>> "billykao" <billy.kao@gmail.com> writes:
>>
>>>> On May 14, 8:42 pm, cincy...@gmail.com wrote:
>>>>
>>>>> You should be fine, assuming that the signal you sampled is
>>>>> appropriately bandlimited (i.e. you should bandpass filter the 55-85
>>>>> MHz band before sampling). Your sampling frequency is 100 MHz, so
> the
>>>>> signal of interest will "alias" down to a center frequency of 30 MHz
>>>>> after sampling. Your signal will then occupy the 15-45 MHz band in
> the
>>>>> sampled signal, which is all below the Nyquist rate. No problems
>>>>> there.
>>>>>
>>>>> Jason
>>>> To the original poster: the keyword is "bandpass sampling".
>>>>
>>>>
>>> For 30 MHz I signal, the maximum available bandwidth is 15 MHz. So is
> Q
>>> signal. If using I and Q, can I make it 30 MHz bandwidth? Thank you.
>> Yes. You essentially utilize the bandwidth in -Fs/2 to +Fs/2 with an
>> I/Q signal, whereas with a real signal the usable bandwidth is from 0
>> to +Fs/2.
>> --
>> % Randy Yates % "And all that I can do
>> %% Fuquay-Varina, NC % is say I'm sorry,
>> %%% 919-577-9882 % that's the way it goes..."
>> %%%% <yates@ieee.org> % Getting To The Point', *Balance of
> Power*, ELO
>> http://home.earthlink.net/~yatescr
>>
>
> I appreciate your answer. Could you explain why Q signal shows bandwidth
> from -Fs/2 to 0?
It doesn't. The two together give you the higher bandwidth.
> Is there any article or book talking about this? Thank
> you very much.
Most books on sampling will explain. I like Richard Lyons:
"Understanding Digital Signal Processing" for a serious introduction.
(Rick is a comp.dsp regular.
Beiefly, the I and Q samples are independent. 30,000 I samples/sec and
30,000 Q samples/sec together provide 60,000 samples/sec. That's
(barely!) adequate for a 30 KHz bandwidth.
Jerry
--
Engineering is the art of making what you want from things you can get.
¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
Reply by Jerry Avins●May 15, 20072007-05-15
billykao wrote:
>> On May 14, 8:42 pm, cincy...@gmail.com wrote:
>>
>>> You should be fine, assuming that the signal you sampled is
>>> appropriately bandlimited (i.e. you should bandpass filter the 55-85
>>> MHz band before sampling). Your sampling frequency is 100 MHz, so the
>>> signal of interest will "alias" down to a center frequency of 30 MHz
>>> after sampling. Your signal will then occupy the 15-45 MHz band in the
>>> sampled signal, which is all below the Nyquist rate. No problems
>>> there.
>>>
>>> Jason
>> To the original poster: the keyword is "bandpass sampling".
>>
>>
>
> For 30 MHz I signal, the maximum available bandwidth is 15 MHz. So is Q
> signal. If using I and Q, can I make it 30 MHz bandwidth? Thank you.
Yes. In that special case, you may add.
Jerry
--
Engineering is the art of making what you want from things you can get.
¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
Reply by Randy Yates●May 15, 20072007-05-15
"billykao" <billy.kao@gmail.com> writes:
> [...]
> I appreciate your answer. Could you explain why Q signal shows bandwidth
> from -Fs/2 to 0?
Sure, but it may take some studying on your part to understand it. It's
not extremely difficult, though.
It's not quite that Q supplies the info "from -Fs/2 to 0". In reality,
ALL sampled signals have bandwidth (in the digital domain) from -Fs/2
to +Fs/2. However, if you read up on the properties of the discrete-time
Fourier transform, you will see that if x[n] is a real signal, then
X(F) is "Hermitian symmetric." Basically that means that the information
in the frequency domain from -Fs/2 to 0 is a copy of that from 0 to +Fs/2.
However, when x[n] is complex, then the resulting spectrum X(F) is no
longer Hermitian symmetric, and the spectrum from -Fs/2 to 0 carries
more information than the spectrum from 0 to +Fs/2 alone.
> Is there any article or book talking about this? Thank
> you very much.
Any textbook on linear systems will have this. My favorite (though
now somewhat dated) is [signalsandsystems].
--Randy
@BOOK{signalsandsystems,
title = "{Signals and Systems}",
author = "{Alan~V.~Oppenheim, Alan~S.~Willsky, with Ian~T.~Young}",
publisher = "Prentice Hall",
year = "1983"}
--
% Randy Yates % "She tells me that she likes me very much,
%% Fuquay-Varina, NC % but when I try to touch, she makes it
%%% 919-577-9882 % all too clear."
%%%% <yates@ieee.org> % 'Yours Truly, 2095', *Time*, ELO
http://home.earthlink.net/~yatescr
Reply by lanbaba●May 15, 20072007-05-15
>>>>"billykao" <billy.kao@gmail.com> writes:
>>>>
>>>>>>On May 14, 8:42 pm, cincy...@gmail.com wrote:
>>>>>>
>>>>>>> You should be fine, assuming that the signal you sampled is
>>>>>>> appropriately bandlimited (i.e. you should bandpass filter the
>>55-85
>>>>>>> MHz band before sampling). Your sampling frequency is 100 MHz, so
>>>the
>>>>>>> signal of interest will "alias" down to a center frequency of 30
>>MHz
>>>>>>> after sampling. Your signal will then occupy the 15-45 MHz band
in
>>>the
>>>>>>> sampled signal, which is all below the Nyquist rate. No problems
>>>>>>> there.
>>>>>>>
>>>>>>> Jason
>>>>>>
>>>>>>To the original poster: the keyword is "bandpass sampling".
>>>>>>
>>>>>>
>>>>>
>>>>> For 30 MHz I signal, the maximum available bandwidth is 15 MHz. So
>>is
>>>Q
>>>>> signal. If using I and Q, can I make it 30 MHz bandwidth? Thank
>>you.
>>>>
>>>>Yes. You essentially utilize the bandwidth in -Fs/2 to +Fs/2 with an
>>>>I/Q signal, whereas with a real signal the usable bandwidth is from 0
>>>>to +Fs/2.
>>>>--
>>>>% Randy Yates % "And all that I can do
>>>>%% Fuquay-Varina, NC % is say I'm sorry,
>>>>%%% % that's the way it goes..."
>>>>%%%% <yates@ieee.org> % Getting To The Point', *Balance of
>>>Power*, ELO
>>>>http://home.earthlink.net/~yatescr
>>>>
>>>
>>>I appreciate your answer. Could you explain why Q signal shows
>bandwidth
>>>from -Fs/2 to 0? Is there any article or book talking about this?
>>Thank
>>>you very much.
>>>
>>>
>>>
>>>_____________________________________
>>>Do you know a company who employs DSP engineers?
>>>Is it already listed at http://dsprelated.com/employers.php ?
>>>
>>
>>
>>It is the interrelation between the I and Q channels that is able to
>>distinguish between positive and negative frequencies. Search for
>>"negative frequency" and you can find some lengthy discussions of it on
>>this group.
>>
>>-Joe
>>
>>_____________________________________
>>Do you know a company who employs DSP engineers?
>>Is it already listed at http://dsprelated.com/employers.php ?
>>
>
>In some commercial RF digitizers, like e.g. Aeroflex PXI-3035 and
National
>Instruments PXI-5661, the real-valued IF signal is digitized with a
>sampling frequency of 100 MHz, followed by IQ-separation and digital
>downconversion (DDC) to the baseband. I remember the maximum signal
>bandwidth, that can be acquired, in both cases is 20 MHz. Still
wondering
>if it is a theoretical limit of implementation constraint.
>
>LBB
>
>
>_____________________________________
>Do you know a company who employs DSP engineers?
>Is it already listed at http://dsprelated.com/employers.php ?
>
"theoretical limit or implementation constraint" I mean.
_____________________________________
Do you know a company who employs DSP engineers?
Is it already listed at http://dsprelated.com/employers.php ?
Reply by lanbaba●May 15, 20072007-05-15
>>>"billykao" <billy.kao@gmail.com> writes:
>>>
>>>>>On May 14, 8:42 pm, cincy...@gmail.com wrote:
>>>>>
>>>>>> You should be fine, assuming that the signal you sampled is
>>>>>> appropriately bandlimited (i.e. you should bandpass filter the
>55-85
>>>>>> MHz band before sampling). Your sampling frequency is 100 MHz, so
>>the
>>>>>> signal of interest will "alias" down to a center frequency of 30
>MHz
>>>>>> after sampling. Your signal will then occupy the 15-45 MHz band in
>>the
>>>>>> sampled signal, which is all below the Nyquist rate. No problems
>>>>>> there.
>>>>>>
>>>>>> Jason
>>>>>
>>>>>To the original poster: the keyword is "bandpass sampling".
>>>>>
>>>>>
>>>>
>>>> For 30 MHz I signal, the maximum available bandwidth is 15 MHz. So
>is
>>Q
>>>> signal. If using I and Q, can I make it 30 MHz bandwidth? Thank
>you.
>>>
>>>Yes. You essentially utilize the bandwidth in -Fs/2 to +Fs/2 with an
>>>I/Q signal, whereas with a real signal the usable bandwidth is from 0
>>>to +Fs/2.
>>>--
>>>% Randy Yates % "And all that I can do
>>>%% Fuquay-Varina, NC % is say I'm sorry,
>>>%%% % that's the way it goes..."
>>>%%%% <yates@ieee.org> % Getting To The Point', *Balance of
>>Power*, ELO
>>>http://home.earthlink.net/~yatescr
>>>
>>
>>I appreciate your answer. Could you explain why Q signal shows
bandwidth
>>from -Fs/2 to 0? Is there any article or book talking about this?
>Thank
>>you very much.
>>
>>
>>
>>_____________________________________
>>Do you know a company who employs DSP engineers?
>>Is it already listed at http://dsprelated.com/employers.php ?
>>
>
>
>It is the interrelation between the I and Q channels that is able to
>distinguish between positive and negative frequencies. Search for
>"negative frequency" and you can find some lengthy discussions of it on
>this group.
>
>-Joe
>
>_____________________________________
>Do you know a company who employs DSP engineers?
>Is it already listed at http://dsprelated.com/employers.php ?
>
In some commercial RF digitizers, like e.g. Aeroflex PXI-3035 and National
Instruments PXI-5661, the real-valued IF signal is digitized with a
sampling frequency of 100 MHz, followed by IQ-separation and digital
downconversion (DDC) to the baseband. I remember the maximum signal
bandwidth, that can be acquired, in both cases is 20 MHz. Still wondering
if it is a theoretical limit of implementation constraint.
LBB
_____________________________________
Do you know a company who employs DSP engineers?
Is it already listed at http://dsprelated.com/employers.php ?
Reply by pompano●May 15, 20072007-05-15
>>"billykao" <billy.kao@gmail.com> writes:
>>
>>>>On May 14, 8:42 pm, cincy...@gmail.com wrote:
>>>>
>>>>> You should be fine, assuming that the signal you sampled is
>>>>> appropriately bandlimited (i.e. you should bandpass filter the
55-85
>>>>> MHz band before sampling). Your sampling frequency is 100 MHz, so
>the
>>>>> signal of interest will "alias" down to a center frequency of 30
MHz
>>>>> after sampling. Your signal will then occupy the 15-45 MHz band in
>the
>>>>> sampled signal, which is all below the Nyquist rate. No problems
>>>>> there.
>>>>>
>>>>> Jason
>>>>
>>>>To the original poster: the keyword is "bandpass sampling".
>>>>
>>>>
>>>
>>> For 30 MHz I signal, the maximum available bandwidth is 15 MHz. So
is
>Q
>>> signal. If using I and Q, can I make it 30 MHz bandwidth? Thank
you.
>>
>>Yes. You essentially utilize the bandwidth in -Fs/2 to +Fs/2 with an
>>I/Q signal, whereas with a real signal the usable bandwidth is from 0
>>to +Fs/2.
>>--
>>% Randy Yates % "And all that I can do
>>%% Fuquay-Varina, NC % is say I'm sorry,
>>%%% 919-577-9882 % that's the way it goes..."
>>%%%% <yates@ieee.org> % Getting To The Point', *Balance of
>Power*, ELO
>>http://home.earthlink.net/~yatescr
>>
>
>I appreciate your answer. Could you explain why Q signal shows bandwidth
>from -Fs/2 to 0? Is there any article or book talking about this?
Thank
>you very much.
>
>
>
>_____________________________________
>Do you know a company who employs DSP engineers?
>Is it already listed at http://dsprelated.com/employers.php ?
>
It is the interrelation between the I and Q channels that is able to
distinguish between positive and negative frequencies. Search for
"negative frequency" and you can find some lengthy discussions of it on
this group.
-Joe
_____________________________________
Do you know a company who employs DSP engineers?
Is it already listed at http://dsprelated.com/employers.php ?
Reply by billykao●May 15, 20072007-05-15
>"billykao" <billy.kao@gmail.com> writes:
>
>>>On May 14, 8:42 pm, cincy...@gmail.com wrote:
>>>
>>>> You should be fine, assuming that the signal you sampled is
>>>> appropriately bandlimited (i.e. you should bandpass filter the 55-85
>>>> MHz band before sampling). Your sampling frequency is 100 MHz, so
the
>>>> signal of interest will "alias" down to a center frequency of 30 MHz
>>>> after sampling. Your signal will then occupy the 15-45 MHz band in
the
>>>> sampled signal, which is all below the Nyquist rate. No problems
>>>> there.
>>>>
>>>> Jason
>>>
>>>To the original poster: the keyword is "bandpass sampling".
>>>
>>>
>>
>> For 30 MHz I signal, the maximum available bandwidth is 15 MHz. So is
Q
>> signal. If using I and Q, can I make it 30 MHz bandwidth? Thank you.
>
>Yes. You essentially utilize the bandwidth in -Fs/2 to +Fs/2 with an
>I/Q signal, whereas with a real signal the usable bandwidth is from 0
>to +Fs/2.
>--
>% Randy Yates % "And all that I can do
>%% Fuquay-Varina, NC % is say I'm sorry,
>%%% 919-577-9882 % that's the way it goes..."
>%%%% <yates@ieee.org> % Getting To The Point', *Balance of
I appreciate your answer. Could you explain why Q signal shows bandwidth
from -Fs/2 to 0? Is there any article or book talking about this? Thank
you very much.
_____________________________________
Do you know a company who employs DSP engineers?
Is it already listed at http://dsprelated.com/employers.php ?
Reply by Randy Yates●May 15, 20072007-05-15
"billykao" <billy.kao@gmail.com> writes:
>>On May 14, 8:42 pm, cincy...@gmail.com wrote:
>>
>>> You should be fine, assuming that the signal you sampled is
>>> appropriately bandlimited (i.e. you should bandpass filter the 55-85
>>> MHz band before sampling). Your sampling frequency is 100 MHz, so the
>>> signal of interest will "alias" down to a center frequency of 30 MHz
>>> after sampling. Your signal will then occupy the 15-45 MHz band in the
>>> sampled signal, which is all below the Nyquist rate. No problems
>>> there.
>>>
>>> Jason
>>
>>To the original poster: the keyword is "bandpass sampling".
>>
>>
>
> For 30 MHz I signal, the maximum available bandwidth is 15 MHz. So is Q
> signal. If using I and Q, can I make it 30 MHz bandwidth? Thank you.
Yes. You essentially utilize the bandwidth in -Fs/2 to +Fs/2 with an
I/Q signal, whereas with a real signal the usable bandwidth is from 0
to +Fs/2.
--
% Randy Yates % "And all that I can do
%% Fuquay-Varina, NC % is say I'm sorry,
%%% 919-577-9882 % that's the way it goes..."
%%%% <yates@ieee.org> % Getting To The Point', *Balance of Power*, ELO
http://home.earthlink.net/~yatescr
Reply by billykao●May 15, 20072007-05-15
>On May 14, 8:42 pm, cincy...@gmail.com wrote:
>
>> You should be fine, assuming that the signal you sampled is
>> appropriately bandlimited (i.e. you should bandpass filter the 55-85
>> MHz band before sampling). Your sampling frequency is 100 MHz, so the
>> signal of interest will "alias" down to a center frequency of 30 MHz
>> after sampling. Your signal will then occupy the 15-45 MHz band in the
>> sampled signal, which is all below the Nyquist rate. No problems
>> there.
>>
>> Jason
>
>To the original poster: the keyword is "bandpass sampling".
>
>
For 30 MHz I signal, the maximum available bandwidth is 15 MHz. So is Q
signal. If using I and Q, can I make it 30 MHz bandwidth? Thank you.
Billy
_____________________________________
Do you know a company who employs DSP engineers?
Is it already listed at http://dsprelated.com/employers.php ?
Reply by julius●May 14, 20072007-05-14
On May 14, 8:42 pm, cincy...@gmail.com wrote:
> You should be fine, assuming that the signal you sampled is
> appropriately bandlimited (i.e. you should bandpass filter the 55-85
> MHz band before sampling). Your sampling frequency is 100 MHz, so the
> signal of interest will "alias" down to a center frequency of 30 MHz
> after sampling. Your signal will then occupy the 15-45 MHz band in the
> sampled signal, which is all below the Nyquist rate. No problems
> there.
>
> Jason
To the original poster: the keyword is "bandpass sampling".