HF RFID Reader for Mouse Identification - Study of

Magnetic Coupling between Multi-Antennas and a

Ferrite Transponder

C. Ripoll

, P. Poulichet

, E. Colin

, C. Maréchal

and A. Moretto

Université de Paris-Est, Esiee, Esycom

Cité Descartes, BP99, 93162 Noisy le Grand Cedex

Esigetel- Département de Télécommunications

1 rue du port de Valvins 77215 Avon-Fontainebleau

Abstract. This paper depicts an optimized RFID system operating at 13.56

MHz, used to recognize a mouse. When it passes near an antenna, the small

transponder (1x6 mm

) placed in its body communicates with the antenna. The

distance of communication is small (around 3 centimeters) and our objective

was to increase this distance. The transponder receives the signal from a coil

wound around a magnetic circuit. The characteristics of the ferrite are very

important for the communication. All the elements of the chain are taken into

account in ADS simulation and we determine the value of the minimum voltage

necessary for remote bias. Finite-element analysis is employed to extract the

values of the generated magnetic field. The paper shows how to correlate the

influent parameters to increase the communication distance. To improve this

reading rate, a novel differential receiving antenna has been designed allowing

an improved decoupling with the close transmitting antenna. An electrical

model of the different parts has been performed. This includes the tag, the

reader antennas and the coupling between them. These models have been

validated by measurements and used for the simulations of the complete

system.

1 Introduction

The principle used to identify a mouse is depicted on the figure 1. The biocompatible

transponder of 1 mm in diameter and 6 mm in length is introduced under the skin of

the mouse with a syringe. When the mouse passes in the vicinity of the transmitting

Antenna, the transponder receives the signal from the antenna. The amplitude of the

signal at 13.56 MHz has to be sufficient to supply the transponder and to transmit

coded signal. It reacts by sending a coded signal in order to identify the mouse.

The inductive coupling RFID (Radio Frequency Identification) systems operating

at 13.56 MHz are present nowadays in a large variety of applications, like access

control, mass transport, e-purse and secured ID cards. In a conventional system, they

are basically composed of one or more transponders dialoging with a base station by

means of a coupling through the close-field magnetic induction. The base station

produces a magnetic field that supplies power to the transponder, interrogates the tag

Ripoll C., Poulichet P., Colin E., Maréchal C. and Moretto A. (2008).

HF RFID Reader for Mouse Identiﬁcation - Study of Magnetic Coupling between Multi-Antennas and a Ferrite Transponder.

In Proceedings of the 2nd International Workshop on RFID Technology - Concepts, Applications, Challenges, pages 32-42

DOI: 10.5220/0001728900320042

 SciTePress

and decodes the information received.

Fig. 1. Communication between the mouse and the Emitter antenna.

The two parts communicate in half-duplex mode: during the uplink, the base station

sends a request (ASK-modulated, 100% or 10% modulation index, depending on

radio interface type [1]) then it continues to transmit an unmodulated 13.56 MHz

carrier, which supplies power to the transponder IC; during the downlink, the

transponder responds by varying its load impedance, and thus modulating the carrier

sent by the base station (847.5 kHz OOK or BPSK-modulated subcarrier, depending

on radio interface type [1], switches between the impedance states), mechanism called

load modulation. The carrier will thus be modulated in amplitude (AM) and phase

(PM).

Current state-of-the-art in RF characterization of 13.56 MHz inductive coupling

identification systems only involves a basic check of the standard compliance [1], [2].

Although reference transponder circuits are used, no electrical parameter

measurement is performed.

In biomedical domain, where inductive links are used to deliver power and

exchange information with implanted stimulators, results have been published [3] –

[5], regarding the remote power supply in an inductive coupling system and its

variation when the coupling changes. Concerning the measurement methods, a

detailed analysis of coupled resonators is performed by Kajfez in [6], [7]. Majority of

proximity coupling applications concern the magnetic resonance imaging and a very

few for the RFID applications.

The RFID system can be modeled as a double-tuned transformer, whose primary

represents the base station and secondary the transponder. The transponder, often

called tag, is made out of an IC (integrated circuit) which capacitive input impedance

resonates with an antenna coil that collects the magnetic flux.

Proximity Radio Frequency Identification (RFID) systems physically relies on the

space evolution of their loop antennas mutual coupling, i.e. their shape, dimensions

and relative position. In order to optimize the antennas design, one needs to predict

the coupling evolution within the desired operating zone. Previous works in

biomedical domain focused on the coupling variation, mainly for circular loops.

2 System Description and Design Keypoints

A well-designed RFID system should ensure a safe functioning within a precise

geometric volume (i.e. voltage supply to the transponder IC and modulation depth

compliant with the manufacturer's specifications).

Three entities can be distinguished, whose interdependence should be understood: the

base station (i.e. antenna topology and detection threshold), the RF channel (i.e.

magnetic radiation pattern) and the transponder (i.e. IC and antenna).

When the transponder moves in the field of the base station, the induced voltage at the

IC input varies over a large range [4] (e.g. the magnetic field, varying between 1.5

and 7.5 A/m for systems compliant with [1], may induce a voltage of 5 to 35 V at the

IC). Moreover, the logic part of the IC needs a regulated voltage supply and a current

source, to avoid parasitic load modulation during its functioning. In these conditions,

a voltage-and-current regulation system varies the input capacitance and load

resistance of the IC. Thus, the overall functioning passes through a wide range of

states, given that the load mismatch represented by the transponder on the base station

antenna varies.

The aim of an RFID system used in biological experiments for identifying laboratory

mice or small animals is to be able to read the ID code in the largest volume zone

with the smallest transponder as possible to avoid stress and so unpredictable

behavior of the animal. For this reason, the transponder used here is an ultra small

transponder measuring only 1 by 6 mm encapsulated in a glass capsule. The system

operates in the world-wide ISM band of 13.56 MHz and not the classical 134 kHz

band used for animals, so as to reduce the size of the transponder antenna. With such

a tiny small transponder, it becomes compulsory to use a coil antenna wounded on a

ferrite rod to increase the induced voltage.

Before introducing the key points of this study, let us recall that the system operates

with the same minimal constraints as in any conventional proximity passive RFID

system. This means that we should take into account the remote biasing of the

transponder but as well the data exchange between the reader and the tag in a Listen

Before Talk protocol. In our case, to remote bias the transponder, first we must

radiate a high magnetic field, which means that a high current circulates through the

transmitting coil and second, the Q factor of the transponder should be optimized to

be rather high but not too high to avoid degrading the backscatter modulation bands

placed at +/- 848 kHz. These constraints call for a precise modeling of the transponder

parameters and particularly the Q factor. We note that in contrary to a smart card

reading system, we can here neglect the variations of the transfer function (i.e. band-

pass shape, resonance and Q-factor) as the coupling varies [3], [6].

Unfortunately, modeling such a tiny rod is theoretically unreliable due to the

difficulty to measure the magnetic losses, which are expressed by the µ

’’ of the

material. Consequently, the modeling will be based on the measurements of a

resonant transponder.

Another key point in this study was to avoid the coupling between transmitter and

receiver. In the transmitter path, very high voltage exists due to the matching circuits

and to the high current as we already mentioned. To reduce dramatically this problem,

we used a differential antenna on the receiver side. This design allows the perfect

cancellation of the carrier transmitted by the transmitting antenna when the coils are

ideally balanced. As a drawback for this geometrical configuration, we should

mention that the data exchange becomes impossible for a central position of the

mouse. This is of little importance because the mouse ID code has the opportunity to

be read at many points when it enters the bean-shaped reading zone. Obviously, the

coupling is finite and the topology of the receiver should take this into account.

This work is organized as follows. Section 3 presents the magnetic simulations of the

transmitting antenna. It shows the influence of the permeability of the transponder.

Section 4 describes the electrical modeling of the three parts (transponder, transmitter

and receiver). In Section 5, we perform electrical simulations of the complete system.

3 Magnetic Simulations of the Antennas

The figure 2 shows the shape of the transmitting antenna. It is realized with a round

PCB of diameter of 3 cm. The outside circular trace of each side of the PCB is used to

generate the transmitting magnetic field. The two inner spirals that constitute the

differential antenna used for the reception are connected together and one point is

connected to the ground.

Fig. 2. Shape of the transmitting antenna and differential antenna.

The simulation is operating using 2 dimensions FEM (Finite Element Modeling) with

Ansys [8]. Figure 3 represents the model used for FEM. It represent a cut along the

axis A-A of the figure 2. The permeability µ

’ (120 for ferrite of the transponder), the

resistivity σ of the materials and the current in the conductors are taken into account

in order to determine the magnetic field in magneto dynamic simulations.

Fig. 3. Geometry for the FEM.

A measurement of the inductance of the transmitting antenna shows that the spiral

inductance has no effect on the magnetic field so they are not taken into account in

the simulation.

Figure 4 shows the module of the magnetic induction (B) determined in axis-

symmetric simulation. The high permeability of the transponder concentrates the

magnetic field generated by the current in the transmitting antenna. This permeability

is a key point design because a high value increases the distance of reception.

However, it is difficult to obtain a high value at 13.6 MHz without loss.

We determine the magnetic induction (B) at distances d of 0.5 cm, 1 cm, 2cm and 3

cm. The curve is represented on the figure 5.

0 0.005 0.01 0.015 0.02 0.025 0.03

x 10

-4

distance x (m)

B (T)

Induction B(T) obtained by FEM as a function of distance

0.5 c m

1 cm

2 cm

3 cm

radius

Fig. 4. B values near the transmitting

antenna.

Fig. 5. Magnetic induction B (T) with respec

to distance d on a parallel line for x=0.

VGMOS

Vout VDMOS

Vrx

Eqn

Var

MU T I N D MUT I ND

R=1

L=1 uH

V_AC

SRC7

Freq=freq

Vac=polar(6,0) V

R=0.3

L=390 nH

R11

R=1 MOhm

C26

C=22 pF

R10

R=820 Ohm

C25

C=100 nF

R=120 Ohm

R=37 Ohm

L=22 uH

sp_sms_BF998_1_19920901

SNP1

Frequency ="{0.05 - 2.00} GHz"

Bias="Mosf et: Vds=5V Vg2s=3.5V Id=10mA"

C24

C=12 pF

R=5k Ohm

L=1 uH

C23

C=100 pF

R=1

L=1 uH

Fig. 6. Differential receiver schematic.

When the distance d increases, the magnetic induction B is reduced. When the

distance x is higher than the radius of the transmitting antenna, the induction B is

reduced.

4 Characterization Receiver, Transmitter and Transponder

To be able to optimize the whole system, we have to study the individual components.

4.1 Measurements of the Parameters of the Receiver

The electrical circuit of the receiver is shown in figure 6. Its transfer function is

simulated by injecting a signal through a non resonant generator inductively coupled

(L9 transmitting antenna) to the two coils representing the differential spiral antennas

in figure 2 and L5-L6 in figure 6. The received signal is transmitted through a band-

pass (C23-L7-C24) to the MOS transistor used to amplify the signal. The circuit is

designed so that its transfer function is centered on 14.4 MHz frequency that

corresponds to one of the symmetric backscattered bands around the 13.56 MHz

carrier. We have to mention that the parallel resistance R8 allows the setting of the

bandwidth, so is influent for the filtering of the carrier residual and will have an

impact on the distortion of the modulated bandwidth.

4.2 Measurements of the Parameters of the Transmitter

The schematic of the transmitter is represented in figure 5. M2 is a MOS transistor

operating in a low loss switching mode. It generates a high current circulating in the

transmitting antenna L1. C13, C22//C21//C12 and L1 constitute a parallel resonant

circuit. C21 or Ctx is a variable capacitor adjusted to facilitate the remote bias of the

transponder. Frequency analysis has to be taken into account as a first step to adjust

the resonant frequency. Furthermore, the shape of the voltage across the transmitting

antenna has to respect a standard waveform (see 4.2.2).

4.2.1 Frequency Analysis

The transfer function is simulated with a sinusoidal excitation, then we derive for

which value of Ctx, we obtain a maximized current (I_Probe) in the transmitting coil.

The resonant frequency has to be set up at 13.56 MHz to induce the necessary voltage

at the input of the IC tag. This is achieved through the use of the Ctx parallel

capacitor.

Self 3.3uH / 0.5 Ohm mesure

Vtx

V_DC

SRC4

Vdc=12 V

VAR

VAR5

RF_freq=13.56 MHz

CTX=70e-12

Eqn

Var

R=0 Ohm

R=0. 5

L=3.3 uH

R=1.3

L=390 nH

C12

C=47 pF

C21

C=CTX

C22

C=150 pF

ap_nms_IRFD020_19930601

C13

C=100 pF

Fig. 7. Transmitter schematic. Fig. 8. Variation of the Transmitter transfe

function with variable Ctx values.

As shown in figure 8, the transfer function is a band-pass type which quality

coefficient has to be around 30 to avoid degradation of the 200 kHz modulation

bandwidth around the carrier.

Figure 8 represents the shift of the central frequency as a function of Ctx. We note

that it is preferable for the transmitter not to operate at 13.56 MHz (Ctx=70pF) but

rather near this frequency. For example, we obtained best results for a value of 14.4

MHz (Ctx=35pF).

freq=

mag(Vtx)=40.897

CTX=6.000000E-11

13.50MHz

12 14 16 1810 20

0.5

1.0

0.0

1.5

freq, MHz

mag(I_ant.i)

mag(Vtx)

freq=

mag(Vtx)=40.897

CTX=6.000000E-11

13.50MHz

4.2.2 Temporal Analysis

We should consider the time domain response because, to operate properly, any

Mifare UltraLight IC at the tag should be able to detect the presence of a pause in the

transmitted signal. This means that the matching / filter circuit should respect the

waveform shown in figure 9.

Fig. 9. Standard waveform and Measured voltages at and near resonance (Ctx=35pF and 70

pF).

The table 1 shows that times t2 and t4 depend largely of the value of Ctx. In the case

of 35 pF, t2 is near of the minimal value and with 70 pF, t2 is too low. For Ctx = 70

pF, the distance of detection of the transponder is likely to be too small.

Table 1. Comparison between ISO14443 and simulated results for transmitter operating at and

near resonance.

t1=2 uS Min Max Ctx=35 pF

(near)

Ctx=70 pF

(at resonance)

t2 (5% decay time) 0.7 uS t1 0.6 us 0.5 us

t3 (90% rise time) 0 1.5 uS 0.33 uS 1.22 uS

t4 (60% rise time) 0 0.4 uS 0.19 uS 0.51 uS

4.3 Measurements of the Parameters of the Transponder

Because the parameters have to be extracted in a non destructive test, we use a

contactless methodology. The transponder was placed in a test primary coil shown in

figure 10. The resistance Re and inductance Le of this coil have been characterized

with an impedance analyser. Then the mutual coefficient between this primary coil

and a transponder without its IC has been extracted by measurement of the induced

voltage at the open transponder coil.

Fig. 10. Set-up to measure the parameters of the transponder.

The equation giving the impedance at the primary coil is:

ωω

ppstt

jCRRjL

RjLZp

+++

++=

(1)

Rt, Lt: Series resistance and inductance of the transponder coil without ferrite.

Lt: Inductance of the transponder coil with ferrite.

Rps: Series resistance due to magnetic losses.

Cp: input capacitance of the IC.

By measuring the resonance frequency, input resistance and reactance at the primary

coil, one can extract the frequency at which denominator cancels. So, because the

input capacitance is known, the value of secondary inductance and resistance are

extracted.

Fig. 11. Measurement of input impedance at primary coil.

Vdata

Vi n_rx VDMOS

VGMOS

Vrx

R=8

L=2.65 uH

R=1 kOh m

SRC

SRC5

C=50 pF

R=1 Ohm

R=1.5 kOhm

MUTI ND

R2 2

R=5 0 Ohm

R1 2

R=820 Ohm

C2 6

C=2 2 p F

C2 5

C=100 nF

R1 0

R=500 Ohm

C2 4

C=7 0 pF

R=0 . 5

L=1 uH

MUTINDMUTIND

R= 1

L=1 uH

R=1

L=1 uH

sp_sms_BF998_1_19920901

SNP1

Frequency="{0.05 - 2.00} GHz"

Bias="Mosfet: Vds=5V Vg2s=3.5V Id=10mA"

R1 3

R=120 Ohm

R=37 Ohm

L=22 uH

C2 3

C=100 pF

Fig. 12. Complete system (Transmitter not represented but all mutual inductances are present).

5 Characterization of the Complete System

Now we can proceed to the simulation of the complete system. All the individual

elements have been optimized according to known criteria as for example to

k(TX-TPR)=2.5%

k1(TPR-RX)=1%

k2(TPR-RX)=2%

k1(TX-RX)=50.5%

k2(TX-RX)=49.5%

maximize the radiated H field or to center the RX filter onto one backscattered side

band. The position of the mouse when crossing the reading volume is taken into

account by considering variable coupling coefficients between the transponder TPR

and receiver RX. As shown in figure 5, the magnetic induction received by the

transponder changes and the coupling coefficient follows.

In figure 12, the complete system is represented (note that transmitter does not appear

for clarity) with all the mutual coefficients between transmitter TX, receiver RX and

transponder TPR.

The coupling coefficients k (TX-RX) between TX and RX are made variable. The

tests are performed for weak coupling (transponder in the farthest zone) and for a non

ideal coupling (zero coupling) between transmitter antenna and differential receiver

antennas:

The differential antenna system has been modelized by two coils in series which

induced currents are inverted according to the reference dot (simulates the inverted

windings of the spirals). In the case of ideal coupling between Tx and Rx, the induced

voltages in the spiral antennas is null. To account for the non ideal case, we consider

non equal mutual coefficients.

We note important results in table 2. Even in the case where there is no coupling

between the transponder and the receiver (f1), a modulated signal is present as shown

in figure 14. Actually, it comes from the transmitter which is coupled to the

transponder. So, it means that even in the case of null coupling, because the

transmitter and the receiver antennas cannot be perfectly uncoupled, data exist in the

receiver.

Table 2. Received modulated voltage with different TPR to RX coupling coefficients.

Vrx

No coupling

to TPR-RX

With equal coupling

k1=0.01 k2=0.01

With unequal coupling

k1=0.01

k2=0.02

With no Tx-Rx

coupling

k1=0.02 k2=0.01

60mV f1 60mV f2 62mV f3 140mV f4

Fig. 13. Waveforms at the receiver for 4 case studies:

f1: with no coupling between Tpr-Rx and coupling between Tx-Rx.

f2: with equal coupling between Tpr-Rx and coupling between Tx-Rx.

Fig. 14. Waveforms at the receiver for 4 case studies (cont.):

f3: with unequal coupling between Tpr-Rx and coupling between Tx-Rx.

f4: with unequal coupling between Tpr-Rx and no coupling between Tx-Rx.

We can even add that this is the main coupling mechanism because in the case of

coupling between transponder and receiver (f2, f3), the modulation voltages are of the

same order. Besides, this extra coupling has the effect of multipath summation in

propagation. It means that we can distort the shape of the received data. This is

confirmed by the last case (f4) in figure 14 where we nullified the coupling between

transmitting and receiving antennas (ideal operation mode). We observe that

amplitude is more than double and much closer to the generated square waveform

from the transponder.

6 Conclusions

To optimize an RFID reader magnetically coupled to a tiny ferrite transponder,

compromises have to be made. On the transmitter side, the matching circuit between

the switching MOS transistor and the coil antenna is optimum and should be tuned at

13.56 MHz to maximize the radiated H-field so as to maximize the transponder

induced voltage. This matching circuit should be tuned slightly above 13.56 MHz if

waveform of the modulated signal (pause) is to be fit with the ISO14443

recommendation. On the receiver side, the matching circuit between the differential

coil antennas and the Low Noise Amplifier is to be centered on the side band (13.56

MHz plus the subcarrier frequency = 14.4 MHz) with a bandwidth that should

preserve the integrity of the modulation data. As for the transponder, it is a key

element and the ferrite itself should be chosen with great care due to the detrimental

effect of too high magnetic losses. In the extracted model, we found 8 Ohms for the

equivalent resistance which is the primary factor to determine the Q, here 10. This

quality factor means a bandwidth of around 1.5 MHz, which is necessary to preserve

a sufficient bandwidth to receive and send the data from and to the transmitter. Due to

the shape of the rod, the effective permeability is much less than the relative

permeability. We computed and measured an effective µ

’ of 14 for a relative µ

’of

120. To determine the induced voltage, magnetic simulations have been performed so

as to calculate the B field at any point in the reading volume.

A differential receiving antenna to avoid the strong voltage which occurs in classical

HF systems has been designed. We managed to obtain a reduced coupling with safe

voltage values. We noticed as well that this imperfect coupling means an intermediate

path for the data to couple from the transponder to the receiver. In consequence, the

reading volume is larger than foreseen, because there is no dead zone but as a

drawback, the data may be distorted. To reduce the Tx-Rx coupling, the Tx antenna is

built with two coils, one on each side of the board to insure the symmetry with the

spiral antennas.

In conclusion, an HF reader has been simulated, built and optimized to reach a

reading distance of about 3 cm for an ultra-small ferrite transponder injected under the

skin of the mouse.

References

1. "Identification cards - Contact less integrated circuit(s) cards - proximity cards", ISO / IEC

14443 – 2 international standards, Part 2 "Radio frequency power and signal interface".

2. "Identification cards – test methods", ISO / IEC 10373-6 international standard, Part 6

"proximity cards".

3. N. Chaimanonart, K. R. Olszens, M.D. Zimmerman, W.H. Ko, D.J. Young, “Implantable

RF Power Converter for Small Animal In Vivo Biological Monitoring”

Engineering in Medicine and Biology Society, 2005. IEEE-EMBS 2005. 27th Annual

International Conference, 2005 pp 5194 – 5197.

4. M. Soma, D. C. Galbraith, R. L. White, "Radio-frequency coils in implantable devices:

misalignment analysis and design procedure", IEEE Trans. on biomedical engineering, vol.

BME-34, No. 4, pp 276-282, April 1987.

5. C. M. Zierhofer, E. S. Hochmair, "Geometric approach for coupling enhancement of

magnetically coupled coils", IEEE Trans. on biomedical engineering, vol. 43, No. 7, pp

708-714, July 1996.

6. D. Kajfez, "Dual resonance" IEEE Proceedings, vol. 135, Pt. H, No 2, pp 141-144, April

1988.

7. D. Kajfez, "Q-factor measurement with network analyzer", IEEE Transactions on

microwave theory and techniques, vol. MTT-32, No. 7, pp 666-670, July 1984.

8. Ansys Software. Magneto-dynamic simulation. www.ansys.com