new babies to RFID design want to ask some question

[nospam]

I am a software engineer and I know nothing about hardware
design....

Recently I am asked to collect some information about the components
of
RFID reader that can read ISO-18000-6 (Type A) tag. The tag works in
the UHF
band (860-930 MHz). I have some question but I don't have any friend
that
have RFID experience. I would really appreciate if you could kindly
answer
the questions below.....

1) The operational principle of this kind of tag is "backscatter
coupling". However, I am not sure how the reader distinguish the
forward
signal and the backscattered signal, which of the same frequency. I
read
some information from internet and someone mentioned about the
"directional
coupler". Is directional coupler really used in a typical UHF RFID
reader?
If so, could you kindly recommend a brand and the model number of
directional coupler that can be used in RFID reader?

2) The reader is supposed to be used in a cashier of shop. The range
of
distance of the reader to the tag is about 30cm. Do you have some
typical
power values that the reader need to send and the power of the
backscattered
signal? I need this kind of figure because I was ask to choose the
transceiver IC and I wonder if the power is enough. Do you have any
recommendation of transceiver IC?

3) About the antenna. I found some paper mention about 4 phased
array, which is good for sending RF signal in one direction. Have
anyone used it before? If yes, where can I found it? Any brand name or
model number recommend?

Thanks in advance.

NOSPAM

 

[Luo XiaoZen]

1) not sure;
2) check with Broadcom;
3) check with TI at
http://www.ti.com/tiris/docs/products/antennas/antennas.shtml

Luo

 

[Bob Agnew]

What frequency range does the directional coupler have to work in. The
reason that I ask is that, at microwave frequencies, one can print a pattern
on a PC board to implement a direcvtional coupler. Also one can buy MMIC
and hybrid coupler components.

Another way to implement a -3db. dircetional coupler is to use a
quarter-wavelength of Sage wireline.

There are several books and computer programs that can help you determine
the printed circuit line widths and spacings. I haven't had to design one
for
several decades now, so things may have changed a little.

 

[Andy]

Andy replies:

Backscatter Reflection --

The incident power hits a modulated reflector. The reflector is
varied between
a state of high reflection and low reflection at a data rate, often
using FSK.
The TIRIS system uses 600 Khz and 1200 Khz for example......
The receiver picks up the incident signal as well as the
reflected signal.
The reflected signal contains sidebands equal to the modulation rate,
while the
incident signal does not. Typically the receiver is a homodyne
receiver, using
I and Q detectors,and the output is the original data modulating signal
which
was applied to the modulated reflector . The data is then processed...

RANGE- For 30 cm, the powere transmitted to the backscatter reflector
can
be in the milliwatt region. 50 mw should be plenty. This can be
calculated
using the reflection characterics of the reflector and the RADAR
equation.
Refer to the article " Data Communications by means of
Reflective Transmission"
which appeared in Microwaves and Wireless, by C A Sharpe ----
sometime around
1994 or so.... The equations are given there, as well as an
explanation of the
TIRIS system and the Titile 21 requirements for California
toll tags.
The power required depends on the change in RCS (radar cross
section) of
the modulated reflector ( the TAG). With an RCS change of
200 sq cm, only a
few mw will be needed at 30 cm... This change can be
accomplished using
a simple dipole at 915 Mhz which is opened and shorted in the
middle
with a diode to do the backscatter modulation.

ANTENNA - Typically a patch arrangement is used at this frequency
range. The
TIRIS system originally used an array of 4 patches in a line to
give the necessary
beamwidths. It was manufactured by SEAVEY ENGINEERING in
Massachusetts.
I am not sure if the antenna presently used has changed, but it
was a good,
robust design....

DIRECTIONAL COUPLER --- A circulator would be a good choice, but they
are
expensive. A microstrip or stripline directional coupler is a
suitable alternative.
They are inexpensive, but not as sensitive. Since your distance
is so small,
you can afford to lose the sensitivity. If you tap the microstrip
at quarter
wavelengths with a diode, you can do the I and Q on a single
coupler.
It is not efficient, but is very cheap, and works great at small
distances.

Good luck with this. I have my doubts that a software designer has

sufficient background to create and efficient design for the RF
circuitry involved
tho. One needs a thorough and wide background to make the
approximations
and shortcuts needed to create a cheap and dirty transceiver, in my
opinnion.
Perhaps you can copy an existing desigh. If you can get a schematic of

the TIRIS programmer for the TAG, you would have a good starting
point, but
you need to get hold of one to see how it is mechanically laid out. At
microwave,
the layout is much more important than the schematic.....

C A Sharpe (Andy)

Retired Systems Engineer and inventor of the TIRIS 915 tag system for
Texas Instruments.

 

[Luo XiaoZen]

Andy replies:

Backscatter Reflection --

The incident power hits a modulated reflector. The reflector is
varied between
a state of high reflection and low reflection at a data rate, often
using FSK.
The TIRIS system uses 600 Khz and 1200 Khz for example......
The receiver picks up the incident signal as well as the
reflected signal.
The reflected signal contains sidebands equal to the modulation rate,
while the
incident signal does not. Typically the receiver is a homodyne
receiver, using
I and Q detectors,and the output is the original data modulating signal
which
was applied to the modulated reflector . The data is then processed...

RANGE- For 30 cm, the powere transmitted to the backscatter reflector
can
be in the milliwatt region. 50 mw should be plenty. This can be
calculated
using the reflection characterics of the reflector and the RADAR
equation.
Refer to the article " Data Communications by means of
Reflective Transmission"
which appeared in Microwaves and Wireless, by C A Sharpe ----
sometime around
1994 or so.... The equations are given there, as well as an
explanation of the
TIRIS system and the Titile 21 requirements for California
toll tags.
The power required depends on the change in RCS (radar cross
section) of
the modulated reflector ( the TAG). With an RCS change of
200 sq cm, only a
few mw will be needed at 30 cm... This change can be
accomplished using
a simple dipole at 915 Mhz which is opened and shorted in the
middle
with a diode to do the backscatter modulation.

ANTENNA - Typically a patch arrangement is used at this frequency
range. The
TIRIS system originally used an array of 4 patches in a line to
give the necessary
beamwidths. It was manufactured by SEAVEY ENGINEERING in
Massachusetts.
I am not sure if the antenna presently used has changed, but it
was a good,
robust design....

DIRECTIONAL COUPLER --- A circulator would be a good choice, but they
are
expensive. A microstrip or stripline directional coupler is a
suitable alternative.
They are inexpensive, but not as sensitive. Since your distance
is so small,
you can afford to lose the sensitivity. If you tap the microstrip
at quarter
wavelengths with a diode, you can do the I and Q on a single
coupler.
It is not efficient, but is very cheap, and works great at small
distances.

Good luck with this. I have my doubts that a software designer has

sufficient background to create and efficient design for the RF
circuitry involved
tho. One needs a thorough and wide background to make the
approximations
and shortcuts needed to create a cheap and dirty transceiver, in my
opinnion.
Perhaps you can copy an existing desigh. If you can get a schematic of

the TIRIS programmer for the TAG, you would have a good starting
point, but
you need to get hold of one to see how it is mechanically laid out. At
microwave,
the layout is much more important than the schematic.....

Good point!