DESIGN AND CHARACTERIZATION OF MINIATURIZED UHF TAG ANTENNAS

DESIGN AND CHARACTERIZATION OF MINIATURIZED UHF TAG ANTENNAS

BONG FWEE LEONG

DOCTOR OF PHILOSOPHY IN ENGINEERING

LEE KONG CHIAN FACULTY OF ENGINEERING AND SCIENCE

UNIVERSITI TUNKU ABDUL RAHMAN

NOVEMBER 2017

DESIGN AND CHARACTERIZATION OF MINIATURIZED UHF TAG ANTENNAS

By

BONG FWEE LEONG

A thesis submitted to the Institute of Postgraduate Studies and Research, Lee Kong Chian Faculty of Engineering and Science,

Universiti Tunku Abdul Rahman,

in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Engineering

November 2017

ii ABSTRACT

DESIGN AND CHARACTERIZATION OF MINIATURIZED UHF TAG ANTENNAS

Bong Fwee Leong

In this thesis, four projects have been conducted for studying tag miniaturization techniques in the UHF band. First, a flexible folded-patch is proposed for a metal- mountable tag antenna. It is formed by connecting a serrated patch to a high- impedance loop-shaped stub, where the serration and the stub are used to fine- tune and scale down the resonant frequency. The second project proposes a compact coin-shaped folded dipole for designing a universal UHF tag. It is composed of a matching loop, which is encircled by two patch-shaped radiating arms, within an embedded structure that has high compactness. Slots are etched in the radiators for fine-tuning the resonant frequency. Highly inductive matching loop is included to conjugate match the capacitive radiating element and the microchip. A miniaturized dipolar patch antenna is proposed in the third project.

It consists of a dipolar patch which is split by a meandered slot into two halves.

The slot is highly reactance and it can be used for scaling the self-resonant frequency of the tag as the first effort to shrink down the antenna structure. In the design, the patches are stub-shorted to ground, where highly inductive thin stubs are used to further miniaturizing the antenna by lowering the tag resonant frequency. It should be mentioned that the thin stubs can provide frequency fine- tuning on site. Finally, a small orientation insensitive metal-mountable tag is proposed to overcome the tag misplacement issue in the actual applications. It

iii

consists of two pairs of orthogonal dipolar patches placed in such the way that their inherent blind spots and null points are removed, making the tag antenna readable in all directions in the boresight. Multiple high-inductive stubs are inserted into the patches and they are shorted to ground for increasing the tuning range of the tag resonant frequency.

iv

ACKNOWLEDGEMENTS

First and foremost, I would like to acknowledge the invaluable suggestions, guidance and advices from Dr. Lim Eng Hock and Dr. Lo Fook Loong throughout the entire period of my research projects. Countless of conversations and discussions have been done to stimulate new ideas for the development of these new devices. In addition, they were very supportive and always ready to give consultation anytime and anywhere.

Special thanks is extended to UTAR for providing lab facilities that needed for measurements and simulations during the research. Moreover, UTAR also gave me freedom to access online database such as ProQuest and IEEE Xplore where I could get all the important literatures.

My deepest thankfulness goes to my family for their persistent love and support throughout my research. This thesis would be impossible without them.

I am indebted to my wife Hui Lin for her understanding, encouragement, quiet patience, and unwavering love are undeniably the bedrock which make me here today. I thank my parents and parents in law for their faith in me and allowing me to be as ambitious as I desired.

v

APPROVAL SHEET

This thesis entitled ‘DESIGN AND CHARACTERIZATION OF MINIATURIZED UHF TAG ANTENNAS’ was prepared by BONG FWEE LEONG and submitted as partial fulfillment of the requirement for the degree of Doctor of Philosophy in Engineering at Universiti Tunku Abdul Rahman.

Approved by:

___________________________

(Assoc. Prof. Dr. Lim Eng Hock) Date:………..

Supervisor

Department of Electrical and Electronic Engineering Lee Kong Chian Faculty of Engineering and Science Universiti Tunku Abdul Rahman

___________________________

(Assoc. Prof. Dr. Lo Fook Loong) Date:………..

Co-supervisor

Department of Electrical and Electronic Engineering Lee Kong Chian Faculty of Engineering and Science Universiti Tunku Abdul Rahman

vi

LEE KONG CHIAN FACULTY OF ENGINEERING AND SCIENCE UNIVERSITI TUNKU ABDUL RAHMAN

Date: 1 November 2017

SUBMISSION OF THESIS

It is hereby certified that BONG FWEE LEONG (ID No: 13UED07152) has completed this thesis entitled “DESIGN AND CHARACTERIZATION OF MINIATURIZED UHF TAG ANTENNAS” under the supervision of Dr. Lim Eng Hock (Supervisor) from the Department of Electrical and Electronic Engineering, Lee Kong Chian Faculty of Engineering and Science (FES), and Dr.

Lo Fook Loong (Co-Supervisor) from the Department of Electrical and Electronic Engineering, Lee Kong Chian Faculty of Engineering and Science (FES).

I understand that University will upload softcopy of my thesis in pdf format into UTAR Institutional Repository, which may be made accessible to UTAR community and public.

Yours truly,

(BONG FWEE LEONG)

vii

DECLARATION

I hereby declare that the thesis is based on my original work except for citations and quotations which have been duly acknowledged. I also declare that it has not been previously and concurrently submitted for any other degree or award at UTAR or other institutions.

(BONG FWEE LEONG)

Date: 1 November 2017

viii

TABLE OF CONTENTS

Page

ABSTRACT ii

ACKNOWLEDGEMENTS iv

APPROVAL SHEET v

PERMISSION SHEET vi

DECLARATION vii

LIST OF TABLES xi

LIST OF FIGURES xii

CHAPTER

1 INTRODUCTION 1

1.1 Background and Issues 1

1.2 Current and Future Markets for Passive RFID Tag 7

1.3 Challenges in UHF RFID 10

1.4 Research Objectives and Motivation 12

1.5 Publications 15

1.6 Thesis Overview 16

2 PASSIVE UHF TAG DESIGN AND CHARACTERIZATION 18

2.1 Introduction 18

2.2 Impedance Matching Techniques for UHF Tag Antennas 19

2.2.1 T-Match Method 19

2.2.2 Inductively-Coupled Method 21

2.2.3 Nested-Slot Method 22

2.2.4 Microstrip Line Coupling Method 23

2.3 Tag Performance Criteria 24

ix

2.4 Typical Tag Design Requirements 27

2.5 Tag Design Process 28

2.6 Tag Antenna Characterization 32

2.6.1 Read Pattern Measurement 37

3 FLEXIBLE FOLDED-PATCH ANTENNA WITH SERRATED EDGES FOR METAL-MOUNTABLE UHF RFID TAG 39

3.1 Introduction 39

3.2 Configuration and Equivalent Circuit 40

3.3 Measurement Setup 46

3.4 Results and Discussion 48

3.5 Conclusion 55

4 COMPACT FOLDED DIPOLE WITH EMBEDDED MATCHING LOOP FOR UNIVERSAL TAG APPLICATIONS 56

4.1 Introduction 56

4.2 Configuration and Equivalent Circuit 59

4.3 Measurement Setup 64

4.4 Results and Discussion 65

4.5 Conclusion 81

5 MINIATURIZED DIPOLAR PATCH UHF TAG ANTENNA WITH NARROW MEANDERED SLOTLINE 83

5.1 Introduction 83

5.2 Configuration and Equivalent Circuit 86

5.3 Results and Discussion 97

5.4 Conclusion 109

6 COMPACT ORIENTATION INSENSITIVE DIPOLAR PATCH FOR UHF METAL TAG 111 6.1 Introduction 111

6.2 Configuration and Equivalent Circuit 114

6.3 Results and Discussion 122

x

6.4 Conclusion 136

7 SUMMARY AND DISCUSSION 137

BIBLIOGRAPHY 140

APPENDICES 151

xi

LIST OF TABLES

Table Page

1.1 Comparison between the barcode and RFID

technologies. 5

3.1 Comparing the performances of several UHF-band metal-mountable tag antennas that are made of patch

resonator. 55

4.1 Comparing the performances of several universal UHF tag antennas that can be used for the metallic and non-

metallic objects. 81

5.1 Comparing the performances of the UHF metal-

mountable tags. 106

6.1 Comparing the performances of the UHF metal-

mountable tags. 136

xii

LIST OF FIGURES

Figure Page

1.1 The use of backscatter radiation to communicate

with the ground. 1

1.2 Basic components of a typical RFID system. 4

1.3 UHF RFID frequencies by countries. 6

1.4 2012–2018 UHF Market Unit Growth—Retail vs.

Rest of Market (‘Other’) 8

2.1 The T-match configuration and its corresponding

equivalent circuit. 20

2.2 The inductively-coupled configuration and its

corresponding equivalent circuit. 21

2.3 The nested-slot configuration and its corresponding

equivalent circuit. 23

2.4 The microstrip line coupling feed configuration. 24 2.5 The typical impedance characteristics of tag antenna,

where the Ra, Xa, Rc , Xc and fc are the antenna resistance, antenna reactance, chip resistance, chip

reactance and tag resonant respectively. 25 2.6 The typical performance chart of tag antenna, where

( )

r₀ = λ/4π /

. 26

2.7 Passive UHF RFID tag antenna design process. 29 2.8 A simplified passive UHF RFID measurement

system setup, in which the reader antenna is stayed at fixed location and the tag can be rotated around its

axes. 33

2.9 Experimental setup for read pattern measurement. 38 3.1 Configuration of the proposed tag antenna with

serrated edges. 42

3.2 Tag antenna with an integrated chip. 42

xiii

3.3 Equivalent circuit of the tag antenna. (LLS = 25.2nH, LLr = 0.0945nH, CLr = 0.0413pF, Ra = 4.0 KΩ, La

= 9.7nH, and Ca = 3.19pF). 44

3.4 Effects of b on the patch inductance and capacitance.

Other parameters remain unchanged. 46

3.5 Tag Measurement Setup in Anechoic Chamber 47 3.6 Input impedance measurement using a balun probe 47 3.7 Simulated, measured, and modelled input impedance

of the tag antenna. 48

3.8 Simulated power transfer coefficient as a function of

b. 49

3.9 Measured and simulated realized gain (Ptag) and the measured tag sensitivity (Gr) in the direction of θ =

40o and φ = 0o. 50

3.10 (a) Surface current distribution, (b) electric field

distribution at 920MHz. 51

3.11 Measured read range. 52

3.12 Measured read range for different (a) plate length W, (b) plate width L in the direction of θ = 40o in the xz

plane. 53

3.13 (a) NXP reference materials. (b) Measured read

range for the NXP reference materials. 54 4.1 (a) Top and front views. (b) anatomical view of the

proposed UHF tag antenna. 61

4.2 Prototype of the proposed UHF tag antenna. 62 4.3 Equivalent circuit of the tag antenna. (Rm = 0.14 Ω,

Lm = 6.12 nH, Cm = 0.21 pF, Ra = 1.8 Ω, La = 50.5

nH, and Ca = 0.5 pF). 63

4.4 Simulated and modelled input impedances of the tag

antenna. 64

4.5 Tag measurement setup in an anechoic chamber. 64

xiv

4.6 Simulated antenna impedance and measured tag sensitivity when the tag antenna is placed. (a) on a Styrofoam block. (b) at the center of a 20 cm × 20

cm metal plate. 67

4.7 Electric field distributions when the tag antenna is placed. (a) in free space. (b) on a 20 cm × 20 cm

metal plate. 68

4.8 Surface current distributions on the proposed tag antenna when it is placed. (a) in free space. (b) on the

20cm × 20cm metal plate. 69

4.9 Realized gains and read distances which are measured when the tag antenna is placed on the Styrofoam (free space) and on the metal plate (20 cm

× 20 cm). 71

4.10 Measured radiation patterns when the the tag antenna is tested in free space and on a 20 cm × 20 cm metal

plate. (a) xy plane. (b) xz plane. (c) yz plane. 73 4.11 Changes in the input impedance by (a) increasing b3

and b4 simultaneously. (b) reducing f2. and (c)

decreasing a2. 75

4.12 Simulated matching charts of the proposed tag antenna in the range of (a) (0.6 mm ≤ b3 ≤ 2.2 mm and 1.9 mm ≤ b4 ≤ 3.5 mm) and (b) (6.6 mm ≤ a3 ≤

9.0 mm and 1.4 mm ≤ a4 ≤ 3.8 mm) at 915 MHz. 76 4.13 (a) NXP reference materials. (b) Read distances

when the proposed tag is tested using the NXP

reference materials. 78

4.14 (a) The proposed tag antenna is tested on 7 different metallic containers. (b) The proposed tag antenna is tested on 7 different non-metallic objects. (c) Measured read distances when the tag antenna is tested on the metallic containers and non-metallic

objects. 79

5.1 (a) Configuration of the proposed tag antenna with design parameters L1= 23, W1 = 16, G = 22.9, a = 18, b = 0.3, c = 1.6, d1 = 0.9, d2 = 3.7, e = 4.07, f = 2.4, g = 0.3, h = 2.1, i = 0.3, j = 0.415 and k = 11.35

(all in mm). (b) Naked inlay. 87

5.2 The proposed tag antenna with integrated chip. 87

xv

5.3 Equivalent circuit of the tag antenna. (Rp = 1.2 KΩ, Cp = 1.08pF, Lp = 4.2nH, Rs = 0.093Ω, Ls = 3.92nH,

and Cm = 0.834pF). 89

5.4 Simulated and modelled input impedances of the

proposed tag antenna. 90

5.5 Simulated power transmission coefficient for

different thin stub lengths d2. 92

5.6 Effects of the (a) meandered slot length lds and, (b) gap b on the tag reactance. (c) gap b on the realized

gain. Other parameters remain unchanged. 93 5.7 Effects of the f on the tag reactances. 94 5.8 (a) Folded dipolar patch with a straight slot. Design

parameters are lds = W1 = 16 mm, b = 0.3mm, d1 = 0.9mm, L1= 23mm, W1 = 16mm, G = 22.9mm. (b) Input impedances for the cases of straight slot and meandered slot. (c) Input impedances for the cases with/without thin stub. (d) Input impedances for the

cases with/without U-shaped slots. 96

5.9 Measured and simulated realized gains when the tag

antenna is placed on a copper plate of 20cm × 20cm 98 5.10 Surface current distribution at the resonant frequency

of 910MHz. 99

5.11 Measured read distance in the xz, yz and xy planes. 100 5.12 Measured read ranges for different plate sizes when

(a) varying L, (b) varying W. 102

5.13 (a) The NXP reference materials. (b) Read ranges when the proposed tag is tested using the NXP

reference materials. 103

5.14 (a) Metallic objects (b) Read ranges on the metallic objects (c) Read ranges with “Autotune” function

enabled. 105

5.15 (a) Antenna structures designed using the techniques in (Yang et al., 2011; Svanda and Polivka, 2015;

Calabrese and Marrocco, 2008) for comparison. (b)

Realized gains for the six cases. 109

xvi

6.1 (a) Configuration of the proposed tag antenna with design parameters L1= 30, W1 = 30, G1 = 29, a = 0.3, b = 0.49, c = 0.31, e = 7.75, f = 0.5, g = 14.85, h =

14.85, t = 1.6 (all in mm). (b) Naked inlay. 116 6.2 The proposed tag antenna with an integrated chip. 116 6.3 Input impedances at the two ports of the proposed

two-port dipolar patch antenna. Also shown is the input impedance for the single-port isolated dipolar

patch antenna with the same dimension. 117 6.4 Equivalent circuit of the dipolar patch antenna. (Ra

= 0.6 KΩ, Ca = 1.24pF, La = 2.1nH, Rst = 0.38Ω,

Lst = 15.5nH, and Cm = 0.32pF). 119

6.5 Simulated and modelled input impedances of the

proposed tag antenna. 120

6.6 Effects of (a) the tuning stub length e (b) the cross- slot gap a and (c) the stub width b on the antenna

impedance. Other parameters remain unchanged. 122 6.7 (a) Measurement setup in the anechoic chamber. (b)

Orientation of the tag antenna. 124

6.8 Measured and simulated realized gains when the tag

antenna is placed on a copper plate of 20cm × 20cm 125 6.9 Surface current distributions when (a) only the dipole

pair (Patch A – Patch B) is turned on, (b) only the dipole pair (Patch C – Patch D) is turned on, and (c) both the dipole pairs are excited simultaneously, at

the simulated resonant frequency of 924MHz. 127 6.10 Measured tag sensitivities in the xy-plane along the

z-direction when the dipolar patches are excited

individually and simultaneously. 128

6.11 Measured read distances in the (a) xy plane, (b) xz plane, (c) yz plane in free space at 895MHz and on

the 20cm x20cm metal plate at 918MHz. 131 6.12 Measured read ranges for different plate sizes when

varying L and W. 133

6.13 (a) The NXP reference materials. (b) Read ranges when the proposed tag is tested using the NXP

reference materials. 134

xvii

6.14 (a) Metallic objects (b) Read ranges on different

metallic objects. 135

A.1 Detailed dimensions of the loop-shaped stub. 152

B.1 ABCD matrix for the π network. 153

CHAPTER 1

INTRODUCTION

1.1 Background and Issues

The history of radio frequency identification (RFID) technology can be traced back to War World II (Dobkin, 2008). Germans discovered that the reflected radio signals would change when pilots were rolling their planes as they were reflected back to ground. At that time, the radar crews on the ground used this rough method to identify if the incoming planes belonged to the Germans, and this was also considered as the first passive RFID system.

Figure 1.1: The use of backscattered radiation for communication with the ground station.

The Scottish physicist Sir Robert Alexander Watson-Watt (Mark R., 2005) headed a secret project to develop the first active system to identify friend or foe (IFF). Engineers installed such a transmitter on each of British aircraft.

When radio signals detected by the radar stations on the ground, planes which

2

were installed with transmitters would broadcast signals back to indicate the planes whether they were friendly. An active RFID system works in the similar way. The advancement of RF communications systems continued through 1950s and 1960s, and many scientists actively involved in doing researches to investigate the uses of RF waves to identify objects remotely. In January 23, 1973, Mario W. Cardullo (Mark R., 2005) received the first U.S. patent in active RFID tag with writable memory. Meanwhile, a patent for passive transponder used for keyless door was also awarded to a California entrepreneur, Charles Walton. In the disclosure, an access card with RFID antenna embedded in the vicinity was used to access a reader near a door. Once the reader detected a valid identity stored in the card, the door would be unlocked. In 1970s, the U.S. government requested Los Alamos national laboratory to develop a system for tracking nuclear materials. They had proposed a concept to affix transponders on trucks so that readers located at suitable gates would be able to secure the loaded facilities. The reader at the gate worked by transmitting RF energy to wake up the transponder which in turn responded to the reader with identification data.

The working concept was further enhanced and modified into the automated toll payment systems and it was commercialized in the mid 1980s. Such systems have been widely used nowadays on roads, bridges, and tunnels around the world.

With the advantages of long read range (up to 20 feet) and fast data transfer, the IBM engineers developed and patented an ultra-high frequency RFID system in the early 1990s. However, IBM sold its patents to Intermec during financial crisis in the mid-1990s. The RFID technology was then broadly deployed in various fields such as warehouse tracking and farming, but

3

unfortunately it failed to attract the interests from intended users due to high setup costs. There was little progress in the RFID technology development from years 1980 to 1998. The technology was not in good demand due to high material costs in tag fabrications, expensive system setup, and the lack of international standards, which had hindered the worldwide adoption of this technology.

UHF RFID development started to boost again in 1999, when the Uniform Code Council and the European Article Number (EAN) International funded Massachusetts Institute of Technology (Mark R., 2005) to establish an Auto-ID Center. Two professors, David Brock and Sanjay Sarma, had been assigned to explore the possibility to affix low-cost tags on all finished products to track them through the supply chain. The Auto-ID Center started to develop two air interface protocols (class 0 and 1), the Electronic Product Code (EPC) numbering scheme, and the network architecture which had defined the physical and logical requirements of the interrogator and the tag of the passive RFID system. It also defined the physical and logical requirements of the RFID system from 1999 to 2003. The EPCglobal was then established to commercialize the EPC technology. Nowadays, many of the best-known companies in the world, including Google, Mitsubishi, Wal-Mart, Siemens, General Electric, Coca-Cola, and Virgin, have already broadly employed the RFID technology to track their goods. Particularly, healthcare and pharmaceuticals players have also widely adopted the RFID technology, and the usage of the RFID tags are expected to rise rapidly from $94.6 million in 2009 to $1.43 billion in 2019 (RFID For Healthcare, 2009). The EPCglobal adopted EPC Gen 2 Class 1 UHF standard (Gen 2 ISO18000-6C) after getting approval from the International Standards

4

Organization (ISO) in January 2005. Today, most of the RFID measurement instruments are developed based on this standard. There are many types of RFID microchips such as Monza 4, Monza 5, and Monza R6 from Impinj, lnc. as well as Ucode 7, Ucode 7XM, and Ucode G2iL from NXP, and etc.. Each of the chips has its own read/write sensitivities, impedance characteristics, and memory capacities. Figure 1.2 shows the configuration of a contemporary RFID system consisting of a tag and a reader (also called the interrogator) that is connected to a workstation for information storage and retrieval. Communication between the reader and tag operates wirelessly using electromagnetic waves. Through multiple exchanges of commands between the reader and the tag, the RFID reader can identify the information stored in that particular tag.

Figure 1.2: Basic components of a contemporary RFID system

Ethernet

RFID Reader

RFID Tag RF Antenna Network Workstation

5

Table 1.1: Comparison of the barcode and the RFID technologies.

Barcode System RFID System

1. Line of sight is needed for the scanner to read the tag.

Reader does not require line of sight to read the tag.

2. The scanner can only read one tag at a time.

Multiple tags can be easily read simultaneously

3. The physical tag must be in clean condition, free from dust, so that it is readable.

Wireless detection is not affected by dirty environment.

4. A barcode can only do simple classification such as telling the type of item.

RFID tag contains much detailed information such as the dates and contents of a specific item.

5. Information in a barcode is permanent.

RFID tag information can be easily altered anytime.

The main application of the RFID is for automated identification which is aimed at replacing the role of the conventional optical identification technique - barcode. To visualize the advantages that the RFID has brought, the author has included a comparison table for the RFID and the barcode technologies, as shown in Table 1.1, for quick reference (RFID Vs Barcode, 2012). As depicted in the table, the barcode tag must be located in the line of sight of scanner to read but this is not a requirement for the RFID where radio wave is emitted from the reader to activate multiple passive tags in its readable region. Since only one barcode can be read at any one time, the inventory process can be very time- consuming if it involves a large data pool. In contrast, the RFID reader can read multiple tags within less than a second and the data can be obtained simultaneously. Also, it is not unusual to encounter a situation that a barcode tag cannot be read because it is scratched or dirty. This doesn’t happen to the RFID tags where the physical conditions where they are placed are not critical. In the barcode system, a single type of commercial item can only be assigned with one barcode. This is not convenient if there are multiple items of the same type are

6

to be registered. On the other hand, the RFID tagging system allows each of the same items to be assigned with a unique identification number electronically.

Barcodes are usually printed a small piece of white paper and the printed information is unchangeable. On the other hand, the information on an RFID tag is electronically stored and changes can easily be done.

In view of the benefits stated above, implementing the RFID technology will surely bring enormous changes to the retailers on the way they work, especially on item-level tagging. The main advantage of item-level tagging is that it allows rapid checkout and automatic inventory control in retail industries such as hypermarkets and department stores. It helps to improve the efficiency of the supply chain from manufacturers to consumers, preventing excessive stock inventory.

Figure 1.3: Regulated UHF RFID spectrums by countries.

7

The regulated operating spectrums for the UHF RFID are different among different countries in the world, as shown in Figure 1.3 (UHF RFID By Country, 2014). Not a single frequency band is optimally applicable across the world. The European Union adopts a UHF bandwidth ranges from 865 MHz to 868 MHz with the interrogator’s maximum allowable transmitted power of 2W ERP (Effective Radiated Power). While the UHF spectrum bandwidth in the North Americas is from 902 MHz to 928 MHz, and the maximum permitted power for the reader to operate is 4W EIRP. Australia has restricted the UHF RFID technology to operate in the frequency range of 920 MHz to 926 MHz.

Meanwhile, China has allocated two different spectrum ranges for the UHF RFID, namely 840 MHz - 845 MHz and 920 MHz - 925 MHz. The users have to select the tags that work well in the designated frequency bands in their own countries.

1.2 Current and Future Markets for Passive RFID Tag

In recent years, the passive UHF RFID technology has been undergoing rapid grow rate with high production volume, benefiting from reduction in manufacturing cost and maturity in semiconductor technology. According to ChainLink Research 2015, a total of 6.2 billion tags sold in the market, showing a steady growth from 4.9 billion in 2014 and indicating a nearly 27% grow rate in the volume. About 70% of the tags were consumed by the retail industry, which still occupies the largest market segment in passive UHF RFID tag applications nowadays (State, 2015).

Many retailers have successfully made use of RFID tags to lower their operation costs and improve the inventory visibility. Since so far only a small

8

portion of apparel items are being tagged, and Asian market has just started to adopt RFID tags for registering their retail goods, the use of RFID tagging technology is expected to experience a much steeper growth in the short future as many more potential retailers are expected to take part. As illustrated in Figure 1.4, the usage of tags in the retail industries will grow for another 34% in the next two years, which makes up of 29% in the overall tag compound growth rate.

Figure 1.4: 2012–2018 UHF Market Unit Growth—Retail vs. Rest of Market (‘Other’)

Apart from the retail market, other industries are also experiencing high growth in using RFIDs recently. As shown in the IDTechEx research, RFID tags can be used on animals such as pig, cattle, and sheep. Since pets can get infected with different diseases easily, as a result, many developed countries (RFID Forecast, 2016) have legally required tagging to be made compulsory in livestock for minimizing the possible epidemic outbreak. This sector has consumed hundreds and millions of tags in 2016. Substantial growth is also seen in other market segments such as healthcare, transportation, asset tracking/management, access control, and automotive. The latest emergences of wireless technologies for the instance sensor networks and the internet of things (IOT) have opened up new opportunities for the UHF RFID tag applications. Additional sensing

9

features have been successfully integrated into various RFID tags since a couple of years back. The advancement of UHF tags as sensing devices has broadened the horizon of the RFID tags for many new applications, ranging from military surveillance and sophisticated tracking systems, disaster warning, home automation to medical management and tele-health. Sensing features are usually incorporated into the microchip of the tag antenna. For instance, Smartrac N.V.

has recently released a passive sensor tag, called sensor Tadpole, for automotive industries especially for car manufacturing lines. The sensor-tag is to pinpoint the location of water leakage inside the car body and it works as a capacitive tuned circuit to detect changes in humidity. The working principle is that changes of air moisture causes the tag input impedance of the microchip to vary, and the tag microchip will then translate it into a sensor code. Finally, the RFID reader will interpret the moisture levels from the obtained sensor code (Water Leakage, 2016).

Development of UHF tags has been undergoing significant improvement in pricing and performance in the past decade. Tags with various sizes and performances have been explored extensively in recent years. Tags with unit price as low as 5¢ are commercially available now pertaining to the persistent efforts made by chip, tag, and materials designers, as well as the low-cost manufacturing plants in Asia, especially in China. As continuously lowering tags price is not a good option for future strategy, many tag manufacturers have started to look for alternatives such as providing full system integration to get rid of this dilemma.

10

The future of RFID market is indeed very promising, as more and more new users are joining the rank. As found in the IDTechEx research, the total RFID market in 2015 worth $10.1 billion, which has seen increases from 6.3%

and 14.8% in 2014 and 2013, respectively. Rapid developments are seen in all the types of tags, readers and software/services. Particularly, the IDTechEx has forecasted that the total market value of RFID is expected to rise to $13.2 billion in 2020.

1.3 Challenges in UHF RFID

The RFID technology has been growing fast recently. However, the RFID engineers have to face many new challenges. The latest challenges involves high tag failure rates in mass production due to issues such as microchip misplacement and antenna etching inconsistency. Quality and reliability of the RFID implementation rely on many factors such as tag size, reader antenna size, tag orientation, tag overlapping, effects of metallic substances in the vicinity of tag and so forth. Since the RFID tag is so critical in determining the success and failure of an RFID system implementation, the RFID tag designers need to equip themselves with sound understanding of the RFID working principles such as the microchip sensitivity, the backscattering performance, the impedance matching, the desired operating frequencies (such as FCC, ETSI or both included), the objects to be tagged, and the environments the tagging are taking places.

Tradeoffs between the tag size and read performance are always to be considered in the design process as well.

11

Dipole is the most popular type of antenna in the UHF RFID (Retailers, 2016) industry. But, unfortunately the tag performance degrades significantly when the tag is brought close to metallic surface (Ukkonen et al., 2004; Yao et al., 2011; Gao and Yuen, 2011), and the tag performance deteriorates further if it is required to have low profile and small in size. This is an emerging research area which has attracted many researchers’ since a decade ago. Many new concepts have been proposed to tackle the design problems. In the early days, some scholars introduced a thick dielectric substrate to separate the tag from the metal surface (Kanan and Azizi, 2009). Inverted-F structures (Ukkonen et al., 2004; Hirvonen et al., 2004; Chen and Tsao, 2010) and patch antennas (Cho et al., 2008) were then suggested where a ground layer was inserted to isolate the radiating patch from the attaching conductive object. The shortcomings of these antenna structures include large size, high profile, rigidness, and high cost.

The object or item, where the tag is to be attached to, is the first factor to be considered before initiating the tag design. When a specific tag antenna is applied on an object with dielectric constant different from what it was designed for, the resonant frequency will get detuned, eventually leading to impedance mismatch and performance loss (Xi and Ye, 2011; Girbau, 2010). Different materials have different absorption and reflection characteristics to radio waves, which can affect the read range of a tag significantly. Therefore, tag antenna design is usually object sensitive and application specific. In recent years, the application of item-level tagging technology has been growing rapidly in retail industry. As of now, the retail industry is making up the largest market segment in the applications of the passive UHF RFID tags (State, 2015). Since many of

12

the item-level consumable goods can come in different shapes and sizes. A small- sized tag is more preferable as it can be easily attached to the surface. However, in practice, there is always trade-off in between miniaturizing antenna size and maximizing radiation performance. The larger the antenna, usually, the better the tag sensitivity and the stronger the backscattered fields returning to the reader.

As a rule of thumb, a larger tag has better performance with higher system detection reliability. Nevertheless, reducing tag size while maintaining tag performance is one of the major challenges in the RFID implementations.

1.4 Research Objectives and Motivation

The motivation of this research is to design miniaturized UHF tag antennas for mounting on metallic surface. This thesis is to achieve three research objectives.

The first objective is to explore the use of different techniques such as serrated edges, loop-shaped stub, matching loop, and meandered slot for designing and tuning a couple of metal-mountable tag antennas. For all these projects, the commercially available flexible polyethylene (PE) foam will be employed as tag’s substrate to separate the patch radiator from the metal surface. It does not require the use of rigid printed-circuit-boards (PCB) and expensive ceramic substrates (Babar et al., 2012). The second research objective is to explore different miniaturization methods such as folding, embedding, and using of high- impedance stubs for designing compact UHF tag antennas. Folding method can reduce the tag size significantly and it does not require the use of metallic vias.

This helps to reduce manufacturing complexity. Embedding the matching loop and meandered slot into the folded-patches are found to be able to reduce the footprint significantly. Use of high-impedance stubs will be shown useful for

13

scaling down the tag resonance for achieving high compactness. The final objective is to design highly compact UHF tag antennas using the flexible thin PET (polyethylene terephthalate) substrate, which is commercially available.

PET is flexible and it can be used for designing a miniaturized tag antenna that can be stuck on a small surface. However, it will be shown that warping and deformation can cause variation in the impedance performance, and careful handling is needed.

To meet all the research objectives and motivation, four projects have been conducted. In the first project, a new UHF tag structure consisting of a folded-patch with serration along its edges and a vertical loop-shaped stub is demonstrated. For the first time, the loop-shaped stub is introduced to drastically scale down the tag resonant frequency, resulting in much smaller structure. The use of serration technique as an effective fine-tuning mechanism on the folded patch is studied in detail. An equivalent-circuit-based procedure has been proposed to estimate the resonant frequency of the proposed tag.

In the second project, a coin-shaped folded dipole is proposed for designing a universal tag for the first time. The matching loop can be embedded into the radiator structure for achieving miniaturization structurally. The dipole arms are folded to form a compact structure that can be used for designing a universal tag antenna. Also, slots are introduced into the dipolar radiator for tuning the impedance of the proposed tag antenna. It is worth mentioning the resonant frequency remains almost constant when the tag applied to both metallic and dielectric surfaces, fulfilling the stringent requirement of the universal tag.

14

The third project is devoted to studying the use of meandered slot for miniaturizing dipolar patch antenna. To demonstrate the design idea, a highly inductive narrow meandered slotline is tactfully incorporated into a folded dipolar patch, splitting it into two halves, for miniaturizing the antenna size and tuning the resonant frequency. Parametric studies show that the narrow meandered slot does not affect the characteristics of the folded dipolar patch much. Also, thin stubs and the U-shaped slots are included to further miniaturize the proposed tag and to provide tuning mechanisms for the tag resonant frequency.

Lastly, the fourth project is to design a compact tag antenna, which consists of two pairs of dipolar patches, for achieving orientation insensitive on metal. The two dipolar patches are tactfully combined in orthogonal for eliminating their nulls points. They are then fed by balanced feed ports at the center for generating radiation patterns which are orientation insensitive. Also, a long inductive tuning stub is integrated into each of the patches for miniaturizing the antenna size and tuning the tag resonant frequency.

All of the proposed tag can be directly imprinted using the PET (polyethylene terephthalate) substrate and it does not require drilling and milling processes. To demonstrate the validity of the design ideas, prototypes have been fabricated and measured. Complete parametric analysis has been conducted to analyze the effects of the design parameters.

15 1.5 Publications

In this thesis, four tag antennas have been designed and evaluated. These novel structures have exhibited good performance-size ratio characteristics for the metal-mountable applications. The contributions of thesis are summed up in the following technical publications:

1) Bong, F. L., Lim, E. H., and Lo, F. L., 2017. Flexible Folded-Patch Antenna with Serrated Edges for Metal-Mountable UHF RFID Tag. IEEE Transactions on Antennas and Propagation. 65(2), pp. 873-877.

2) Bong, F. L., Lim, E. H., and Lo, F. L., 2017. Compact folded dipole with embedded matching loop for universal tag applications. IEEE Transactions on Antennas and Propagation. 65(5), pp. 2173-2181.

3) Bong, F. L., Lim, E. H., and Lo, F. L., 2017. Miniaturized dipolar patch antenna with meandered slotline for UHF tag. IEEE Transactions on Antennas and Propagation. (accepted as a full paper on 28Jun17).

4) Bong, F. L., Lim, E. H., and Lo, F. L., 2017. Compact orientation insensitive dipolar patch for UHF metal tag. IEEE Transactions on Antennas and Propagation. (under review).

16 1.6 Thesis Overview

This thesis consists of 7 chapters. The first chapter is initiated with discussions on the backgrounds and issues of the conventional RFID systems. The author has briefly shared the current and future market trends as well as the challenges the RFID is facing in the real implementation scenario. The research objectives and motivation are clearly explained in the later part of the same chapter.

Chapter 2 mainly focuses on the tag design and measurement methodology. Since the quality of impedance matching between the tag antenna and the microchip is critical, it is first discussed in this chapter. Here, four types of the most commonly applied matching techniques are studied. The key tag performance indicators such as gain and sensitivity are described. This is followed by discussing the typical tag design requirements. Here, the tag design process is elaborated step-by-step with the use of the design flow chart. Lastly, the tag measurement methods and the derivation for each of the performance parameters are discussed comprehensively.

Chapter 3 reports on a novel small-sized folded-patch tag antenna for metal mountable. It has demonstrated how to design a folded tag antenna by wrapping a flexible inlay around a soft dielectric foam block. The benefits of the serration and the highly inductive loop-shaped stub are discussed in detail. They are used for fine-tuning the resonant frequency and scaling down the self- resonant frequency of the proposed tag antenna. In this chapter, the derivation steps of equivalent circuit are explained in detail. Simulation and experiment have been conducted to characterize the radiation performance.

17

Chapter 4 focuses on the design of a coin-shaped folded dipole UHF tag that can be placed in free space and on metal. The use of embedded matching loop to achieve high compactness is presented. It also shows that the slots can be used for fine-tuning the resonant frequency. The simulation and experiment data have been analyzed to understand the characteristics of the proposed tag antenna.

Also, it is also found that the resonant frequency of the proposed tag antenna is stable and not affected much by its backing object.

A compact dipolar patch that is designed for mounting on metal is discussed in Chapter 5. It shows that meandered slot can help for miniaturizing the tag structure, without significantly affecting the tag characteristics. This chapter also introduces the use of inductive thin stubs for shorting the patches to ground, where the stubs can again be used for frequency tuning. Also, a simple equivalent circuit has been derived for studying the tag impedance characteristics.

Chapter 6 presents a compact orientation insensitive patch for metal- mountable tag. It can be seen that two identical dipolar patches which are placed orthogonally can remove the null points created by a single dipolar patch, resulting in orientation insensitive radiation patterns. In the design, both ends of each of the patches are stub-shorted to ground, where the highly inductive thin stubs are used for size miniaturization and frequency fine-tuning.

Chapter 7 gives a summary of the work and reports conclusions. It also discusses the significance of my works to the RFID fields.

18 CHAPTER 2

PASSIVE UHF TAG DESIGN AND CHARACTERIZATION

2.1 Introduction

A tag is an information-carrying device attached to an object to be identified. The performance of the tag plays a major role in determining the efficiency of the entire RFID system. As the RFID systems are getting more and more popular, end users are pushing toward the use of UHF tags which have higher operating frequencies, high data rate, long read range, low profile and small form factor as compared to LF and HF band RFID system.

Since the past decade, many types of antenna have been proposed for designing various RFID tags in the UHF band. The dipole structures such as folded dipoles and meandered dipoles are widely used by many commercial applications since they can be easily printed on thin and low-cost PET films.

However, the main problem with these types of antennas is that, when they are moved close to the metallic surfaces, the antenna performance will degrade drastically due to changes in the antenna impedances. Since the UHF RFID tag is a passive device where it contains no power source, it is important to ensure maximum power transfer between the microchip and the tag antenna for generating an optimum backscattered signal. To achieve this condition, a good impedance match is desired in between the microchip and the tag antenna.

Designing a passive antenna which can match well to the complex chip impedance is one of the most challenging tasks in tag design. This is because the

19

microchip has inherently high quality factor, with low series resistance and large capacitive reactance. Furthermore, the tag impedance changes when it is mounted on different objects, and a tag antenna has to be designed to be readable in various platforms without severe degradation in performance. In order to obtain a stable tag performance, minimizing the loading effects caused by the backing objects is very important. Design and characterization of the passive UHF tag antenna will be discussed in details in this chapter.

2.2 Impedance Matching Techniques for UHF Tag Antennas

The RFID microchip is an energy storage device and most of the commercially available RFID microchips exhibit highly capacitive reactance roughly ranging from -100 Ω to 400 Ω (Marrocco, 2008). On the other hand, the resistance in series is about an order-of-magnitude smaller, usually in the range of 5 Ω to 20 Ω. Since the chip impedance is capacitive, the antenna impedance has to be

inductive so as to achieve good conjugate matching. Since the RFID tags must be manufactured with low cost, it is not feasible to use lumped components to form external matching circuits in the tag antenna design. Usually, the matching mechanisms have to be formed within the antenna structure itself. Below are the brief discussions of the four types of impedance matching techniques that are widely deployed by the RFID tags.

2.2.1 T-Match Method

To introduce tunable input impedance to the planar dipole of length l, with reference to Figure 2.1, a centered short-circuit stub can be incorporated with a

20

second dipole of length a to form the T-match structure. The antenna source is connected to the center of the second dipole with a distance b from the first or larger dipole. The planar dipole T-match configuration and its equivalent circuit are shown in Figure 2.1. The T-match acts as an impedance transformer.

Referring to (Marrocco, 2008), the input impedance Zin at the chip bonding pads is given by,

A t

A t

in

Z Z

Z

Z Z ₂

2

) 1 ( 2

) 1 ( 2

α α + +

= + , (2.1)

where Zt= jZ⁰tanka/2 is the input impedance of the short-circuited stub together with it second short dipole, and Z0 denotes the characteristic impedance of the transmission line. With the T-match condition removed, the dipole impedance taken at its center will become ZA. The term

α = ln( b / r

) ln( b / r

)

represents current division factor between two conductors, and r_e =0.25wand

' '

8 . 25 w

r

=

Figure 2.1: The T-match configuration and its corresponding equivalent circuit.

l l

b a/2

w' w

2Z_t Z^A Zin

(1+ ) : 1

α

21 2.2.2 Inductively Coupled Method

The inductively-coupled technique consists of a feed loop with two terminals as chip bonding pads as well as a long dipole placed in the vicinity, as shown in Figure 2.2. The communication mechanism between the feed loop and the radiator is through mutual coupling. The corresponding equivalent circuit is shown in Figure 2.2.

Figure 2.2: The inductively-coupled configuration and its corresponding equivalent circuit.

With reference to Figure 2.2, the components of the equivalent circuit are further elaborated. The resistances of the feed loop and the radiating body are represented by Rloop and Rrb, respectively. Meanwhile, the Lloop andM denote the self-inductance and mutual-inductance of the feed loop. The elements Rp and Cp

are used to represent the effects of substrate. Assuming that the effects of substrate are negligible, the real Rzin and imaginary Xzin parts of the antenna input impedance (Marrocco, 2008) can be predicted by,

loop rb

zin R

R M

R = f +

2

0 )

2 ( π

(2.2)

l a/2

w'

w

IC RFID Antenna M R

C

R

L

L

C

R

22 X_in = 2

π

As can be seen from the two impedance elements above, the resistance is mainly affected by M and Rrb, and the reactance is controlled by Lloop. This implies the resistance and reactance can be adjusted independently. A design strategy can be used to match the antenna to an arbitrary microchip with complex impedance effectively. The impedance matching procedure can be started by selecting a proper loop size to neutralize the capacitive chip reactance. The next step is to vary the loop-dipole distance to match the chip resistance.

2.2.3 Nested-Slot Method

A useful impedance matching strategy for large planar dipoles or suspended patches (Marrocco, 2008) is through nested-slot structure shown in Figure 2.3.

The slot profile is able to transform the antenna impedance, where each tooth provides energy storage and radiation. In this structure, the tag antenna gain is mainly determined by the patch length l; while varying the slot dimensions a and b will change the tag impedance.

23

Figure 2.3: The nested-slot configuration and its corresponding equivalent circuit.

2.2.4 Microstrip Line Coupling Method

A microstrip line is located at a distance of d from the microstrip patch, and the line is short-circuited at both ends. The microchip feed point is at the center of the microstrip line, forming a pair of equal-length strips as illustrated in the Figure 2.4 (Tseng et al., 2012). The two microstrip lines are used to excite the radiating patch through electromagnetic coupling. To achieve maximum power transfer, the impedances of the chip and the tag antenna must be matched conjugately. The input reactance of the antenna is contributed by the pair of short-circuited microstrip lines. The length L of microstrip line (either side of the mictrostrip pair) can be linked to the antenna reactance by,

 



 

=



0 1

0 . 5

2 tan Z

L X

π λ

, (2.3)

a b

g

b g/2 g/2 b

Z

l

24

Where Xant and λ are the expected inductive reactance and the guided operating wavelength of the antenna. As for the input resistance, it can be adjusted by controlling the distance d between the radiating patch and the microstrip line.

Figure 2.4: The microstrip line coupling feed configuration.

2.3 Tag Performance Criteria

There are several key performance indicators for the passive UHF tag antenna such as tag sensitivity, read distance, and radiation patterns. The tag sensitivity Ptag is the most commonly used parameter for describing the tag performance. It is the minimum power that is required to turn on the tag, and sometimes it is also called power-on-tag-forward. The read distance Rmax is the maximum detectable distance between the reader antenna and the tag under test. With the use of the Friis equation, the Rmax can be calculated by using the maximum reader

Substrate h

S^x

g S^y Chip

D

Shorting Pin

Shorting Pin d

t

Metal Plate

25

transmitting power (locally regulated radiation power) and the received power of the tag. Such measurement method is able to provide a good approximate read distance under certain controlled environment, and it is broadly used by the RFID community as a site-independent benchmark (Virtanen et al., 2013; Bjorninen et al., 2013). On top of that, the tag’s radiated power can also be characterized by its radiation patterns, which is a graphical representation of the radiation properties around the antenna under test. It provides information on the best read angle of a tag antenna. The derivations of these three key performance indicators, together with other useful parameters such as realized gain and power transfer coefficient, will be discussed in details in section 2.6 later.

Figure 2.5: The typical impedance characteristics of a tag antenna, where Ra, Xa, Rc , Xc and fc are the antenna resistance, antenna reactance, chip resistance, chip reactance, and tag resonant, respectively.

The typical characteristics of the antenna impedance, microchip impedance, and read distance of a tag can be described as functions of frequency as illustrated in Figure 2.5 (Rao et al., 2005). The pure antenna impedance characteristic can be obtained by using single-port, (Z11) simulation or

ImpedanceRange

Minimum required range

Peak range

Tag resonant

Antenna self-resonant

R^a

R^c X^a

-X^c

Bandwidth f^c

26

measurement, by applying the 50 Ω measuring port across the feeding pads. The peak of the Ra curve is the antenna’s self-resonance. The interception point of the Xc and Xa curves, which corresponds to the peak read distance, denotes the tag’s resonance. Meanwhile, the bandwidth of the tag can be defined as the frequency band at which the tag can meet the minimum read range requirement.

Figure 2.6: The typical performance chart of a tag antenna, where

( )

A tag antenna with perfect match, small footprint, low profile, and good read range is always highly desirable, but unfortunately, it is unrealizable in many practical cases. Designing a passive UHF tag antenna involves unavoidable compromises between gain, impedance matching, antenna size, and bandwidth.

The scholars in (Rao et al., 2005) presented a tag performance chart for helping tag designers to estimate the tradeoffs between the impedance matching and

Poor Tag

Better Match

Better Gain

Good Tag

Power Transmission Coefficient

G ai n (d B i)

27

antenna gain. Figure 2.6 shows that several gain-power transmission combinations are possible for the same read range.

2.4 Typical Tag Design Requirements

Designing a tag always start with good understanding of the design requirements.

Since multiple factors can come together to affect the tag’s performance, a designer should obtains details of the actual applications so that a more robust tag can be designed. The key design requirements are summarized as follows:

• Regulated frequency and power: The regulated spectrums and maximum powers for the commercial UHF RFID operations can differ from one country to another. For example, the approved frequency band and operating power for the United States are 902 MHz – 928 MHz and 4W EIRP, China are 920.5 MHz -924.5 MHz and 2W ERP, and Europe are 865.6 MHz -867.6 MHz and 2W ERP. The operating frequency of the tag must be designed to conform to the approved spectrum and power of the country it is intended for.

• Tag reading performance: In the commercial RFID design, the potential customers will always specify the minimum read distance and radiation patterns of the tag. The antenna designer can then make use of these requirements to select the right type of antenna structures.

• Tag dimension or size. The larger the tag antenna, the better it is to receive signal from the reader, and the stronger the backscattered power. However, it is also very desirable to have a compact tag to make it fixable to a small object.

• Material properties of tagged objects: No tag will be applied in free space.

A different backing object will introduce different loading effects to the tag antenna due to the change of dielectric constants. This will cause frequency

28

detuning and performance attenuation on the tag. Therefore, the proposed tag has to be fine-tuned such that it works excellently on the intended tagged object.

• Tag orientation: Read range is dependent on the tag orientation. Best read range can usually only be obtained in the boresight of the tag. In certain applications, orientation insensitive tags may be required for detection in all angles.

• Tag conductor: The most commonly used conductors in tag design are copper, aluminum, and silver. Among all, silver shows the best electrical conductivity, following by copper and aluminum. However, aluminum is widely used commercially due to its low price and reasonably good RF performance.

• Tag substrate: Many applications require the tag antennas to have low profile and to be easily mounted/embedded on any objects. In the commercial market, many tag antennas are usually made by depositing a thin layer of copper or aluminum onto the top surface of a piece of flexible thin paper, transparent film, polyethylene terephthalate (PET), or fabric, each of which has its own dielectric constant and tangent loss.

• Costing factor: RFID tags must be low in cost. The allowable per-unit-cost will affect the choices in materials and the tag’s configuration.

2.5 UHF Tag Design Process

The microchip’s read and write sensitivities, together with its intrinsic complex impedance, strongly affect the tag antenna’s performance. Tag sensitivity and

29

radiation pattern must be closely monitored during the design process to ensure all the design requirements are met. Since the tag antenna’s footprint, impedance matching quality, and operation frequency impose constraints on the maximum attainable gain and bandwidth (Rao et al., 2005), tradeoffs must always be made for optimizing its performances. Usually, a small antenna structure coming with some convenient tuning mechanisms are preferred for tag design. This is to provide tuning possibility as a tag is subject to tolerances during the fabrication processes. More importantly, frequency tuning flexibility is required for adjusting the tag resonance so that it is useable for different backing objects.

Figure 2.7: Design process of the passive UHF tag.

Select the application

Understand the design requirements and application environments

Select antenna structure

Perform impedance matching, parametric study and optimization

Design requirements met ?

Design is ready Build and measure prototype

Define tag requirements

30

The flow chart in Figure 2.7 illustrates the design process of a passive tag antenna, which will be further elaborated here. It provides a basic guideline for designing a new working tag. Once a certain application is identified, a designer can first study the design requirements as discussed in Section 2.4. A tag antenna can be strongly influenced by its surroundings. For example, a container filled with water or a metallic jug that is placed closely to a tag is going to affect the performance of the RFID system (read range and read rate).

However, the properties of metal can be properly made use for enhancing the read range if the correct antenna type is selected.

All of the aforementioned design requirements and the potential effects coming from the surroundings are then translated into the tag requirements. Choices of materials are the first to come in the actual commercial design as they determine the tag size and costs directly. A microchip with appropriate read and write sensitivities (given in the datasheets) is then selected to match the input impedance of the antenna to the microchip for maximizing the power transfer.

The next step is to select a proper antenna structure that can generate the desired radiation patterns and read range. The tag antenna structure must be able to work well in the designated surroundings. If the tagged items are metallic packaged, antennas such as microstrip patch, PIFA, and slot antennas are more appropriate. On the other hand, thin planar dipole structures are more suitable for designing tags that are targeted for dielectrics.

31

With the use of electromagnetic simulation tools such as CST Microwave Studio and Ansys HFSS, the selected antenna structure will be put through the design steps such as impedance matching, parametric study, and performance optimization until all the tag requirements are fulfilled. Like most of the RF circuit designs, the tag performances are too complicated to be estimated by using the commonly seen analytical solutions. Therefore, application of fast numerical electromagnetic computation tools is crucial for efficient tag design.

In the design process, electromagnetic software will be used to set up a simulation model which mimics the actual environments. In this thesis, the CST Microwave Design Studio is selected as it provides a simple setup template specifically for designing the UHF RFID tag antennas. The boundary condition is set to be a quarter wavelength from the metal plate edges for far-field simulation, and the absorptive boundary is assumed to be no reflection, which has been verified by the CST technical expert and set to be a default setting. The simulation is done with power sweeping from 800 MHz to 1000 MHz where the frequency-domain solver and tetrahedral mesh are employed. To imitate the actual measurement, the simulation model of the proposed tag antenna is also placed at the center of a piece of 20 cm × 20 cm copper plate. A 50-Ω discrete port is first applied across the bonding pads and, subsequently, the port is normalized to complex impedance using the post-processing procedure of the software. Important design parameters such as read range and realized gain can be directly extracted from the electromagnetic software. During the full-wave simulation stage, the tag’s gain, read range, impedance matching quality, and radiation patterns will be closely monitored. The parameters will provide the designer a good understanding of the tag’s characteristics.

32

In the final stage of the design process, the antenna prototype is built based on the optimized dimensions generated by the simulations. All of the key performance parameters must be evaluated and the tag design is considered successful if all the requirements are met. Otherwise, the antenna structure has to be modified or even changed, and further optimizations will be carried out until all the tag requirements are met.

2.6 Tag Antenna Characterization

There are two methods to characterize the tag performance. The first method is done by keeping the transmitted power Ptx a constant and varying the tag distance d from the reader in the direction normal to the reader antenna. This method is often used in the realistic environment; however, the accuracy of measured data is questionable as the measurement method is exposed to multipath phenomena (Nikitin et al., 2012). The second method is to fix the distance d between the reader and the tag while the transmitted power Ptx is varied. When performing this procedure, the transmitted power is varied gradually until the tag is activated.

Such measurement method is able to provide a good approximate read distance under certain controlled environment, and it is broadly used by the RFID community as a site-independent benchmark (Virtanen et al., 2013; Bjorninen et al., 2013).

33

Figure 2.8: A simplified passive UHF RFID measurement system, where the reader antenna is fixed at a location and the tag can be rotated around its axes.

In order to have insight into the second measurement method, let’s carefully evaluate the generic setup of the UHF RFID measurement system, as shown in Figure 2.8. The reader antenna and tag are placed face-to-face on a straight horizontal plane with a distance at least one wavelength away to ensure far field condition. During the measurement process, the position of the reader antenna is kept stationary, while the tag is allowed to rotate about the axis according to the spherical coordinate system illustrated in same figure. The tag antenna parameters are assumed to be a function of (θ, φ).

According to Friis’s transmission equation, the forward link power on the tag microchip Pchip (θ, φ) (Collela et al., 2016) can be calculated as,

Pchip (θ, φ) = P