of nearby businesses. This data can be used to
graphically illustrate your location on a map, and help you locate nearby shops and restaurants.
For example, you could scan an RFID tag on a street corner and then enter a query, Find the
nearest Chinese restaurant. By keeping track of a sequence of tags along the way, the mobile
computer could also determine the direction you are walking, and if incorrect, provide updated
instructions about the route.
9.2.2 Memory and Time
RFID tags with embedded writable memory can be used to log information, but unlike a
computer le system, they do not have the ability to automatically timestamp the data unless the
interrogator itself can provide timing information. This is because a clock requires continuous
P1: JYS
power to keep it operational and once an RFID tag leaves the readers eld, energy is no
longer available. However, a timestamp is important for many applications, allowing data to
be merged and synchronized, and it can be used to prevent data-logs from being falsied. An
example application of writable RFID tags along with interrogator time-stamping, is provided
by the sport of Orienteering, and is described below:
Orienteering
The sport of orienteering, although not well known in the US, has a strong following in Europe,
and is beginning to grow in North America. The sport combines cross-country running with
map and compass-work. At the start of a race, contestants are presented with a detailed map
marked with a route containing way-points (also called controls), the locations of which are only
previously known to the organizers. A runner must visit each control in the order shown on the
map and complete the course in the fastest time in order to win the race. The starts are staggered
to make it less likely runners will followeach other. Traditional orienteering races supply runners
with a course card and use mechanical punches at each of the controls; a runner must punch
the card at each control to prove that he or she has been there. Each punch contains a unique
pattern of pins, which is also not known to the runner, and therefore effectively unforgeable
before the race. A modern orienteering event provides runners with an alternative to the card, a
nger mounted RFID tag (see Fig. 9.2a) that can be inserted into a tag writer housed in a small
battery-powered box at each control (see Fig. 9.2b). As a runner visits each control, the tag
accumulates a set of unique numbers identifying the controls, and the time they were recorded.
This information can be downloaded at the nish to satisfy the race ofcials that the course has
been completed correctly, and to determine the total running time. As with the marathon races
described in Section 5, the split times calculated for intermediate points along the course are
FIGURE9.2: (a) RFID thumb tags. (b) A control in the forest with RFID writer station and position
to insert the thumb tag
P1: JYS
68 RFIDEXPLAINED
also of great interest to the contestants. At the end of the event, the tags memories are wiped
clean, ready for the next race.
9.2.3 Another Use of Time
Even though read/write RFID tags must rely on an interrogator to generate timestamps, it is
still possible to use the parameter of time to police good behavior. For example, interrogator
timestamps can be used to help detect falsied events. Consider two interrogators that create
write events sequentially, each providing their own timestamp. The time recorded for the rst
event constrains the earliest time that can be recorded for the second event. Thus, if the second
interrogator wishes to falsify its log, events that occur both before and after its own recording
provide limits on the time of the forgery.
The value of this technique can be illustrated using a supply chain example. Consider a
tagged case of commercial goods that are being transported by a shipping company between
two cities, and along the route the goods must pass through several checkpoints. However, at
one checkpoint a nefarious operator decides to remove some of the merchandise, and to cover
his tracks tries to falsify the time that the goods were in his hands. An automatically generated
timestamp in the tag would prevent this type of forgery. But, as this is not possible with passive
RFID and because the timestamp must be provided by the interrogator, the nefarious operator
has the opportunity to change his timestamp at will. However, if each event is recorded into a
write-once memory in the tag, the nefarious operator can only claim his event occurs after the
previous write event, and cannot project the time too far into the future, or it might conict with
the checkpoint that follows. RFID tags that create a progressive log of data using an append-
only memory can therefore be used to detect some types of supply chain anomaly without the
need for an active clock in the tag itself.
9.2.4 Facilitating Wireless Connections
In Section 3.4 we described how NFC could be used as a side channel to aid wireless discovery
as well as serve as a communication channel in its own right. Bluetooth and WiFi are standards
for localized wireless communication (WLAN) that can benet from this capability. RFID
memory without any special extensions can also serve as a side channel to aid in communication
by passing auxiliary information. Mobile devices can use this information to decrease association
time and reduce power consumption. For example, if mobile device A would like to connect
to another device B that uses a short-range RFID tag to store its wireless MAC address, it
can discover and connect to it by reading these parameters when brought close by. If several
other devices are present in the locality (C, D, and E), A no longer needs to contend for the
wireless medium in order to discover them all and then decide which one to connect to. A
wireless discovery process is inherently unreliable because a mobile device does not know what
P1: JYS
other devices are actually in the locality, and therefore the value of retrying a discovery request
that had a null result is of limited value. That is to say, the messages may have been lost due
to wireless interference or contention, or there may not have been any other devices present at
all. In practice engineers design a wireless discovery mechanism so that a statistical argument
can be made about the likelihood of a successful discovery, e.g., for the Bluetooth protocol [24]
the typical discovery process is carried out for 10.25 s with a greater than 95% chance that all
nearby devices will be found. For Bluetooth, an RFID side-channel can improve on this long
discovery time, by allowing users to deliberately bring devices together in close proximity in
order to initiate the wireless link. By involving users and making use of unambiguous physical-
proximity, wireless connection time can be shortened and the results accurately reect the user
intention.
Power savings are also possible using an RFID side channel. Radio standards such as
WiFi were not designed with a low power mode that enables them to be discovered without the
radio being turned on. This results in a considerable quiescent power consumption, which for
small mobile devices dramatically shortens battery life. A modied RFID tag can serve a wake-
on-wireless capability [52] for WiFi or other similar radios. This is achieved by building an
electronic switch to turn the WiFi radio on and off, and controlling the switch from a modied
RFID tag. In short, the RFID circuit can be extended to provide an external (wake-up) signal,
the logical state of which is dened by the value of an internal memory register written by
an RFID interrogator. Thus, an RFID interrogator can wake-up a near-by WLAN radio by
writing into the tags memory.
In addition to using RFID to pass MAC addresses and wake-up information, there
are higher levels of interaction that can be usefully communicated through its memory. The
discovery mechanisms described above are specically related to wireless hardware. In the case of
discovering a simple device, such as a wireless printer, it is expected to provide a print service
and nothing else. However, when discovering a general purpose device such as computer, the
list of services available will not be known. In wired networks a service discovery protocol is used
to determine the services provided by a networked computer, examples include UPnP and Jini,
but these mechanisms assume a shared network connection already exists. In the wireless world
a connection must be made before service discovery can be determined. To explain the problem,
consider a Bluetooth enabled PDA (A) that is trying to connect to a wireless music service
running on one of three nearby Bluetooth enabled PDAs (B, C, and D). The three devices are
all capable of advertising their music service using UPnP. By initiating the standard Bluetooth
discovery protocol, A will discover B, C, and D, and it will learn they are PDAs and will nd
the set of proles available (note: the Bluetooth spec uses the term prole to describe a protocol
it supports). Lets assume they all support the PAN prole that allows an IP based protocol to
be established between them. In order for A to determine if a music service is available on B, it
P1: JYS
70 RFIDEXPLAINED
must create a Bluetooth connection to B, establish an IP connection, and then listen for a UPnP
broadcast announcing the music service. If nothing is found, the link must be disconnected
and a new connection made to C and D in turn, until the service has been found. This process
is time consuming, and power inefcient. RFID memory can save time and power by storing
service names and TCP port numbers alongside device types and MAC addresses. If PDA (A)
now reads each of the RFID tags attached to PDA (B, C, and D), in a short period of time it
can decide which device to connect to, and only then incur the overhead of a connection to the
device that can actually provide the music service [53].
9.3 SUMMARY
Through the examples described here it will become apparent that there are numerous applica-
tions for RFID memory, many more than can be explored in this article. However, this section
provides representative examples to illustrate the general scope of memory applications. Further
reading can be found in the cited references.
P1: JYS
71
C H A P T E R 1 0
Challenges, Future Technology,
and Conclusion
There are three main issues that are holding back the widespread adoption of RFID: design, cost
and public acceptance. So much commercial interest has been building around this technology
that adoption is reaching a tipping point and the remaining technical problems are the focus of
much attention. New developments are described in the press on a weekly basis, and progress
towards workable solutions is likely to be swift. Belowwe consider the challenges in more detail.
10.1 CORECHALLENGES
10.1.1 Design
Designing tags and readers so that they guarantee highly reliable identication is not a solved
problem. The solutions must be tolerant of tag orientation, packaging materials, and checkout
congurations that can be found in typical stores. Improved tag antenna design can solve some
of these issues. Tag readers can also be designed to exhibit antenna diversity by multiplexing
their signals between a number of antenna modules mounted orthogonally, or by coordinat-
ing multiple readers. In the latter case, care needs to be taken to avoid what is sometimes
called The Reader Collision Problem [54], as interrogation signals will interfere with each other.
Multiple readers can be used to provide interrogation diversity if a strict time division scheme is
used.
10.1.2 Cost
Pricing plays a critical role in any business decision: Traditional labeling solutions are still
considerably lower cost than any electronic tagging solution. Even though RFID tags are now
available at prices as low as 13 cents each, the market analysts cannot agree what the tipping-
point price might be, and argue among themselves that a 10-cent, 5-cent, or even 1-cent tag
is going to be needed before the market begins to cascade into adoption. Consider a 50-cent
candy barif a 10-cent RFID tag replaces a 0-cent barcode (it can be printed on the wrapper
P1: JYS
72 RFIDEXPLAINED
itself ), then there may not be any remaining prot. As a result RFID tags are likely to have
their rst deployments with high-prot items.
10.1.3 Acceptance
Some of the general reactions that have been levied from the press and civil libertarians were
described earlier. There are genuine concerns here, and it is important that we proceed cautiously
to build in the necessary safeguards to protect us against RFID misuse. In 2003, one author
proposed An RFID Bill of Rights [55] that laid down a set of guidelines that retailers should
adhere to in order to protect the rights of our citizens. At present there are no laws regulating
how tags can be used, and to gain full public acceptance legislation might be required. However,
in the meantime, the early adopters such as Wal-Mart and Tesco (UK) could help defuse the
current concerns by publicly adopting their own Bill of Rights as an open policy.
10.2 ADDITIONAL CHALLENGES FORSHORT-RANGERFID
When describing short-range RFID applications, we often assume it is known where a tag has
been placed on an object, so that a reader can be brought close-by for interrogation. However, an
advantage of RFID is that it can be hidden in packaging and placed behind conventional labels,
and therefore does not spoil the aesthetic appearance of the product. But invisibility also means
that anybody unfamiliar with the product does not know where to look for an attached tag and
thus where to place a reader. In the case of RFID tags that contain memory which may remain
active throughout a products lifetime, this is an important issue to resolve for the consumer as
well as the manufacturer and retailer. However, a standardized solution to this problem has yet
to be found [26]. One possible solution is to establish a convention for RFID placement. For
example, it could always be located behind the manufacturers logo, or in a location, such as
the topside of the product. Alternatively there might be a discreet, but unmistakable, symbol
that is placed in front of where the tag is embedded. In comparison to barcode technology, this
symbol would be much smaller and attract less attention, but still allow a user to visually search
for the symbol in order to position the reader.
10.3 FUTURETECHNOLOGIES
Although conventional silicon chips bonded to spiral copper coils, or UHF dipoles etched
into a copper/acetate substrate, can be cost reduced to the point where they are viable for
the supply chain market, there will be commercial pressure to reduce costs further. However,
ultimately the assembly, or encapsulation processes employed will constrain the minimum price
of a tag. In order to overcome this limitation, Alien Technology Inc., has been experimenting
with self-assembly techniques for joining the tag silicon with the antenna. Although still in an
early stage of development, they have been able to use a uidic assembly process to streamline
P1: JYS
CHALLENGES, FUTURETECHNOLOGY, ANDCONCLUSION 73
FIGURE10.1: Self-assembly techniques being pioneered by Alien Technology Inc.
tag manufacture. To do this, the silicon substrate is formed into an inverted pyramidal shape,
a nano-block (see Fig. 10.1), and then in quantity the nano-blocks are added to a solution
allowing each one to move randomly until it nds a similarly shaped well in an intermediate
substrate (non-silicon). The two pieces slot together, and the resulting combination is held in
place by surface effects. From a manufacturing perspective the new structure is larger, easier to
pick-up, and maneuver into place to bond with the antenna. As a result the cost of assembly
will drop accordingly.
In an even bolder approach to lowering costs, Philips (Eindhoven, Netherlands) has been
experimenting with an all plastic RFID tag [56], see Fig. 10.2a, the raw materials and manu-
facturing process being less costly than silicon. By depositing organic semiconductor materials
directly onto an acetate substrate, it is possible to design and build arbitrary plastic circuits.
Early active devices made from organic polymers were only able to switch at low frequencies
(100 kHz), but recent improvements have enabled a 13.56 MHz RFIDtag to be built entirely
out of plastic, and return a unique 64-bit code to the reader. The tag is made froma plastic called
pentacene, and this material along with smaller dimensions used to build the active devices, has
resulted in the necessary speed improvement. About 2000 transistors are used in the prototype
(Fig. 10.2b). This is an important development because 13.56 MHz is the same frequency used
(a) (b)
FIGURE10.2: (a) Philips experimental set-up for testing a plastic RFID Tag. (courtesy Philips Elec-
tronics, N.V.) (b) A prototype plastic RFID tag up-close (courtesy Philips Electronics, N.V.)
P1: JYS
74 RFIDEXPLAINED
by the near-eld based ISO standards 15 693 and 14 443, and creates an opportunity to use a
large pre-installed reader base to exploit this new lower cost tag. This project has demonstrated
signicant progress, but there are still many problems to overcome before commercial tags can
be mass-produced. For example, the prototype tag shown in Fig.10.2b can only transmit its ID
to a reader over a few millimeters, and was built using a conventional lithographic process. The
range needs to be increased to a few centimeters for effective use; and to be cost competitive,
an ink-jet deposition process must be used to assemble the organic transistors. This will allow
less expensive materials and a scalable manufacturing process to replace todays silicon based
approach. The potential of the ink-jet process has been shown by other groups [57], and
if perfected will enable the antenna to be deposited on a substrate at the same time as the
electronics, thus not requiring any second stage assembly. The results presented by Philips at
ISSCC06 are nonetheless very encouraging and will spur other researchers to work in this area.
10.4 CONCLUSION
RFIDis continuing to make inroads into inventory control systems and it is only a matter of time
before the component costs fall belowa point that, whenweighed against the advantages, make it
an attractive economic proposition. While engineering challenges still exist, there are extensive
developments underway to build tag-reading systems that have sufciently high accuracy to
perform acceptably. There may even be some economic pressure from the larger distributors to
modify product packaging to integrate RFID and improve the read accuracy. At this delicate
stage, while the technology is being trialedby major corporations, media reaction, andoutspoken
privacy groups have the opportunity to inuence the rules by whichthe technology is used. Given
there is nowlegislation in place among most developed countries to protect personal information
held in computers at banks and other commercial organizations, there is no reason why RFID
data management cannot acquire a similar code of conduct [55]. The potential benets of RFID
are enormous, and as long as the use of tag data is handled appropriately, we are certain to see
many novel and surprisingly useful applications of this technology in the future.
P1: OIY
MOBK032-REF MOBK032-Want.cls December 8, 2006 17:59
75
References
[1] Uniform Code Council, Inc., UPCSymbol Specication Manual, Dayton, OH, Reprinted
May 1995.
[2] D. J. Collins and N. N. Whipple, Using Bar CodesWhy Its Taking Over. Data Capture
Institute, ISBN 0-9627406-0-8.w.
[3] C. Wu, Tagged out, Science News, in Taggants: Barcodes for Bombs, Sept. 14, 1996.
[4] K. Finkelzeller, The RFID Handbook, 2nd edn., Wiley, 2003, ISBN 0-470-84402-7.
[5] R. Want, The magic of RFID, ACM Queue Magazine, vol. 2, no. 7, pp. 4148, Oct.
2004.
[6] R. Want, RFID: The key to automating everything, Scientic American, pp. 5665,
Jan. 2004.
[7] Alien Technology, Inc., www.alientechnology.com.
[8] Reed Electronics Group, RFIDtags and chips: Opportunities in the second generation,
Rep. No. IN0502115WT, www.instat.com, Dec. 2005.
[9] R. Want and D. Russell, Ubiquitous electronic tagging, IEEE DS-Online.
[10] D. J. Moore, R. Want, et al., Implementing phicons: Combining computer vision with
infrared technology for interactive physical icons, in Proc. ACMUIST99, Ashville, NC,
pp. 6768, Nov. 810, 1999.
[11] R. Want, A. Hopper, V. Falcao, and J. Gibbons, The active badge location system,
ACM TOIS, vol. 10, no. 1, pp. 91102, Jan. 1992.
[12] J. Paradiso and M. Feldmeier, A compact self-power push button controller, in Ubicomp
2001, Springer Verlag, Berlin, pp. 299304, 2001.
[13] EPCglobal, Inc., www.epcglobalinc.org.
[14] IBM, AssetID, Information Brief, www.pc.ibm.com/ww/assetid/index.html, Nov. 1999.
[15] Indala, (Motorola), BiStatix White Paper, Version 4.1, www.mot.com/LMPS/Indala/
bistatix.htm, March 1999.
[16] Microchip Technology, RFID System Design Guides and Application Notes,
www.microchip.com, 1999.
[17] Trovan RFID, Method and apparatus for modulating and detecting a sub-carrier signal
for an inductively coupled transponder, www.trovan.com, US Patent #5095309.
[18] TIRIS, Tag-it Inlays, Product Bulletin, Texas Instruments, www.tiris.com.
P1: OIY
76 ALECTUREONRADIOFREQUENCY IDENTIFICATION
[19] W. H. Hayt, Jr., Engineering Electromagnetics, 5th edn., McGraw-Hill, New York, ISBN
0-07-027406-1.
[20] G. Moore, VLSI: Some fundamental challenges, IEEE Spectrum, vol. 16, p. 30, 1979.
[21] Specication of Air InterfaceEPCglobal. EPC
TM
Radio-Frequency Identity Proto-
cols Class-1 Generation-2 UHF RFID Protocol for Communications at 860960 MHz
Version 1.0.9, Jan. 2005.
[22] Impinj, Inc., www.impinj.com.
[23] Near Field Communication (NFC) Forum, www.nfc-forum.org.
[24] The Bluetooth Special Interest Group (SIG), www.bluetooth.org.
[25] WiFi IEEE802.11 wireless standard, http://standards.ieee.org/getieee802/802.11.html.
[26] R. Want, K. Fishkin, B. Harrison, and A. Gujar, Bridging real and virtual worlds with
electronic tags, in Proc. ACM SIGCHI, Pittsburgh, pp. 370377, May 1999.
[27] M. Weiser, The computer for the 21st century, Scientic American, vol. 265, no. 3,
pp. 94104, Sept. 1991.
[28] A. Tannenbaum, Computer Networks, Prentice Hall, Englewood cliffs, NJ, 1981.
[29] P. Karn, MACAA new channel access method for packet radio, in Proc. 9th
ARRL/CRRL Amateur Radio Computer Networking Conf., London, Ontario, Canada,
p. 134, Sept. 22, 1990.
[30] M. Alesky, Cyborg 1.0, Wired Magazine, Issue 8.02, pp. 144151, Feb. 2000.
[31] R. Want, Enabling ubiquitous sensing with RFID, IEEE Comput., vol. 37, no. 4,
pp. 8486, April 2004.
[32] Microsensys, Temperature Sensing and Logging, www.microsensys.com/english/emicros
.htm.
[33] K. Opasjumruskit, et al., Self-powered wireless temperature sensors exploit RFID
technology, IEEE Pervasive Comput., pp. 5461, Jan.March 2006.
[34] KSW Microtec, www.ksw-microtec.de.
[35] D. E. Culler and H. Mulder, Smart sensors to network the world, Scientic American,
pp. 8591, June 2004.
[36] M. Philpose, J. R. Smith, B. Jiang, A. Mamishev, S. Roy, andK. Sundara-Rajan, Battery-
free wireless identication and sensing, IEEEPervasive Comput., vol. 4, no. 1, pp. 3745,
Jan.March 2005.doi:10.1109/MPRV.2005.7
[37] J. Smith, et al., A wirelessly powered platform for sensing and computation, In 8th
Int. Conf. Ubiquitous Computing, Orange Country, CA, USA, pp. 495506, Sept. 1721,
2006.
[38] Metro Group Store Future Store Initiative, www.future-store.org.
[39] Internet Retailer, www.InternetRetailer.com Article id=17448.
P1: OIY
REFERENCES 77
[40] C. Legner and F. Thiesse, RFID-based maintenance at Frankfurt airport, IEEE Per-
vasive Comput, vol. 5, no. 1, pp. 3439, Jan.March 2006.doi:10.1109/MPRV.2006.14
[41] M. Philipose, K. Fishkin, D. Patterson, M. Perkowitz, D. Hahnel, D. Fox, and H. Kautz,
Inferring activities from interactions with objects, IEEE Pervasive Comput., vol. 3,
no. 4, 2004.
[42] J. R. Smith, K. P. Fishkin, B. Jiang, A. Mamishev, M. Philipose, A. D. Rea, S. Roy,
and K. Sundar-Rajan, RFID based techniques for human-activity detection, Commun.
ACM, vol. 48, no. 9, pp. 3944, Sept. 2005.doi:10.1145/1081992.1082018
[43] K. Fishkin, M. Philipose, and A. Rea, Hands on RFID: Wireless wearables for detecting
use of objects, in Proc. 9th Int. Sym. of Wearable Computers, ISWC05, Oct. 2005.
[44] F. Thiesse, E. Fleisch, and M. Dierkes, LotTrack: RFID-based process control in the
semiconductor industry, IEEE Pervasive Comput., vol. 5., no. 1., pp. 3439, 4753,
Jan.March 2006.doi:10.1109/MPRV.2006.14
[45] G. Borriello, W. Brunette, M. Hall, C. Hartung, and C. Tangney, Reminding about
tagged objects using passive RFIDs, in 6th Int. Conf. Ubiquitous Comput., Nottingham,
UK, pp. 3653, Sept. 2004.
[46] M. Ohkubo, K. Suzuki, and S. Kinoshita, RFIDprivacy issues and technical challenges,
Commun. ACM, vol. 48, no. 9, pp. 6671, Sept. 2005.doi:10.1145/1081992.1082022
[47] E. Batista, Step back for wireless ID tech? Wired News, http://www.wired.com/news/
wireless/0,1382,58385,00.html, April 8, 2003.
[48] R. Singel, American passports to get chipped, Wired News, www.wired.com/news/
privacy/0,1848,65412,00.html, Oct. 19th, 2004.
[49] A. Juels, R. L. Rivest, and M. Szydlo, The blocker tag: Selective blocking of RFID
tags for consumer privacy, in 8th ACM Conf. Comput. Commun. Security, ACM Press,
pp. 103111, 2003.
[50] P. Hewkin, Smart tagsThe distributed memory revolution, IEEE Rev. (UK), June
1989.
[51] T. Kindberg, et al., People, places and things: Web presence of the real world, ACM
MONET (Mobile Networks and Applications Journal), 2002.
[52] P. Shih, V. Bahl, and M. J. Sinclair, Wake on wireless: An event drive energy saving
strategy for battery operated devices, in Proc. ACM MobiCom 2002.
[53] T. Pering, R. Ballagas, and R. Want, Spontaneous marriages of mobile devices
and interactive spaces, Commun. ACM, vol. 48, no. 9, pp. 5359, Sept. 2005.
doi:10.1145/1081992.1082020
[54] D. W. Engels, The Reader Collision Problem, Auto-ID Center White Paper, MIT-
AUTOID-WH-007, Nov. 1, 2001.
P1: OIY
[55] S. Garnkel, An RFID Bill of Rights, Technol. Rev., p. 35, Oct. 2002.
[56] Philips demonstrates world-rst technical feasibility of 13.56-MHz RFID tags based on
plastic electronics. http://www.research.philips.com/newscenter/archive/2006/060206-
rd.html.
[57] PARC Research, Jet-printed plastic transistorsA solution for the display industry,
http://www.parc.xerox.com/research/projects/lae/plastic.html.
P1: OIY
MOBK032-GLOSS MOBK032-Want.cls December 8, 2006 17:36
79
Glossary
AFI Application Family Identier
ASK Amplitude Shift Keying
BPSK Binary Phase Shift Keying
CRC Cyclic Redundancy Code
CSMA Carrier Sense Multiple Access
DARPA Defense Advance Research Program Agency
ECMA European Computer Manufacturers Association
ECU European Currency Unit
EPC Electronic Product Code
ETSI European Telecommunications Standards Institute
FCC Federal Commission of Communication
HF High Frequency
ID Identication/Identity
ISO International Standards Organization
ISM Industrial Scientic Medical
JINI A Discovery Service created by Sun Microsystems
LSB Least Signicant Bit
LF Low Frequency
MACA Media Access Collision Avoidance
MAC Media Access Control
MSB Most Signicant Bit
NFC Near Field Communication
PC Personal Computer
PSK Phase Shift Keying
RFID Radio Frequency Identication
ROM Read Only Memory
SNR Signal Noise Ratio
UPC Universal Product Code
UHF Ultra High Frequency
UPnP Universal Plug and Play
P1: OIY
MOBK032-GLOSS MOBK032-Want.cls December 8, 2006 17:36
WiFi IEEE 802.11a/b/g Standard
WLAN Wireless Local Area Network
WORM Write-Only, Read-multiple Memory
XOR Exclusive-OR
16RN 16-bit Random Number
P1: OIY
MOBK032-AU MOBK032-Want.cls December 8, 2006 17:35
81
Author Biography
Roy Want is a Principal Engineer at Intel Research. Interests include embedded systems,
mobile computing and automatic identication. Want received a BA in computer science from
Cambridge University, UK in 1983 and earned a Ph.D. in distributed multimedia-systems in
1988.
He joined Xerox PARCs Ubiquitous Computing program in 1991 and managed the
Embedded Systems area, later earning the title of Principal Scientist. He joined Intel Research
in 2000.
Want is the author of more than 50 publications in the area of mobile and distributed
systems; and holds 52 patents. He is a Fellow of both the IEEE and ACM.
P1: OIY
82
P1: OIY
83
Figure Acknowledgments
Figures 1.1, Altek Instruments Ltd, BarcodeMan, Walton on Thames. UK
Figure 1.5, data in table extracted fromRFIDTags andChips: Opportunities inthe 2ndGeneration,
Report # IN0502115WT, Publisher, Reed Business
Figure 5.1, HID Global Corporation, Irvine, CA 92618 U.S.A www.hidcorp.com
Figure 5.2, Inspec Tech, Inc. Valley Head, Alabama www.inspectech.us
Figure 5.5b, The Augusta Chronicle, and the Boston Athletic Association
Figure 5.6a, North Dakota State University, Extension Service, Fargo, North Dakota
Figure 5.6b, Pubaa Animal Clinic, Segamat, Malaysia
Figure 5.9a, Metropolitan Transportation Commission Oakland, California www.mtc.ca.gov
Figure 6.1 and 6.2, KSW Microtec AG, Dresden, Germany, www.ksw-microtec.de
Figures 6.3, 10.2a, 10.2b, Philips Electronics, N.V., Eindhoven, The Netherlands
Figure 7.1, METRO Group Future Store Initiative, METRO AG, www.future-store.org
Figure 7.3, from IEEE, Pervasive Computing, Volume 5, No. 1
Figure 7.5, University of Washington, Seattle, Washington
Figure 8.2a, 8.2b, Notags.co.uk, UKCitizens against the pervasive use of RFIDin our Society
Figure 9.2a, Centre for Orienteering History, www.orienteering-history.info
Figure 9.2b, Deeside Orienteering Club, UK, www.deeside-orienteering-club.org.uk
Figure 10.1a, 10.1b, 10.1c, Alien Technology Corporation, www.alientechnology.com
P1: OIY
84

RFID Explained PDF

Uploaded by

Copyright:

Available Formats

RFID Explained PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

RFID Explained PDF

Uploaded by

Copyright:

Available Formats

P1: IML/FFX P2: IML/FFX QC: IML/FFX T1: IML

MOBK032-FM MOBK032-Want.cls December 8, 2006 17:36

access control: keys and immobilizers;

patrol verication: process management;

supply chain: warehousing and inventory control;

assets: airline luggage, equipment, and cargo.

pharmaceuticals: packaged drugs;

transportation: auto-tolls: FasTrak, EZ-pass;

ticketing: ski passes;

credit/debit cards: PayPass by MasterCard.

smart toys: interactive characters.

You might also like