FPGA Central

hi

i am into a little confusion

the itu-t v.23 says that

"Modulation rates and characteristic frequencies for the forward
data-transmission channel are

for Mode2 up to 1200 bauds

1300 Hz (symbol 1, mark)

and

2100 Hz (symbol 0, space) "

According to my preception 1300 Hz means - if 1300 cycles are received
in one second it means a 'one' is received
and similerly 2100 Hz means if 2100 cycles are received in one second,
a 'zero' is received.

and again according to my perception 1200 bauds mean that the line
changes state 1200 times...


1300 cycles per second means '1' and within that one second the line
changes state for 1200 times. i dont get it. what is this?

please help me with this. and correct me where i am wrong.

thanks in advance.

Somia

Somia wrote:
> hi
>
> i am into a little confusion
>
> the itu-t v.23 says that
>
> "Modulation rates and characteristic frequencies for the forward
> data-transmission channel are
>
> for Mode2 up to 1200 bauds
>
> 1300 Hz (symbol 1, mark)
>
> and
>
> 2100 Hz (symbol 0, space) "
>
> According to my preception 1300 Hz means - if 1300 cycles are
received
> in one second it means a 'one' is received
> and similerly 2100 Hz means if 2100 cycles are received in one
second,
> a 'zero' is received.
>
> and again according to my perception 1200 bauds mean that the line
> changes state 1200 times...
>
>
> 1300 cycles per second means '1' and within that one second the line
> changes state for 1200 times. i dont get it. what is this?
>
> please help me with this. and correct me where i am wrong.
>
> thanks in advance.
>
> Somia


Hello Somia,

Baud basically refers to the number of state changes per second. In
this case, the "state" is frequency. So for this kind of FSK, a single
bit's state lasts for 1/1200th of a second. So when a "1" is sent, 1300
Hz is output for 1/1200th of a second. So barely over a single cycle of
the 1300Hz is output. Likewise for a "0" bit, 2100 hertz is output for
1/1200th of a second. So in this case you get nearly 2 cycles.

A decent FSK modulator will be phase continuous and have some kind of
slew rate limit, so the change of frequency from one bit to the next
isn't instantaneous. But when viewed in the frequency domain, the FSK
signal will smoothly move between 1300 and 2100 Hz.

IHTH,
Clay

"Somia" <[email protected]> wrote in message
news:[email protected] ups.com...
> hi
>
> i am into a little confusion
>
> the itu-t v.23 says that
>
> "Modulation rates and characteristic frequencies for the forward
> data-transmission channel are
>
> for Mode2 up to 1200 bauds
>
> 1300 Hz (symbol 1, mark)
>
> and
>
> 2100 Hz (symbol 0, space) "
>
> According to my preception 1300 Hz means - if 1300 cycles are received
> in one second it means a 'one' is received
> and similerly 2100 Hz means if 2100 cycles are received in one second,
> a 'zero' is received.
>
> and again according to my perception 1200 bauds mean that the line
> changes state 1200 times...
>
>
> 1300 cycles per second means '1' and within that one second the line
> changes state for 1200 times. i dont get it. what is this?
>
> please help me with this. and correct me where i am wrong.
>
> thanks in advance.

Somia,

Clay gave you a good answer.
(I don't know how practical it would be to detect frequency in the span of
one cycle though....)

When you say above:
">According to my preception 1300 Hz means - if 1300 cycles are received
> in one second it means a 'one' is received"

Perhaps you understand this and it was only a matter of expression. But
just to make sure you have no misperception:

Above, you seems to imply that your understanding is a symbol exists in the
channel for 1 second. That's not necessary. For a symbol to be received it
is only necessary that the frequency of the signal to be detected. That may
be in 1/1200th of a second as implied in the spec.

1300 cycles per second is a *rate*. It's not tied to a 1-second interval
it's just how rates are expressed with "per second" being a convenient
reference. So, the existence of a 1300Hz tone could be detected in a very
short time and probably a time greater than 1/1300th of a second unless the
signal to noise ratio is very high indeed.

So, you might have said:
If a 1300Hz tone is detected, it means a "one" is received. Then the amount
of time given for the detection of the 1300Hz tone to occur is a matter of
system design and perhaps the standards.

Fred

Somia wrote:
> hi
>
> i am into a little confusion
>
> the itu-t v.23 says that
>
> "Modulation rates and characteristic frequencies for the forward
> data-transmission channel are
>
> for Mode2 up to 1200 bauds
>
> 1300 Hz (symbol 1, mark)
>
> and
>
> 2100 Hz (symbol 0, space) "
>
> According to my preception 1300 Hz means - if 1300 cycles are received
> in one second it means a 'one' is received
> and similerly 2100 Hz means if 2100 cycles are received in one second,
> a 'zero' is received.
>
> and again according to my perception 1200 bauds mean that the line
> changes state 1200 times...
>
>
> 1300 cycles per second means '1' and within that one second the line
> changes state for 1200 times. i dont get it. what is this?
>
> please help me with this. and correct me where i am wrong.
>
> thanks in advance.
>
> Somia

Clay's answer is complete. You need to be sensitive to the difference
between rate and count. You can travel at 30 kilometers per hour for
only a minute, or all day.

Jerry
--
Engineering is the art of making what you want from things you can get.
¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯ ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯

Clay wrote:

(snip)

> Baud basically refers to the number of state changes per second. In
> this case, the "state" is frequency. So for this kind of FSK, a single
> bit's state lasts for 1/1200th of a second. So when a "1" is sent, 1300
> Hz is output for 1/1200th of a second. So barely over a single cycle of
> the 1300Hz is output. Likewise for a "0" bit, 2100 hertz is output for
> 1/1200th of a second. So in this case you get nearly 2 cycles.

> A decent FSK modulator will be phase continuous and have some kind of
> slew rate limit, so the change of frequency from one bit to the next
> isn't instantaneous. But when viewed in the frequency domain, the FSK
> signal will smoothly move between 1300 and 2100 Hz.

I believe this is right for FSK. A PLL can follow the change, and
can demodulate FSK fairly easily. FSK can be demodulated much faster
than the frequency difference, which isn't so obvious. Bell 103A has
a 200Hz shift and is commonly used at 300 baud.

For phase modulation (PSK, PAM, etc.) the changes are discontinuous,
and that is important. Clock recovery can be done by squaring and
filtering the signal, assuming a sufficient density of phase changes.
Scramblers are used to ensure a high density of phase changes even
with low entropy signals.

-- glen