# How can an antenna be used for both rx and tx?

M

#### Michael

Jan 1, 1970
0
share an antenna between rx and tx circuits? I'm thinking of something
like a cell phone, for example. My first thought was that the device
would allot a certain amount of time for sending and a certain amount
of time for receiving (say, every other 5 microseconds, or something
like that). But then making every other device match up with that
would be rather difficult methinks. So then my next idea was that the
device is always both sending and receiving and the received signals
are just extracted from the antenna's signal. But then I got to
thinking about exactly how this would be done. You'd want to subtract
the signal being transmitted from the overall signal present at the
antenna - but where would you be getting the antenna signal? Wouldn't
you be getting it from the antenna connection, where the transmitter
circuitry is connected to the antenna? So how would you be able to
pull out the rx signal?

I suspect this is a terribly uninformed question, but I'm really a
robotics/sensors guy - all this RF stuff seems like black magic to
me

-Michael

A

#### Anthony Fremont

Jan 1, 1970
0
Michael said:
share an antenna between rx and tx circuits? I'm thinking of something
like a cell phone, for example. My first thought was that the device
would allot a certain amount of time for sending and a certain amount
of time for receiving (say, every other 5 microseconds, or something
like that). But then making every other device match up with that
would be rather difficult methinks. So then my next idea was that the

It's way simpler than that.
device is always both sending and receiving and the received signals
are just extracted from the antenna's signal. But then I got to
thinking about exactly how this would be done. You'd want to subtract
the signal being transmitted from the overall signal present at the
antenna - but where would you be getting the antenna signal? Wouldn't
you be getting it from the antenna connection, where the transmitter
circuitry is connected to the antenna? So how would you be able to
pull out the rx signal?

Everything is happening at the same time usually (full duplex). First off,
the transmitter and receiver are on different frequencies. This is usually
careful filtering prevents the transmitter from getting in. Google "antenna
duplexer".
I suspect this is a terribly uninformed question, but I'm really a
robotics/sensors guy - all this RF stuff seems like black magic to
me

When it comes to antennas, you're not too far off. ;-)

M

#### MassiveProng

Jan 1, 1970
0
When it comes to antennas, you're not too far off. ;-)

But you are.

M

#### Michael

Jan 1, 1970
0
Everything is happening at the same time usually (full duplex). First off,
the transmitter and receiver are on different frequencies. This is usually
careful filtering prevents the transmitter from getting in. Google "antenna
duplexer".

Wait - you're saying when an antenna is shared between rx and tx, that
the rx and tx must be on different frequencies? Is that always the
case? Surely, though, they're still at almost the same frequency,
right?

H

#### Homer J Simpson

Jan 1, 1970
0
Wait - you're saying when an antenna is shared between rx and tx, that
the rx and tx must be on different frequencies? Is that always the
case? Surely, though, they're still at almost the same frequency,
right?

If they were on exactly the same frequency it wouldn't work - you'd have to
switch from send to receive "Over"

They are sufficiently far apart that the receiver can ignore the
transmitter.

A

#### Anthony Fremont

Jan 1, 1970
0
Michael said:
Wait - you're saying when an antenna is shared between rx and tx, that
the rx and tx must be on different frequencies?

No I didn't say that. They could be on the same frequency, but then you
would have to go with some type of half duplex as you described in your
first scenario.
Is that always the case? Surely, though, they're still at almost the same
frequency, right?

It depends, for example 2 meter (144-148MHz) HAM radio repeaters usually use
a 600kHz offset where the transmitter and receiver differ in frequency by
600 kHz. 70cm (~450MHz) repeaters have a 5MHz offset. When you get up
around 900MHz, the offset is usually around 45MHz IIRC. As you can see, as
the frequency goes up so does the spacing. There would likely be many radio
systems out there that would not comply exactly with what I just stated.
For example, odd offsets are not uncommon in the commercial radio world.

D

Jan 1, 1970
0
M

Jan 1, 1970
0
C

#### cledus

Jan 1, 1970
0
Michael said:
share an antenna between rx and tx circuits? I'm thinking of something
like a cell phone, for example. My first thought was that the device
would allot a certain amount of time for sending and a certain amount
of time for receiving (say, every other 5 microseconds, or something
like that). But then making every other device match up with that
would be rather difficult methinks. So then my next idea was that the
device is always both sending and receiving and the received signals
are just extracted from the antenna's signal. But then I got to
thinking about exactly how this would be done. You'd want to subtract
the signal being transmitted from the overall signal present at the
antenna - but where would you be getting the antenna signal? Wouldn't
you be getting it from the antenna connection, where the transmitter
circuitry is connected to the antenna? So how would you be able to
pull out the rx signal?

I suspect this is a terribly uninformed question, but I'm really a
robotics/sensors guy - all this RF stuff seems like black magic to
me

-Michael

Your intuition is very good Michael. Actually, both techniques are
used. Various cell phone technologies incorporate both "Time Division
Multiplexing" and "Frequency Division Multiplexing".

In TDM, the TX is on briefly, then the radio switches to RX mode in a
cyclic manner. The transmitter and receiver antenna terminals are
essentially connected to ports on a SPDT switch. The common pole is
connected to the antenna. Timing and synchronization are critical to the
success of this method. And enough "slop" to account for propagation
delays must be factored in. The advantage is that lossy filters are not
required to separate RX and TX. Instead, a less-lossy TX-RX switch may
be used. This works extremely well as evidenced by the success of GSM
technology which uses TDM. Another side benefit is that since the TX is
active for only a small percentage of the time, excellent battery life
can be obtained.

Other technologies, such as the archaic AMPS system and the various CDMA
technologies, continuously receive and transmit at the same time, but at
different frequencies (hence FDM). Filters are used to separate out the
two frequency bands. One terminal of each of the RX and TX filters is
coupled to a common antenna. The disadvantage is that these filters
(aka duplexers) tend to be lossier than the switches used for TDM
technologies. But it doesn't impose the timing/sync restrictions that
TDM must account for.

T

#### Tam/WB2TT

Jan 1, 1970
0
Michael said:
share an antenna between rx and tx circuits? I'm thinking of something
like a cell phone, for example. My first thought was that the device
would allot a certain amount of time for sending and a certain amount
of time for receiving (say, every other 5 microseconds, or something
like that). But then making every other device match up with that
would be rather difficult methinks. So then my next idea was that the
device is always both sending and receiving and the received signals
are just extracted from the antenna's signal. But then I got to
thinking about exactly how this would be done. You'd want to subtract
the signal being transmitted from the overall signal present at the
antenna - but where would you be getting the antenna signal? Wouldn't
you be getting it from the antenna connection, where the transmitter
circuitry is connected to the antenna? So how would you be able to
pull out the rx signal?

I suspect this is a terribly uninformed question, but I'm really a
robotics/sensors guy - all this RF stuff seems like black magic to
me

-Michael
Different systems will use different methods, but the GSM frame structure is
well documented, and you should be able to find answers to all you questions
on the web. Search for GSM.

Tam

J

#### John Larkin

Jan 1, 1970
0
share an antenna between rx and tx circuits? I'm thinking of something
like a cell phone, for example. My first thought was that the device
would allot a certain amount of time for sending and a certain amount
of time for receiving (say, every other 5 microseconds, or something
like that). But then making every other device match up with that
would be rather difficult methinks. So then my next idea was that the
device is always both sending and receiving and the received signals
are just extracted from the antenna's signal. But then I got to
thinking about exactly how this would be done. You'd want to subtract
the signal being transmitted from the overall signal present at the
antenna - but where would you be getting the antenna signal? Wouldn't
you be getting it from the antenna connection, where the transmitter
circuitry is connected to the antenna? So how would you be able to
pull out the rx signal?

I suspect this is a terribly uninformed question, but I'm really a
robotics/sensors guy - all this RF stuff seems like black magic to
me

-Michael

The old analog cell phones shared the antenna on both ends, and
routed the transmit signal out and the receive sig in, based on
frequency differences. Digital cell phones pack the voice signals into
packets, and time-share the antenna so they don't have to transmit and

Radar was the classic single-frequency, single-antenna system.
Diplexers, using waveguide tricks and gas-discharge t-r switches, kept
the megawatt transmit pulses from blowing out the delicate receive
diode mixers.

John

M

#### MassiveProng

Jan 1, 1970
0
I disagree. Just where do you believe he was wrong?

In calling it black magic.

And just in the fact the he is simply off... Too far.

M

#### MassiveProng

Jan 1, 1970
0
Sorry, but hes spot on.
The reference was to the "black magic" remark, and in that, no he
isn't.

M

#### [email protected]

Jan 1, 1970
0
share an antenna between rx and tx circuits? I'm thinking of something
like a cell phone, for example. My first thought was that the device
would allot a certain amount of time for sending and a certain amount
of time for receiving (say, every other 5 microseconds, or something
like that). But then making every other device match up with that
would be rather difficult methinks. So then my next idea was that the
device is always both sending and receiving and the received signals
are just extracted from the antenna's signal. But then I got to
thinking about exactly how this would be done. You'd want to subtract
the signal being transmitted from the overall signal present at the
antenna - but where would you be getting the antenna signal? Wouldn't
you be getting it from the antenna connection, where the transmitter
circuitry is connected to the antenna? So how would you be able to
pull out the rx signal?

I suspect this is a terribly uninformed question, but I'm really a
robotics/sensors guy - all this RF stuff seems like black magic to
me

-Michael

Ah, but haven't you wondered about how you could send an analog phone
signal in both direction on one twisted pair? ;-)

L

#### Lionel

Jan 1, 1970
0
The reference was to the "black magic" remark, and in that, no he
isn't.

He said it was *like* black magic, dipshit, not that it *is* black
magic.

M

#### MassiveProng

Jan 1, 1970
0
He said it was *like* black magic, dipshit, not that it *is* black
magic.

No, idiot. The other poster said that, and he agreed.

I wouldn't expect an adolescent retard to be able to comprehend the
written word, however, much less know who wrote what.

L

#### Lionel

Jan 1, 1970
0
No, idiot. The other poster said that, and he agreed.

I didn't specify who first said, TinyDong, just that it was correct, &
that.
I wouldn't expect an adolescent retard to be able to comprehend the
written word, however, much less know who wrote what.

....says the dipshit who was too stupid to realise what he was being
corrected for.

M

#### MassiveProng

Jan 1, 1970
0
...says the dipshit who was too stupid to realise what he was being
corrected for.

Said the utter idiot that couldn't handle correction if his life
depended on it.

I'm sure you'll end up in the lake of fire, boy.

L

#### Lionel

Jan 1, 1970
0
Said the utter idiot that couldn't handle correction if his life
depended on it.

I'm sure you'll end up in the lake of fire, boy.

You think I'm religious, Fuckchops? What a fucking moron.

J

Jan 1, 1970
0
Ah, but haven't you wondered about how you could send an analog phone
signal in both direction on one twisted pair? ;-)

This is a very good point, although to anyone thinking about it: The
difference with radio antennas is that you typically have, e.g., one watt
being transmitted and say, a microwatt being received. With telephone
lines, the transmitted and received poweres are roughly comparable.

M
Replies
21
Views
2K
joseph2k
J
Replies
9
Views
3K
Replies
17
Views
355
Replies
10
Views
874
Replies
11
Views
1K