Vous êtes sur la page 1sur 78




Electronics and Electrical Engineering Department
Fontys University of Applied Science
BEE Tracking System Using RFID


Task24, B.V

Antoniustraat 9, Eindhoven

The Netherlands

Company Mentor

Egbert Touw

Performance Improvement Competence Leader

School Mentor

Piet Van Roij

Fontys Hogeschool Engineering


Safee Hussain

Fontys Hogeschool Engineering Dept. (EEE)

Student-number: 2137177

Graduation Session: 7th July 2009

Document History

Version Date Remarks Review by

1.0 June 7, 2009 Piet Van Roij

1.1 June 10,2009 Piet Van Roij

1.2 June 12,2009 Piet Van Roij

2.0 June 16,2009 Piet Van Roij


The contents of this document must be treated as confidential. Changing

the contents of this document is prohibited. No part of this document may
be reproduced, stored in a retrieval system or transmitted in any form or
by any means, electronic, mechanical, photocopying, recording or
otherwise, without the prior written permission of Authors/Mentor.

Management Summary

Over recent years farmers and beekeepers who breed bees in

conventionally commercial ways and stand direct beneficiaries of the
animal keeping through its marvelous abilities of pollinating plant seeds
and nectar collection, have encountered a problem of colony collapse in a
massive effect. Thousands of bees are disappeared in a season, which
greatly affect food and honey production. Our domain specifically
contributes by detecting bees and tracking their movement in mass. The
solutions rendered in previous expeditions point towards a solution
through the technology termed as RFID. This requires us to arrange a
system including a tag to be pasted on bee, a reader to register bee
entering and leaving of Hive and an interface between the reader and a

Cost on a system that has not yet been realized and created in real world
completely is always larger than expected; we’re discussing a problem
that could pose threat directly to 15-30 % of total human food production
so benefit; is huge in its natural sense and is a service in its moral sense.
The cost of for the whole system is suggested to be € 880.5.

Conclusively, our effort shows that the solution of tracking bees; through
RFID is really difficult if not “impossible”. This decision is taken after a
vigorous exercise which is elaborated in our report below, it includes a
research and practical implementation of design proposed to be the best
in given conditions and circumstances. The final decision is of course
taken by the judges and reader himself through arguments given.


We divided the problem in its core systematic order, we required a

transponder with built in antenna and chip small enough to be carried by a
bee, such a tag was available from Microsensys a German company and
since the tag was following internationally recognized protocols, a reader
from Skyetek an American company, was also searched out to be attached
with an antenna design proposed and implemented during the project.

Size of the tag, range to be covered by the antenna, frequency selection,

physical dimensions of hive entrance, communication hurdles intimately
connected RFID and bee’s natural reactive behavior; greatly influence the
antenna design part. This writing offers a comprehensive argument on the
decisions taken to solve the problem of tracking. It should be clear that
the antenna design which we propose is to be attached with reader
equipment to detect the transponder, because there’s no design as such
which has been brought to reality yet.

Our plan was to implement the detection procedure primarily and then
transfer the registered information to a server through GSM modules and
upload the information on a website. Although our antenna design did not
detect the required tag sensed from the reader equipment, but the tag
had clear influence on the magnetic field of antenna in its proximity seen
from the network analyzer. Plus our antenna was able to sense and detect
most common tags in the market. We can only use the passive tags as
bees cannot carry a battery to run the chip on tag.


As the name suggests, the project Bee Tracking System uses RFID for
detection of Bees, use of RFID as technology for solving the problem had
already been suggested in a prototype proposed by John Roodt, a previous
internee on the project. This report demonstrates the practical
implementation and implications that came as results after the pursuit of
problem solution. There is no debate on the choice of technology, but the
report includes a comprehensive debate on physical and practical
exercises made to bring the idea to real world.

RFID comprises of two basic modular structures, one to be termed as

Reader and other to be Tag. We include the in depth analysis of RFID
technology especially in the light of limitations posed by the insect bee, its
behavior and its hive.

Once if the bee is detected and registered correctly, we can send the
information to an internet server using simple GSM modules. Five months
of pursuit doesn’t only include the lab work, but also an internet research
on the selection of right equipment, right frequency of communication and
antenna design research. The frequency was selected to be in the ISM
band 13.56 MHz.

Research part of the project through internet was carried at TASK 24

during the initial period of two months; this also includes the arrangement
of reader equipment from a set of companies which shall be explained
later. For the rest of the three months, the practical part of the project was
pursued at Fontys Telecommunication Laboratory.

Table of Contents
Table of Contents...................................................................................................8
1.1 Problem Description....................................................................................10
1.2 Assignment.................................................................................................10
1.3 Resources and Support...............................................................................10
2.1 RFID Technology.........................................................................................11
2.1.1 Merits of RFID .......................................................................................12
2.1.2 Demerits of RFID ..................................................................................12
2.1.3 Inductive Coupling ............................................................................13
2.1.4 Backscatter Coupling............................................................................16
2.2 RFID Antenna Design (Inductively Coupled)................................................18
2.2.1 Current and Magnetic Fields ...............................................................18
2.2.2 Induced Voltage in Antenna Coil...........................................................20
2.2.3 Inductance of various Antenna Coils.....................................................21
2.2.4 Resonance circuits, Quality factor Q and Band width............................25
2.2.5 Antenna Symmetry...............................................................................28
2.2.6 Field Strength and effects of Planes.....................................................29
2.3 Antenna Matching ......................................................................................31
2.3.1 Significance...........................................................................................31
2.3.2 How to Match .......................................................................................33
2.4 Bee Behavior...............................................................................................36
2.4.1 Bees and Magnetic Fields......................................................................36
2.4.2 Bee and Bee Hive..................................................................................38
2.5 System........................................................................................................ 38
2.5.1 The Tag.................................................................................................40
2.5.3 Antenna Switching Mechanism.............................................................43
2.6 Problem Realization in the light of RFID......................................................45
Implementation.................................................................................................... 46
3.1 Antenna Circuit............................................................................................46
3.1.1 Dimensional calculations of Coil............................................................47
3.1.2 Configurations......................................................................................48

3.1.3 Measurements.......................................................................................49
3.2 Antenna Matching.......................................................................................51
3.3 Testing with Reader....................................................................................54
3.3.1 Results.................................................................................................. 55
3.4 Observations...............................................................................................56
3.5 Cost Analysis...............................................................................................61
3.6 Problem Faced and Tackled........................................................................62
References........................................................................................................... 66
Glossary................................................................................................................ 68
Annex A-1............................................................................................................. 69
Annex A-2............................................................................................................. 74

1.1 Problem Description
Colony Collapse Disorder is the name of the problem being faced by
most of the beekeepers all over Europe and America. Male bees
“drones” are vanishing in large numbers; this directly affects the
fruit and vegetable production. To solve the problem we need to
track bees and see what makes them disappear. Initially, we need to
tag them with any identification and follow their movement by
registering them each time they enter or leave any hive. Although
U.S. beekeepers have been hardest hit by CCD, beekeepers in
Europe have reported similar honeybee problems.

Mr. Egbert being himself as a bee-keeper and member of Beekeeper

Association Netherlands encountered the problem of colony collapse.
Thus, bee tracking holds key role in the problem solution process.

Now that it is established that bee tracking is important, we need to

have a closer look into the situation in terms of Reader Antenna and
Tag selection or even a whole lot of new design based on RFID

1.2 Assignment
Detect and register the tagged bees whenever they enter or leave
the hive.

1.3 Resources and Support

a) Task 24, provided us with a notebook.

b) Mr. Egbert Touw, gave financial support, communicated with

companies and managed the project from company´s end.

c) Mr. Piet Van Roij, technically supported each step and mentored
this project from school´s end.

d) Fontys, provided with equipment support of network analyzer,

spectrum analyzer along with circuit components and laboratory
work place.

2.1 RFID Technology
Radio frequency identification (RFID) is a generic term
that is used to describe a system that transmits the identity (in the form of
a unique serial number) of an object or person wirelessly, using radio
waves. It's grouped under the broad category of automatic identification

A typical RFID tag consists of a microchip attached to a radio antenna

mounted on a substrate. The chip can store as much as 2 kilobytes of
data. For example, information about a product or shipment—date of
manufacture, destination and sell-by date—can be written to a tag.

To retrieve the data stored on an RFID tag, you need a reader. A typical
reader is a device that has one or MORE antennas that emit radio waves
and receive signals back from the tag. The reader then passes the
information in digital form to a computer system

Read range is determined by many factors, but one of the most important
is the method passive tags use to transmit data to the reader. Low- and
high-frequency tags use inductive coupling. Essentially, a coil in the reader
antenna and a coil in the tag antenna form an electromagnetic field. The
tag draws power from the field, uses the power to run the circuitry on the
chip and then changes the electric load on the antenna. The reader
antenna senses the change in the magnetic field and converts these
changes into the ones and zeros that computers understand. Because the
coil in the tag antenna and the coil in the reader antenna must form a
magnetic field, the tag must be fairly close to the reader antenna, which
limits the read range of these systems.

Illustration of RFID in Block diagram 2.1.1 (**http://www.rfid-


2.1.1 Merits of RFID

The following are the benefits of RFID Systems:

• Non-line of sight identification of tags.

• Unattended operations are possible, minimizing human
errors and high cost.
• Ability to identify moving elements that have tags
• Larger area of coverage. Up to several feet.
• Can be used in diverse environments, including live
stock, military, and scientific areas.
• RFID can be used in addition to Bar Code. These two
technologies can be complementing each other.
• Automatic integration with back end software solutions
provide end to end integration of data in real time.
• Enables distributed deployment and edge processing.
• Provides enterprise-class management and security.
• Abstracts low-level hardware interfaces.
• Insulates developers from standards and protocols.
• Speeds development of business logic.
• Simplifies integration with enterprise systems.
• Provides a robust simulator to assist development.
• Offers an advanced architecture for future growth.

2.1.2 Demerits of RFID

• Expensive compared with Bar code.

• Bulkier, due to embedding of electronic components in
the tag. However, with advanced techniques, it is possible to
reduce the size, and weight of the tags to a large extent.
• Prone to physical/electrical damage due to
environmental conditions. For example, tags that are
subjected to space exploration may encounter extreme
temperatures. The tags required to be designed for a given
application, and may be costly when designed for use under
extreme environmental conditions.

The benefits of this technology are very attractive and technically catchy
enough to overrun any other technology for tracking a mobile commodity.
The management and organizational abilities of RFID make it more reliable
and applicable in different scenario, same is the case with our problem.
But this is not all that we should know about RFID before we want to
implement this system. We should take specific RFID traits to our
consideration; RFID has its own way of operation and posses its own
application hurdles. We include the main RFID topics of technicality as

2.1.3 Inductive Coupling

An inductively coupled transponder comprises of an electronic data
carrying device, usually a single microchip and a large area coil that
functions as an antenna.

Inductively coupled transponders are almost always operated passively.

This means that all the energy needed for the operation of the microchip
has to be provided by the reader. For this purpose, the reader's antenna
coil generates a strong, high frequency electro-magnetic field, which
penetrates the cross -section of the coil area and the area around the coil.
Because the wavelength of the frequency range used (< 135 kHz: 2400 m,
13.56 MHz: 22.1 m) is several times greater than the distance between the
reader's antenna and the transponder, the electro-magnetic field may be
treated as a simple magnetic alternating field with regard to the distance
between transponder and antenna.

A small part of the emitted field penetrates the antenna coil of the
transponder, which is some distance away from the coil of the reader. By
induction, a voltage is generated in the transponder's antenna coil. This
voltage is rectified and serves as the power supply for the data carrying
device (microchip). A capacitor is connected in parallel with the reader's
antenna coil, the capacitance of which is selected such that it combines
with the coil inductance of the antenna coil to form a parallel resonant
circuit, with a resonant frequency that corresponds with the transmission
frequency of the reader. Very high currents are generated in the antenna
coil of the reader by resonance step-up in the parallel resonant circuit,
which can be used to generate the required field strengths for the
operation of the remote transponder.

The antenna coil of the transponder and the capacitor to form a resonant
circuit tuned to the transmission frequency of the reader. The voltage at
the transponder coil reaches a maximum due to resonance step-up in the
parallel resonant circuit.

Figure 2.1.2(**http://www.rfid-handbook.com)

As described above, inductively coupled systems are based upon

a transformer-type coupling between the primary coil in the reader and
the secondary coil in the transponder. This is true when the distance
between the coils does not exceed 0.16 λ, so that the transponder is
located in the near field of the transmitter antenna.

If a resonant transponder (i.e. the self-resonant frequency of the

transponder corresponds with the transmission frequency of the reader) is
placed within the magnetic alternating field of the reader's antenna, then
this draws energy from the magnetic field. This additional power
consumption can be measured as voltage drop at the internal resistance in
the reader antennae through the supply current to the reader's antenna.
The switching on and off of a load resistance at the transponder's antenna
therefore effects voltage changes at the reader's antenna and thus has
the effect of an amplitude modulation of the antenna voltage by the
remote transponder. If the switching on and off of the load resistor is
controlled by data, then this data can be transferred from the transponder
to the reader. This type of data transfer is called load modulation.

To reclaim the data in the reader, the voltage measured at the reader's
antenna is rectified. This represents the demodulation of an amplitude
modulated signal. An example circuit is shown in the chapter "Reader –
Low Cost Layout".

Figure 2.1.3(**http://www.rfid-handbook.com)

Picture: sample circuit of the power supply and load modulator in a


Picture above: If the additional load resistor in the transponder is switched

on and off at a very high elementary frequency fH, then two spectral lines
are created at a distance of ±fH around the transmission frequency of the
reader, and these can be easily detected (however fH must be less than
fREADER). In the terminology of radio technology the new elementary
frequency is called a subcarrier. Data transfer is by the ASK, FSK or PSK
modulation of the subcarrier in time with the data flow. This represents an
amplitude modulation of the subcarrier.

Inductive reader-coil: circle radius vs. field strength

The figure below shows the relationship between the circle radius of the
reader-antenna and the field strength H a distance of 0 .. 10 m. In this
figure the circle radius slowly increases from 1 .. 50 cm, while the current I
is kept constant. It can be seen that while the field strength decreases at
zero distance decreases, the distance in large distance increases. That’s
the reason, that inductive coupled long range systems (ISO 15693) use
large diameter reader antennas, compared to short range systems (ISO
14443) using small diameter reader antennas.

Figure2.1.4 (**http://www.rfid-handbook.com)

2.1.4 Backscatter Coupling

We know from the field of RADAR technology that electromagnetic waves
are reflected by objects with dimensions greater than around half the
wavelength of the wave. The efficiency with which an object reflects
electromagnetic waves is described by its reflection cross-section. Objects
that are in resonance with the wave front that hits them, as is the case for
antenna at the appropriate frequency for example, have a particularly
large reflection cross-section.

Figure 2.1.5(**http://www.rfid-handbook.com)

Picture: Operation principle of a backscatter transponder

Power P1 is emitted from the reader's antenna, a small proportion of which

(free space attenuation) reaches the transponder's antenna. The power
P1' is supplied to the antenna connections as HF voltage and after
rectification by the diodes D1 and D2 this can be used as turn on voltage

for the deactivation or activation of the power saving "power-down" mode.
The diodes used here are low barrier Schottky diodes, which have a
particularly low threshold voltage. The voltage obtained may also be
sufficient to serve as a power supply for short ranges.

A proportion of the incoming power P1' is reflected by the antenna and

returned as power P2. The reflection characteristics (= reflection cross-
section) of the antenna can be influenced by altering the load connected
to the antenna. In order to transmit data from the transponder to the
reader, a load resistor RL connected in parallel with the antenna is
switched on and off in time with the data stream to be transmitted. The
amplitude of the power P2 reflected from the transponder can thus be
modulated (modulated backscatter).

The power P2 reflected from the transponder is radiated into free space. A
small proportion of this (free space attenuation) is picked up by the
reader's antenna. The reflected signal therefore travels into the antenna
connection of the reader in the "backwards direction" and can be
decoupled using a directional coupler and transferred to the receiver input
of a reader. The "forward" signal of the transmitter, which is stronger by
powers of ten, is to a large degree suppressed by the directional coupler.

The ratio of power transmitted by the reader and power returning from the
transponder (P1 / P2) can be estimated using the radar equation.

2.2 RFID Antenna Design (Inductively Coupled)
In an RFID application, an antenna coil is needed for two main reasons:

• To transmit the RF carrier signal to power up the tag

• To receive data signals from the tag

An RF signal can be radiated effectively if the linear dimension of the

antenna is comparable with the wavelength of the operating frequency.
In an RFID application utilizing the VLF (100 kHz – 500 kHz) band, the
wavelength of the operating frequency is a few kilometers (l = 2.4 Km
for 125 kHz signal). Because of its long wavelength, a true antenna can
never be formed in a limited space of the device. Alternatively, a small
loop antenna coil that is resonating at the frequency of the interest
(i.e., 13.56) is used. This type of antenna utilizes near field magnetic
induction coupling between transmitting and receiving antenna coils.
The field produced by the small dipole loop antenna is not a
propagating wave, but rather an attenuating wave. The field strength
falls off with r-3 (where r = distance from the antenna). This near field
behavior (r-3) is a main limiting factor of the read range in RFID
applications. When the time-varying magnetic field is passing through a
coil (antenna), it induces a voltage across the coil terminal. This voltage
is utilized to activate the passive tag device. The antenna coil must be
designed to maximize this induced voltage.

2.2.1 Current and Magnetic Fields

Ampere’s law states that current flowing on a conductor produces a
magnetic field around the conductor. Figure 1 shows the magnetic field
produced by a current element. The magnetic field produced by the
current on a round conductor (wire) with a finite length is given by:

Bφ = µ oI /4π r (cosα 2 – cosα 1) [ Weber /m2]

 Equation 1


I = current

r = distance from the center of wire

µ o = permeability of free space and given as

µ o = 4 π x 10-7 (Henry/meter)

In a special case with an infinitely long wire where

a1 = -180° and a2 = 0°, Equation 1 can be rewritten as:

Bφ = µ oI/ 2π r [Weber /m2] 

Equation 2

Figure 2.2.1(**)

The magnetic field produced by a circular loop antenna coil with N-turns as
shown in Figure 2 is found by:

Bz = (µ 0INa2)/ 2(a2 + r2) 3 2
Equation 3
= (µ 0INa2)/2* (1/ r3) for r2>>a2
• a = radius of loop

Equation 3 indicates that the magnetic field produced by a loop antenna

decays with 1/r3 as shown in Figure 3. This near-field decaying behavior of
the magnetic field is the main limiting factor in the read range of the RFID
device. The field strength is maximum in the plane of the loop and directly
proportional to the current (I), the number of turns (N), and the surface
area of the loop.
Equation 3 is frequently used to calculate the ampere-turn requirement for
read range. A few Examples that calculate the ampere-turns and the field
intensity necessary to power the tag will be given in the coming up section

Figure 2.2.2(**)

Figure 2.2.3(**)

2.2.2 Induced Voltage in Antenna Coil

Faraday’s law states a time-varying magnetic field through a surface bounded by
a closed path induces a voltage round the loop. This fundamental principle has
important consequences for operation of passive RFID devices. Figure 2.5 shows
a simple geometry of an RFID application. When the tag and reader antennas are
within a proximity distance, the time-varying magnetic field B that is produced
by a reader antenna coil induces a voltage called electromotive force or simply
EMF) in the tag antenna coil. The induced voltage in the coil causes a flow of
current in the coil. This is called Faraday’s law. The induced voltage on the tag
antenna coil is equal to the time rate of change of the magnetic flux Ψ .

V = - N dΨ/ dt
Equation 4


• N = number of turns in the antenna coil

• Ψ = magnetic flux through each turn

The negative sign shows that the induced voltage acts in such a way as to
oppose the magnetic flux producing it. This is known as Lenz’s Law and it
emphasizes the fact that the direction of current flow in the circuit is such that
the induced magnetic field produced by the induced current will oppose the
original magnetic field. The magnetic flux Y in Equation 4 is the total magnetic
field B that is passing through the entire surface of the antenna coil, and found

ψ = ∫ B·dS Equation 5


• B = magnetic field given in Equation 3

• S = surface area of the coil

The inner product presentation of two vectors in Equation 5 suggests that

the total magnetic flux y that is passing through the antenna coil is
affected by an orientation of the antenna coils. The inner product of two
vectors becomes maximized when the two vectors are in the same
direction. Therefore, the magnetic flux that is passing through the tag coil
will become maximized when the two coils (reader coil and tag coil) are
placed in parallel with respect to each other.

Figure 2.2.4(**)

2.2.3 Inductance of various Antenna Coils

The electrical current flowing through a conductor produces a magnetic
field. This time-varying magnetic field is capable of producing a flow of
current through another conductor. This is called inductance. The
inductance L depends on the physical characteristics of the conductor. A
coil has more inductance than a straight wire of the same material and a
coil with more turns has more inductance than a coil with fewer turns. The
inductance L of inductor is defined as the ratio of the total magnetic flux
linkage to the current I through the inductor: i.e.

L= Nψ / I Equation 6


• N = number of turns

• I = current

• ψ = magnetic flux

In a typical RFID antenna coil for 13.56 MHz, the inductance is often
chosen as a few (uH) for a tag and from a few hundred to a few thousand
(mH) for a reader. For a coil antenna with multiple turns, greater
inductance results with closer turns. Therefore, the tag antenna coil that
has to be formed in a limited space often needs a multi-layer winding to
reduce the number of turns. The design of the inductor would seem to be
a relatively simple matter. However, it is almost impossible to construct an
ideal inductor because:

The coil has a finite conductivity that results in loss; the distributed
capacitance exists between turns of a coil and between the conductor and
surrounding objects. The actual inductance is always a combination of
resistance, inductance, and capacitance. The apparent inductance is the
effective inductance at any frequency, i.e. inductive minus the capacitive
effect. Various formulas are available in literatures for the calculation of
inductance for wires and coils. The parameters in the inductor can be
measured. LCR Meter can measure the inductance, resistance, and Q of
the coil.

Inductance of a Single Layer Coil

The inductance of a single layer coil shown in Figure 2.6 can be calculated

L = (aN)2/ 22.9l + 25.4a

Equation 7


• a = coil radius (cm)

• l = coil length (cm)

• N = number of turns

Figure 2.2.5(**)

Inductance of a Circular Loop Antenna Coil with Multilayer
To form a big inductance coil in a limited space, it is more efficient to use
multilayer coils. For this reason, a typical RFID antenna coil is formed in a
planar multi-turn structure. Figure 8 shows a cross section of the coil. The
inductance of a circular ring antenna coil is calculated by an empirical

L = 0.31(aN)2/ 6a+9h+10b
Equation 8

• a = average radius of the coil in cm

• N = number of turns

• b = winding thickness in cm

• h = winding height in cm

Figure 2.2.6 a(**)

N = [Lµ H(6a+9h+10b)/ ( 0.31)a2]1/2

Rectangular Antenna Coil

Calculation of Inductance:


Figure 2.2.6 b(**)
• ao, bo = Overall dimensions of the coil
• aavg, bavg = Average dimensions of the coil
• t= track thickness
• w= track width
• g= gap between tracks
• Na= Number of turns
• d= equivalent diameter of the track

Theory Behind
The inductance for the reader antenna coil is typically in the range of a
few hundred to a few thousand micro- henries (uH) for high frequency
applications. The reader antenna can be made of either a single coil that is
typically forming a series resonant circuit or a double loop (transformer)
antenna coil that forms a parallel resonant circuit.
The series resonant circuit results in minimum impedance at the
resonance frequency. Therefore, it draws a maximum current at the
resonance frequency. On the other hand, the parallel resonant circuit
results in maximum impedance at the resonance frequency. Therefore,
the current becomes minimized at the resonance frequency.

Since the voltage can be stepped up by forming a double loop (parallel)
coil, the parallel resonant circuit is often used for a system where a higher
voltage signal is required. (Figure 2.2.7) shows an example of the
transformer loop antenna. The main loop (secondary) is formed with
several turns of wire on a large frame, with a tuning capacitor to resonate
it to the resonance frequency (13.56 MHz). The other loop is called a
coupling loop (primary), and it is formed with less than two or three turns
of coil. This loop is placed in a very close proximity to the main loop,
usually (but not necessarily) on the inside edge and not more than a
couple of centimeters away from the main loop. The purpose of this loop is
to couple signals induced from the main loop to the reader (or vice versa)
at more reasonable matching impedance.
The coupling (primary) loop provides an impedance match to the
input/output impedance of the reader. The coil is connected to the
input/output signal driver in the reader electronics. The main loop
(secondary) must be tuned to resonate at the resonance frequency and is
not physically connected to the reader electronics. The coupling loop is
usually unturned, but in some designs, a tuning capacitor C2 is placed in
series with the coupling loop. Because there are far fewer turns on the
coupling loop than the main loop, its inductance is considerably smaller.
As a result, the capacitance to resonate is usually much larger.

Figure 2.2.7(**)

2.2.4 Resonance circuits, Quality factor Q and Band width

In RFID applications, the antenna coil is an element of resonant circuit and
the read range of the device is greatly affected by the performance of the
resonant circuit. Figure 2.2.7 show typical examples of resonant circuits
formed by an antenna coil and a tuning capacitor. The resonance
frequency (fo) of the circuit is determined by:

Equation 9


• L = inductance of antenna coil

• C = tuning capacitance

The resonant circuit can be formed either series or parallel. The series
resonant circuit has minimum impedance at the resonance frequency. As a
result, maximum current is available in the circuit. This series resonant
circuit is typically used for the reader antenna. On the other hand, the
parallel resonant circuit has maximum impedance at the resonance
frequency. It offers minimum current and maximum voltage at the
resonance frequency. This parallel resonant circuit is used for the tag

Figure 2.2.8 shows a simple parallel resonant circuit. The total impedance
of the circuit is given by:

Z(jw)= jwL/ [(1- w2 LC) +j(wL/R)]

Equation 10


• w = angular frequency = 2*pi*f

• R = load resistor

The ohmic resistance r of the coil is ignored. The maximum impedance

occurs when the denominator in the above equation minimized such as:

w2LC = 1 Equation 11

This is called a resonance condition and the resonance frequency is given


Equation 12

By applying Equation 11 into Equation 10, the impedance at the resonance

frequency becomes:

Z=R Equation 13

Figure 2.2.8(**)

The R and C in the parallel resonant circuit determine the bandwidth, B, of

the circuit.

B = 1/ 2π RC
Equation 14
The quality factor, Q, is defined by various ways such as:

Q = Energy Stored in the System per One Cycle/ Energy Dissipated in the
System per One Cycle

Q = fo/B  Equation 15


• fo = resonant frequency

• B = bandwidth

By applying Equation 12 and Equation 14 into Equation 15, the loaded Q in

the parallel resonant circuit is:

Q=R  Equation 16

The Q in parallel resonant circuit is directly proportional to the load

resistor R and also to the square root of the ratio of capacitance and
inductance in the circuit. When this parallel resonant circuit is used for the
tag antenna circuit, the voltage drop across the circuit can be obtained by
combining Equations 16 and ( Vo = 2π foNQSBocosα )

Vo = 2*pi*foNQSBo * cos a 
Equation 17

=2*pi*foN[ R( C/ L)1/2]SBo* cos a

The above equation indicates that the induced voltage in the tag coil is
inversely proportional to the square root of the coil inductance, but
proportional to the number of turns and surface area of the coil. The
parallel resonant circuit can be used in the transformer loop antenna for a
long-range reader as discussed in "Reader Antenna Coil" (Figure 2.2.7).

The voltage in the secondary loop is proportional to the turn ratio (n2/n1)
of the transformer loop. However, this high voltage signal can corrupt the
receiving signals. For this reason, a separate antenna is needed for
receiving the signal. This receiving antenna circuit should be tuned to the
modulating signal of the tag and detuned to the carrier signal frequency
for maximum read range.

2.2.5 Antenna Symmetry

The symmetry in antenna design is absolutely necessary with respect to
tuning and EMC behavior.

Figure 2.2.9

The following drawing show and example of a symmetric 4-turn antenna


Figure 2.2.10

2.2.6 Field Strength and effects of Planes
The benefit of a ferrite is to shield an antenna against the influence of
metal. A metal plane could be part of the housing of the NFC device or a
ground plane of the NFC device PCB itself, which has to be connected very
near to the antenna. If metal is placed very near to antenna the
alternating magnetic field generates eddy currents in the metal. These
eddy currents absorbed power, and lead to detuning of the antenna due to
a decreased inductance and quality factor. Therefore, for operation of an
antenna in close metallic environment, it is necessary to shield the
antenna with ferrite.

The following examples should give an impression on the influence of

ferrite for the distribution of a magnetic field.

For easy simulation a circular antenna has been used in every case. A
circular antenna is rotation symmetric to the x-axis. Therefore the
simulation can be reduced to a two dimensional mathematical problem.
The simulation shows on the one hand the field distribution of a non-
distributed antenna. Common for all examples: Radius of the antenna 7.5
cm, 1 turn, wire width 1mm copper.

Figure 2.2.11 shows the 2D field of the circular antenna. The right part
shows the field distribution. The highest field strength is generated in the
area of the coil.The left part shows the magnitude of the field strength H
over a distance d. The line of a minimal field strength of Hmin = 1.5 A/m is

Figure 2.2.11

Figure 2.2.12 shows the field distribution of the same antenna with a
metal plane and a ferrite plane near to the antenna.

Figure 2.2.12

2.3 Antenna Matching
Antennas are matched for maximum power transfer from the transmitter
to the antenna. Most of the antennas are matched on 50 ohms and 75
ohms. In our exercise we pursued 50 ohms matching.

The first question arising to one’s mind is that why do we need antenna to
be matched. The answer is simple if an antenna is matched to 50 ohms it
means the total impedance of antenna resonating circuit and matching
circuitry is 50 ohms, this is just a standard one can have different
impedance for matching but the impedance is chosen in such a manner
that the impedance of the whole transmitter equals the impedance of
antenna circuitry, so that half of the power from source is dissipated in the
transmitter circuit and half of it is dissipated in antenna circuit. This is the
only way to supply maximum energy to antenna. In this way we can
achieve VSWR=1 and reflection coefficient to be zero.

2.3.1 Significance
This has to be clarified that why do we want to match or tune our antenna with
previous reader equipment including reader IC, encoder, modulator and EMC
filter etc , this because we want maximum power to be delivered into the air to
achieve better coupling factor among two coils. So, the idea is that half of the
power is dissipated in the circuitry before antenna and half to be delivered to the
antenna and if both are matched, and no RF power is reflected back to the
transmitter then we transmit maximum power into the air, of course this power
that we want to see in the air is also limited by regulatory authorities, we give
brief table later related to allowed energy levels. The diagrams below show how
important it is to tune the antenna to a resistance equal to the resistance of
previous transmitter circuitry. In the graphs shown below the matching
impedance is 30 Ohms.

The Current Vis-à-vis Matching Resistance
The following figure 2.3.1(**) shows three curves at different voltages.

Figure 2.3.1(**)

The Power Vis-à-vis Matching Resistance

Figure 2.3.2(**)

The Power Added Efficiency Vis-à-vis Matching Resistance and
Collective Analysis

Figure 2.3.3(**)

We can clearly ponder in the figure above that power added efficiency is the
greatest with minimum power applied to antenna at 30 ohms.

2.3.2 How to Match

Smith Chart
Smith chart is a normalized impedance plot chart, a classic
convention of plotting impedance of an antenna with reference to 50 ohms.
There´s a method widely used to match a circuit using smith chart, which
abstains the addition of a resistor to the circuit, one can achieve a matched
network by adding lumped elements in the light of movements shown in
Figure^……. Let´s just get familiar with the basic smith chart with both
admittance and impedance plot. As can be seen from the diagram below, red
lines represent admittance plot and black lines illustrate the impedance plot.

Figure 2.3.4(**)

General mathematical forms

• Z= R+jX ------------ Impedance

• Y=G+jB-------------Admittance

Figure 2.3.5(**)

We divide our job matching into few steps explained as under

1. Identify the impedance on smith chart

2. Draw an SWR circle on the impedance placing the compass needle in the
middle of the chart.

3. Identify the admittance value on the chart.

4. Plot the unit circle, also called as normalized impedance plot of Z= 1+jx.

5. Plot the auxiliary circle.

6. A). Follow the jX line to unit circle from the impedance plot, in case of
impedance part. Or

B). Follow the jB line to auxiliary circle from the admittance plot, in case of
admittance part.

It can be seen from the figure above that if we follow part A we add the
element in series and if we follow part B we add the lumped element in

7. Now follow the unit circle or auxiliary circle to reach the centre of plot.

These steps together are shown in the following figure.

Figure 2.3.6(**)

2.4 Bee Behavior
In this type of detections where the animal has to carry the tag with it, we
need take the behavior of the animal to our account.

2.4.1 Bees and Magnetic Fields

Orientation and Navigation of Bees may be disturbed by man-made
electric, magnetic and electromagnetic fields. The conclusions of the
research of other scientists and of my own research result in the following

1. The integuments of bees (and bird feathers) have semi

conductive and piezoelectric functions.

2. This means they act as a transducer of pulse modulated

high frequency microwave fields into an audio frequency
range. Certain parts of the structure of the integument
function as dielectric receptors of electromagnetic radiation
in the microwave region.

3. The presence of magnetite nano particles has been

demonstrated in the abdomen of bees. Magnetite is an
excellent absorber of microwave radiation at frequencies
between 0.5 and 10.0 GHz through the process of
ferromagnetic resonance. Pulsed microwave energy
absorbed by this process is first transduced into acoustic
vibrations (magneto acoustic effect).

4. It has been demonstrated that free-flying honeybees are

able to detect static intensity fluctuations and ultra low
frequency magnetic fields as weak as 26 nT against the
background earth-strength magnetic field.

5. Magnetic field bursts at a frequency of 250 Hz oriented

parallel to the field lines of the Earth’s magnetic field induce
unequivocal jumps of misdirection of up to +10°.

6. The magnetic induction levels in today’s environment range

usually between 1 nT and 170,000 nT in the low frequency
range and between several nT and several 1000 nT in the
high frequency range. Hence, these levels are commonly
higher than the threshold of sensibility of bees to variations
of magnetic fields.

7. In the honeybee, the NO-system in the antenna lobes is
implicated in chemosensory processing and in memory
formation. Disturbance of the NO-system through
electromagnetic fields with low and high frequency is
proved today only in mammals, but it is expected to work in
the same manner also in insects. In this case, olfactory
function and the memory of bees are fading.

On balance, the consequence of all these investigations is that the

orientation and navigation of bees may be disturbed by man-made
technical communication fields.

2.4.2 Bee and Bee Hive
• Honeybees are able to carry up to 70 mg

• After when the wooden slab is placed in the entrance, the total
height to enter remains to be 8mm.

• Bees Height 5mm

• So detection range is 3mm excluding the thickness of tag.

• The width of entrance is around 40 cm.

Figure 2.4.1(**)

2.5 System
Skimming the research of Mr. Roodt, it was found that Microsensys a German
company is the only company offering solutions through RFID for bee tracking
worldwide. The equipment they offer is not applicable, which was quite
understood between them and us while we communicated about the solution.
According to Mr. Sylvo Jeager Marketing Manager of the company, it was possible
to detect the pico Mic 3 Tag from iID Maja Reader or iID Mini Pen reader, but a
new system was required to be launched in May 2009 which would give an
applicable and complete solution. We were able to convince them that we are
seriously following the problem and we need their help. Upon a sponsor letter to
Mr. Jeager, the company replied us with good news of handing us over the
prototype of their upcoming product for a technical feedback and maybe some
help on it. This product was a direct solution of bee detection.

Things could not go as smoothly as planned for, and we took the tag from
microsensys and decided to make a reader antenna that fit our problem and use
an efficient reader from the market to derive our own designed antennas.

The tag and the readers we searched out and found to be relevant to the
situation are shown below. Some of the specifications are mentioned here, while
for the rest one should refer to the annex and references.

2.5.1 The Tag
The suggested Microsensys tag is shown below with its dimensions

Figure 2.5.1(**)

2.5.2 The Reader

Company Product Underlined Specs

Texas Instrument ISO 15693

Evaluation Module for TRF7960

Shanghai Fine ISO 15693
Electronics Ltd. External Antenna

Hf Rfid Middle Range Reader

With External Antenna
Skyetek ISO
External Antenna

SkyeModule M2
Microsensys Handheld antenna
ISO 15693

iID Maja Reader ISO

Zixsys External Antenna
with an interface

13.56MHz ISO/IEC15693
compliant Small size Reader
module From ZIXSYS
Microsensys ISO 15693
USB 2.0 Interface

iID Mini Pen / USB Interface

HF transponder transmission levels, calculated

Most of the calculated average transponder levels in the HF range are - 30
… -33 dBμA/m @ d=10m or lower. The CISPR 16 signal weighting will
reduce the measured level by about 10 dB.
Tag Frequency Frequency H (allowed field
Carrier fc message strength)
signal dBμA/m
ISO-14443, 13.56 MHz fc ± 846 KHz -41 dBμA/m
ISO-15693, 13.56 MHz fc ± 423/484 -36 dBμA/m
vicinity KHz
ISO-18000-3 13.56 MHz Similar to the
below listed
13.56 MHz
I-code (Philips) 13.56 MHz fc ± 423 KHz -36 dBμA/m
Tag-It (TIRIS) 13.56 MHz fc ± 423/484 -36 dBμA/m
Microchip 13.56 MHz fc ± 70 KHz -24 dBμA/m
Gemplus 13.56 MHz fc ± 106 KHz -27 dBμA/m
fc ± 212 KHz -30 dBμA/m
Legic 13.56 MHz fc ± 212 KHz -33 dBμA/m

ISO 15693 (ISO SC17/WG8) - vicinity cards

Power supply 13.56 MHz inductive coupling,
0,15 .. 7,5 A/m
Downlink (data transmission reader Modulation:
> card) 10% ASK, 100% ASK (card supports
Bit coding:
“1 out of 256”, “1 out of 4” (card
supports both)
Baud rate:
1.65 kBit/s, 26.48 kBit/s (card
supports both)
Uplink (data transmission card > Modulation:
reader) Load modulation with subcarrier
Bit coding:
Manchester, the subcarrier is ASK
(423 kHz) or FSK (423/485 kHz)
Baud rate:
6.62 kBit/s, 26.48 kBit/s (selected
by reader)

2.5.3 Antenna Switching Mechanism
If one antenna is able to detect a tag then we can build a whole array of
40 antennas covering the whole entrance, we can use either switch IC or
some interface with microcontroller to switch between the antennas and
scan the whole entrance in micro-seconds to make sure that we do not
miss any tagged bee entering or leaving the facility.

We had to expedite through internet to find some of the related ICs,

capable of handling 4 antennas at the same some of them are as under:

Company/Authorized Picture/Name Specification

Analog devices SP4T


ND) - ICs, Modules,
Accessories for RF, RFID

New Japan Electronics SP4T


Spectrum Microwave SP4T


Using one of these ICs along with a microcontroller would enable us to
switch between around 40 antennas fabricated in each Hive. Since we had
limited time, we could only come up with a prototype of switching
mechanism, but we still need to see the practical implications of the
subject. We did this research to take out the cost analysis of the project.

2.6 Problem Realization in the light of RFID
If we just take a look on the information provided above, we can sort the
problem in simple words. That is to say that, we have to design such an
antenna which is able to detect this microsensys tag (small size/pico tag)
from the range of 3mm at least. The tag is bearable by the bee as it can
carry weight up to 70mg (not 24/7) and weight of the tag is only 2.5mg, so
that bee can carry it and perform its normal functions. Now consider a
situation that around 50 bees are tagged in a hive and around 7 tagged
bees enter the hive at the same time, to detect these tags the size of the
antenna is already to small that it cannot provide multi-detection because
of its size. This means antenna design must have more than one antenna.
But before we design the antenna we need to see what reader should be
attached to it.

After this research work 10 weeks, we decided to use microsensys tag to

be pasted on the bee. Skyetek module M 2 was selected as the reader
equipment. Frequency of operation was selected to be in HF band 13.56
MHz. It was proposed that since skyetek module can take external
antenna matched with 50 ohms, we are going to design our own antenna
system and attach it to the reader module

3.1 Antenna Circuit
We proposed three ideas of building reader antenna in our research. By now we
had to decide which design should be implemented, building a rectangular
antenna and a single coil antenna was not advisable because they require a lot
bigger space, if we increase the size of reader antenna the transponder can’t
couple with it(reader antenna) magnetically thus we won’t be detecting anything
in that cases. So we decided to go for multilayered antenna coil built just like an
inductor but carefully calculated number of turns and its diameter. Using the
equation 9 from section 2.2.4, we made a table and a set of capacitors and
inductors to be used in such a manner that circuit resonates on the target
frequency i.e. 13.56 MHz. The table is as under

Putting the frequency 13.56 MHz we can simplify the equation as

The Formula Inductance measured Appropriate/ Available

set of Capacitors

Use in parallel

Use in parallel

Use in parallel

Use in parallel

Use in parallel

Use in parallel

Use in parallel

3.1.1 Dimensional calculations of Coil

Next step was to build and test the inductance of the antenna from LRC meter.
Following formulae and calculations show how we measured and constructed the
required antenna. We took a multi-row and a multi layered architecture of coil as
shown below. We used a wire of AWG 18 and the diameter of the wire was
calculated to be 1.0278 millimeters

Formula used:

Putting values:

• Turns = 15

• radius = 0.25 cm

• length = 1cm

• depth = 0.01 cm

We find the inductance as:

Using these calculations above we constructed the coil and found the
inductances of different coils to fall in the region of 1 , so that we
just get the idea that how many turns in actual practice we require to fall
in the target region of 13.56 MHz.

3.1.2 Configurations
It is easy to establish the fact that size and range of the antenna in our
case are very small the dimensions of transponder are such small that if
there are two bees passing very close through the entrance of the hive
and going in opposite directions, then bigger or normal patch and micro
strip antennas cannot detect their direction and may not be able to detect
them simultaneously.

Consequently we decided to build the antenna in such a way that one

antenna should take care of only one transponder and covers less area so
that one antenna corresponds to one transponder. In this way we can
detect the direction of passing transponders and can address more than
one transponder at a time.

Moreover since the size and range of the antennas are really small, we
decided to build an array of antennas, and a switching mechanism to
activate each antenna at high frequency.

We started building Inductive antennas with capacitors and resistors that

could be arranged in a number of configurations, the configurations that
were tested and followed during experiments are as under:

Configuration Name Diagram

In series with the resistor
Series resonant circuit

In Parallel with the resistor

Series resonant circuit

Parallel resonant circuit
Using SWR bridge

In series with the resistor

Parallel resonant circuit

In parallel with the resistor

Parallel resonant circuit

All of these configurations make up a resonating circuit with inductor as an

antenna. The results were shown on the screen of spectrum analyzer, and
some of the results are following sections.

As we had no idea that how the antenna circuit of the Microsensys tags
was made, so we had to have these configurations to find by hit and trial
that what type of configuration would work with the tag.

3.1.3 Measurements
Figure 3.1.1 Figure 3.1.2

Seen from the spectrum analyzer these circuits gave us some useful responses,
we had made an antenna resonating on frequencies that are nearby the required
frequency of 13.56 MHz. Figure to the left shows a paralell reonant circuit and to
the right demonstrates a series resonant circuit.

Next problem was to bring these antennas to precisely to 13.56 MHz to get the
best magnetic coupling of the antenna and tags brought in proximity. It is not
possible to change your capacitance as we didn’t have the variable capacitors
available, we could only change the inductance of the coil to increase or
decrease the resonant frequency. As the frequency of resonance is inversely

proportional to the inductance, we observed that increasing the number of turns
on the coil would increase the inductance and thus would decrease the
resonance frequency and vice-versa.

If we develop a relationship of these circuits with our previous RFID

knowledge we can say, if we use series circuit the data is carried by
current and detected through current change while if the same
components are in parallel the data is detected through voltage changes.
In these practices total 8 circuits were built resonating at frequencies near
13.56 MHz.

Now the question arises, that where exactly should this dip lie, the first
answer that comes to our minds is 13.56 MHz both the tag and reader
antennas should be resonating at this frequency, but practically the dip
should come a little before the resonating frequency, in such a manner
that 13.56 MHz falls at the start of second half of dip.

The reason behind this strange claim is that when the tag is brought close
to the reader antenna, due to mutual inductance the inductance of the
reader antenna increases. And we know that increasing the inductance
would cause the dip to shift left. If we want to bring an optimum change in
voltage (in case of parallel circuit), then we can’t place the resonant
frequency (13.56 MHz) to the left of the dip. We have to place the target
frequency slightly to the right of the dip to get maximum change of
voltage. Maximum change in voltage means maximum sensitivity of the
antenna. The microsensys tag is very small it won’t be able to cause huge
factor of magnetic coupling thus we need a better Q factor from the
antenna as well.

Next problem was to increase the Q factor and chose the configuration
that works the best in our case. We have a comparison of series and
parallel circuit response from the Network Analyzer, according these
results shown below; we decided to go for a parallel circuit. To increase
the quality factor we had to use SMD components. The reason for choosing
the parallel circuit was:

a. It seemed to have better quality factor (see figures below).

b. Its VSWR was close to 1 at nearby frequencies.

c. Mostly RFID circuits are implemented in this manner.

Figure 3.1.3 Figure 3.1.4

The figure to left shows two circuits, one in series (yellow) and one in parallel
(blue). The figure at the right hand side shows both of these circuits when
brought to close to each other. We can clearly examine that the circuit with
parallel configuration gives better(higher) quality factor.

3.2 Antenna Matching

Next we have to see whether our circuit is matched to 50 Ohms, because the
reader that has been selected takes the external antenna if it is matched to 50
Ohms. For this we used the Network Analyzer to see the S11 parameter and see
what’s our reflection or what’s the insertion loss or how much energy we lose just
because our antenna is not matched. The resistors we used in our circuits were
50 Ohms, so we had to be in the nearby region. The network analyzer from
agilent needs to be callibrated first and our antenna required the machine to be
callibrated to 50 ohms. We can see these expected results in the following


Figure 3.2.1

The response was over whelming, we fell close to the required impedance. But it
wasn’t exactly up to the mark of 50 Ohms, we can observe the impedance to be
ZL = 51.774 – j1.5524.

We have explained the method of matching in the research section 2.3. The
impedance that has to be plotted in the first step of matching as explained in
sub-section 2.3.2 is ZL found here in the measurement above. Using the smith
chart and exactly following the steps we can match the circuit we had from the
configurations we followed above.

A reader might have following questions

• What changes does matching cause as far as circuitry is concerned?

• What is a solution and results from the smith-chart?

• Number of solutions offered?

• What selects among those solutions?

We have to add an inductor in series and a capacitor in parallel. The value of

capacitor and inductor can be known from the distance (in λ) travelled to reach
the centre of the smith chart.

As told above in the theory research (section 2.3), we have to reach the centre of
the smith chart. And we have two solutions to reach the centre. The distance
travelled on the chart is measured in terms of λ. A semi circular travel is 0.5λ.

Figure 3.2.2

The values of C1 and L2 are calculated from the amount of distance that was
travelled on the imaginary curve to cut the unit circle, and then from there to the
centre of the graph gives the value of L2.

Along with that we need to know the reflection coefficient at the place where the
current reaches antenna itself.

 ZL = 51.774 – j1.5524

 Zo = 50 Ohms

 So,

 And As VSWR = 1 + IrLI/1 - IrLI

 Thus ,VSWR = 1.047

We have to get rid of this reflection coefficient, it is already too low but to
completely put it to zero we need to move from the point of admittance on the
smith chart to the auxiliary, and from there we can follow the auxiliary circle to
the centre of the smith chart.

3.3 Testing with Reader
It was about time that we would have had the reader equipment to
connect these circuits above for testing. On 28th May we received the
reader equipment, next day we switched the external antenna to ENA, this
external antenna was accompanying the reader equipment. Comparison of
the antenna made by skyetech and the antennas are show below

Figure 3.3.1 shows, the external antenna from skyetek to the left and the
antenna to the right is the one made in Fontys.

Figure 3.3.1

-2.6103db at 13.5456 MHzIIIII-43.198db at 13.5456 MHz

When a Tag :- I-code SL1,SL2 1K-bit was brought close to the antennas we

saw these results.

Figure 3.3.2

When our antenna was tested with skyetek reader, it was able to detect
almost all common RFID tags available in the market including these

• Mifare-1k DAC

• Mifare Desfire

• I-Code SL1, SL2 (1k-bits)

In the similar fashion, we tested every tag with both of the antennas first
on the network analyzer to see whether the tag causes any change to
antennas’ log-magnitude response. Then the antenna was attached to the
skyetek reader to see the tag detection on the PC. The reader equipment
has an interface for PC USB or Serial port, and software with the name
skyware needs to be installed on the corresponding machine.

We have measurements results we saw with microsensys tag

Figure 3.3.3 Figure 3.3.4

It should be noted here that the external antenna from skyetek was unable to
show any response to the microsensys tag and the antenna we built in the lab
detects the tag on the Network Analyzer’s log-magnitude graphical view but it
doesn’t detect if used with skyetek reader.

3.3.1 Results
Finally after testing and making all possible changes to circuit that could be make
in the terms of attaching SMD components and decreasing the size of the
antenna, we can say that if it is not really impossible to track the animal through
RFID, then it is really difficult. One has to put more time and resources to make it
possible. Microsensys are solving the same problem on a commercial scale; they
should have had a feasibility to put finances into this project. The design they
have proposed for reader antenna resembles our antenna design, in terms of
using multiple RFID small sized antenna.

3.4 Observations

In the wireless world, antennas are used to transmit radio frequency

energy from location A to location B in the most efficient manner. That is
they do radiate the power that is fed to them. If we were dealing with UHF
RFID systems, this hypothesis would be true. However, at 13.56MHz, the
picture is quite different. The wavelength in free space at 13.56 MHz is
22.12 meters.

A standard ground plane antenna has a length of one quarter of the

wavelength which is 5.53 meters. It has a radiation resistance close to 50
ohms. In the world of 13.56 MHz RFID systems, we are unlikely to come
close to such dimensions. And even if we do, the amount of radiated
power will remain quite small. Let us consider an example. Say that we
have a loop antenna and that its area is one square meter. The radiation
resistance is given by:

RR =31200*[A/lambda2]2

In our case, this yields RR=130 milliohms and we probably have used 4
meters of wire to construct the loop, assuming a square shape… So, if we
do not radiate energy, how do we transfer it to the tag we intend to
communicate with ? The answer is magnetic coupling. Some people refer
to RFID base stations as “couplers”. This terminology is certainly quite
appropriate in our case. We have indeed to consider the RFID system,
antenna plus tag, as a loosely coupled transformer, with the base station
antenna acting as the primary of this transformer. This concept is of
paramount importance for the system designer. One must always
remember it. The tag AND the base station “antenna” constitute THE
system, and cannot be studied separately.

The other point to remember is that if we feed five Watts to a loop RFID
“antenna”, these five watts, being NOT radiated, will have to be
dissipated somewhere…

Minimum bandwidth requirements from the tag point of view

In our case we have no idea about what the inductance of the tag
antenna, how it is coupled with our antenna. When should the tag give
best quality factor? All what we knew about the tag was that it follows ISO
15693 protocol. Through research we came to know that the this protocol
has sub carrier along at 423 KHz, possibly ON/OFF keyed if we use the
single sub carrier modulation at a data rate of 27 kBits/sec. To cater these
side bands and, not to clip them from top more than 3db we have to keep

the quality factor in the middle. If we make it too large we’ll completely
lose information and if we let it go we won’t be able to sense anything

The coupling factor (k) & the quality factor (Q)
Within the given circumstances we have another thing to worry about.
This was something totally unexpected, we consider that if RFID uses
inductive coupling, just like a transformer, then the better these inductors
are coupled the better connection we have the better data rate can be
achieved and maybe better range too. But, the case is a little different this
came to my attention when I saw the reader antenna from skyetek giving
following responses:

Figure 3.4.1

Now when the tag is brought close to the antenna

Figure 3.4.2

The graph is expected to shift left, yes it does but along with that shift we
see as the we bring them close the unexpected reverse dip goes even

Figure 3.4.3

Yes if the coupling factor is too high then we have two resonant
frequencies, to reduce this effect engineers use RF amplifiers (reader
equipment) and improve the quality factor to reduce this effect but it is
almost in evitable. We can also see a response from our own antennas
reaching to this kind of response in figures above. But, we don’t see any
kind of extra dip or even a hint of it when microsensys tag is brought to
proximity. Do we have a required coupling factor with microsensys tag
who knows, we have no information about its inductance to calculate
using following criteria

How to calculate the coupling factor:

The nice thing is that the coupling factor depends only on geometrical
parameters. The inductance values, the coils number of turns, are not
involved. Let us consider the following oversimplified representation of a
base station antenna coil and a tag antenna coil:

Figure 3.4.4
On the left, we have the base station antenna coil, with a diameter d1=2 x
r1. On the right, we have the tag antenna coil, with a diameter d2=2 x r2,
tilted along the axis by an angle a. The coupling factor can be expressed

Of course, this equation is valid only for circular coils, but the design
guidelines we are going to infer from it are valid whatever the shape. From
the equation above, it is obvious that the coupling factor will be maximum
when r1 = r2.

From more sources we came to know that geometry requirements are not
the only one to be met before we calculate the coupling factor it also
requires more information.

The coupling factor is determined by the distance between the inductors

and their relative size. It is further determined by the shape of the coils
and the angle between them. If coils are axially aligned, a displacement
causes a decrease of k. Figure 6 shows this effect for an exemplary
arrangement of planar coils with 30 mm diameter. It shows the measured
and calculated coupling factor for parallel coils at different misalignment
distances at the horizontal axis. Coupling factors in the range of 0.3 to 0.6
are typical. Note that a negative coupling factor means that the receiver
captures the magnetic flux “from behind.
The definition of the coupling factor is given by

Note: This coupling factor and quality factor conflict or clash, suggests
clearly that we cannot consider reader or tag separately, we have to thing
them as a system.

3.5 Cost Analysis
Before we calculate the total cost, we have to consider following quantities
as requirements per system (one system/hive).

• 1 U.S. dollar = 0.725320955 Euros

• 40 antennas per hive

• 10 Switching ICs per hive

• 1 Reader per Hive

Equipment Cost/ Unit Cost/Hive

PIC Controller 0.5 euros 0.5 euros

Skyetek Reader 850 euros 850 euros

ADG904BCP-R-REEL (RF 2.15 $ = 1.55 euros 21.5$ = 15.5 euors

Switch IC)

Antenna 0.5$ = 0.362 euros 20 $ = 14.5

Total Cost/Hive 880.5 euros

3.6 Problem Faced and Tackled
Although at every step and every level of this project, the report explains
the hurdles we faced and the solutions we chose to solve the
corresponding problems, we picked the most time taking and important
problems and described them as under:

a. In the start we had to select the frequency of communication. We

decided to use 13.56 MHz for two main reasons. Bee is paralyzed at
higher frequencies and microsensys tag is the only tag that could be
attached to a bee which resonates on 13.56 MHz.

b. The LCR meter measures the inductance on 1 KHz, inductances

change with frequencies. So we had to gamble with the antenna and
go with 15 turns first then by repeated measurements on Network
Analyzer, increasing or decreasing the turns we tuned the circuit to
required frequency.

c. Initially we were required to build antenna with high quality factor,

but under given equipments and the design we selected we could
not achieve Q factor more than 9. While we were struggling with
2.34. So we used ferrite cores and built the circuit again and again
to place components close to the ground. We could also try to
reduce the skin effect but that needed Litz wires. Another way of
improving quality factor we need to decrease our bandwidth, that
depends on the components used and we were limited in these
terms as well.

d. Calculating the impedance matching circuit from the smith chart is a

tedious and lengthy process. To save time we used some applets
from the internet and software for smith chart simulations.

e. Initially finding the required set of capacitors was a difficult task, we

have limited set of capacitance values available in a lab. So we
made the capacitance constant at 100 pF and made all the changes
in inductor required for resonance.

After four and a half months of experience we are around the end of
project time, under the given circumstances the main goals of project
were not achieved. Mic3 microsensys tag was not detected by our antenna
design. But it was made clear the proposed designed does work with RFID
and has the capability to detect other renowned tags. There might be
several hundred reasons for failure but to my sheer understanding I have
following conclusions.

• This project doesn’t only require a telecom engineer to make it

completely possible. As far is antenna design is concerned, my antenna
clearly detects the microsensys tags on Network Analyzer, so there
might be some problem with reader sensitivity.

• We can even say that right reader was not selected if it is unable to
sense the tag. But the situation in which reader was ordered and the
time that was given to work with the reader was not enough to be 3
weeks only.

• Simultaneously, we cannot just put the blame on reader equipment, as

I have communicated on both ends personally; I think there´s a big
chance that mic 3 tag doesn´t completely follow ISO 15693 protocol.

• If someone else has to continue the project, I would suggest him/her to

communicate with as many companies as possible. I had to wait for a
prototype of reader design which they (Microsensys) promised to hand
over in May, their maja reader (from the new prototype) follows the
same principle as I used for my antenna design. Microsensys response
wasted too much time and kept us miss directed, as I suppose we
would have arranged the same reader (Skyetek) equipment far earlier if
microsensys would have given us a clear picture.

• Antenna design and then testing requires at least two months for a
graduation trainee student I believe, while I had to perform same duty
in one month.

• We cannot take reader and tag separately, together they make up the
whole system, and we need to know the inductance and quality factor
of tag antenna. This has been one of the biggest pull back. There was
no information given on the tag other than it follows a protocol ISO

• The quality factor selection, coupling factor and size of antenna are the
most crucial aspect of this project, they should never be neglected.

For students, looking to take this project forward after me, have to work in
following fields

• The antenna size should be decreased to such an extent that size of this
reader antenna should become comparable to tag, keeping the fact in mind
that the antenna should be able to have least insertion loss and could be
compatible with the reader circuit in terms input power.

• If reader sensitivity is improved somehow, it would greatly enhance the

detection capability.

• Before someone else takes on to microsensys tag, he/she should have more
information about the tag, for instance the inductance of tag antenna, and
how could we increase its inductive coupling with reader antenna?

• They should also investigate about the protocol ISO15693, and see its power
distributions in the air on different antenna designs and implement the best

• Study RFID Antennas

Most of the references are given at the spot in this report but still there is
a lot of work that has been banked and helped by following search results,
my research mainly comes from these web-links and books

• http://news.bbc.co.uk/2/hi/science/nature/7258822.stm

• http://networks.silicon.com/mobile/0,39024665,39170213,00.htm

• http://rfid-handbook.com

• http://users.rcn.com/jkimball.ma.ultranet/BiologyPages/B/BeeDances.html

• http://www.enigmatic-

• http://www.electrocom.com.au/rfid_uhf.htm

• http://www.fuzing.com/qrx/rfid-uhf

• Radio Basics for UHF RFID from the Book, The RF in RFID: Passive UHF
RFID in Practice by Daniel M. Dobkin.

• http://www.rfid-in-action.eu/rfid-referencemodel

• RFID Coil Antenna Design Microchip

• Melexis: 13.56 MHz RFID systems and antennas design guide

Below are the links for the readers from the internet, respectively from the

• http://nl.rs-online.com/web/search/searchBrowseAction.html?

• http://nl.rs-online.com/web/search/searchBrowseAction.html?

• http://fnjudy.en.busytrade.com/products/info/864488/Hf_Rfid_Middle_Ra

• http://www.rfidglobal.org/product/2007_12/15693_rfid_long_range_read

• http://www.skyetex.com/ProductsServices/EmbeddedRFIDReaders/Skye

• http://www.zixsys.com/en/products_zrw.html

Below link gives the applets demos used in the report

• http://www.amanogawa.com/archive/LMatch/LMatch.html

• http://www.amanogawa.com/archive/transmissionA.html

• http://www.dxzone.com/cgi-bin/dir/jump2.cgi?ID=18635

Below links give access to RF Switch ICs from different sources

• http://parts.digikey.com/1/parts/638134-ic-sw-sp4t-4-1-refl-mux-

• http://www.datasheetcatalog.com/datasheets_pdf/N/J/G/1/NJG1519KC1.

• http://www.spectrummicrowave.com/xref.asp?xn=anzac

• Drones : Male Bees

• CCD : Colony Collapse Disorder

• RFID : Radio Frequency Identification

• PLC : Programmable Logic Controller

• VSWR : Voltage Standing Wave Ratio

• HF : High Frequency, a band in radio frequencies

• UHF : Ultra High Frequency, a higher frequency band in radio frequency


• VHF : Very High Frequency, a band of frequency in radio band

• RF : Radio Frequency

• IC : Integrated Circuit

• ASK : Amplitude Shift Keying

• PSK : Phase Shift Keying

• FSK : Frequency Shift Keying

• SMD : Surface Mount Device, a very small circuit component.

• SWR : Standing Wave Ratio

• Hive : Home for bees

• Acoustic : Audible

• Piezoelectric features : sensitive to electromagnetic waves

• ENA : Enhanced Network Analyzer

• SA : Spectrum Analyzer

• TG : Tracking Generator

Annex A-1
Project BEE Tracking

Plan of Approach
Proposed plan of design and implementation of the communication system for
bee tracking is created after keen ponder and analysis of the problem, plan
states itself as follows:

• Project Phases
• Background Information
• Formulation of objectives or definition of the problem
• Project Scope
• Scheduling
• Risk Analysis
• Drawing up a project pictorial plan
The project has been divided into following phases

1. Design and assembly of detection system.

2. Implementation of the detection system.

3. GSM based communication system design.

4. Closing report and final presentation.

Background Information
Only for the sake of background information and complete understanding of the
project, the whole topic of RFID was studied during past six weeks. The research
along with meeting with Mentors concluded many fruitful results. The emphasis
of research lies on following agendas

1. RFID communication basic

2. RFID components.

3. RFID frequency analysis

4. RFID hardware design.

5. RFID Antenna Design.

6. RFID research papers for different fields of life

7. Bee behavior and natural traits

Formulation of objectives or definition of the problem
The objective is that whenever a bee enters or leaves the facility (HIVE), it is
registered. The goal is to track the bees through a proposed communication
system of RFID and GSM. The problem statement mainly comprises of finding out
a way to tag the bees and to detect their entrance through a detection system of
antenna and reader module. Moreover problem continues to emphasize on
communication system that includes a GSM system which enables the data of
incoming and outgoing bees to be transferred to a main server station.

Project Scope
Scope of the project stays within the circle of communication system and
antenna design

Exactly corresponding to project phases a tentative schedule has been
established as under

1. Design and assembly of detection system.

Sub phases Purpose Time span

Communication with Protocol information, 1 week
microsensys and J.Roodt connection establishment
procedures, inquire if they
have a suitable reader
module for us?
Reader Module Research and testing 2 weeks
Antenna Design for Reader To see if the system works 2 weeks
module only for detection before we use the system
purposes on site.
2. Implementation of the detection system.

Sub phases Purpose Time span

Posting tags on bees, glue To see practically if tags 1 week
analysis fulfill the physical
Fabrication of Antenna in To see if the proposed and 2 weeks
the Bee hive designed antenna really
works on bee hive
3. GSM based communication system design.

Sub phases Purpose Time span

Study /Research on GSM To see how do they work, 2 weeks
modules to arrange data transfer
from bee hive to server
Assembly of 4 working To build the final part of 2 weeks
module along with one communication system
server to collect and
process their data.
4. Closing report and final presentation.

Task Purpose Time span
Reporting and To put all the work on 1 week ( starting on 9th
Presentation record and in black and June)

Note: The schedule starts from 16th March 2009 and last till 9th June

Risk Analysis
Each step and each phase has its own risks and loop holes, risks can never be
abolished completely but their probability can be minimized through efficient
approach of solution. Corresponding to each step and phase, hunch and risk are
stated as under:

Phase-1: Design and assembly of detection system.

While designing the reader

for the micro-sensys tag, wee need to know what protocol does microsensys use,
if the protocol is used by some other reader we can use those too, but if the
microsensys have their special protocol for connection establishment then we will
have to order a reader from them. Even if we find the right reader for our
corresponding tag, antenna design is still a fragile process, its matching includes
many attributes to make it ambiguous like electromagnetic inductance and
production of eddy currents.

Phase-2: Implementation of the detection system.

Posting tags and fabricating

antenna on the hive would cause great changes in the system that has already
been tested for detection (and is seen working perfectly well before we use the
system in bee hive), factors like metallic materials in the surroundings affect the
matching and tuning a lot.

Phase-3: GSM based communication system design.

Establishment of
both RFID and GSM system may be easier but here these system not only
communicate independently but also that they communicate with each-other,
this is one of the blind corners of project till now and needs to be looked down
upon during the whole process.

Drawing up the project pictorial plan

NOTE: Each hive would have around 20 tagged bees initially.

Annex A-2

Data Sheets