Prev: Xilinx Pipelined/Streaming FFT Architecure?
Next: Shannon's upper bound for AWGN channel, but what about Multipath?
From: csilva1975 on 9 Apr 2010 07:22
Hello all.

Sorry for my absence, but my time is becoming scarce these weeks.
Good news: the DSP chip has arrived, will be programming it soon to try out
my algorithm.

Vladimir:

>> Yup you guessed it. I will upconvert it with a ne602 controlled by the
pwm
>> output of the DSP/MCU.
>
>OMG. I can only imagine what the output spectrum is going to be.

I forgot to mention that the pwm output will be 50% duty cycle, so I
don't expect much interference. I will have signals spread at harmonics
of my pwm frequency, but they are spaced and can be easily filtered.

>What is the final goal that you try to accomplish?

Well … I remember a while ago when I first saw a simple and effective use
of practical DSP. I hooked up a ne602 to the sound card of the computer and
using software (that I can't remember the name) I could listen to AM
broadcasting, SSB, change the filters etc.

Then I started to look for better ways of doing this. The objective *was*:

Sample the output of a 455 Khz ceramic filter (there will be a mixer before
the filter). Process the signal and output the demodulated signal.

After I started simulating the process, it has occurred to me that I could
easily use the same platform with little modification to transmit.

AM and SSB are relatively easy implementation for my knowledge, the FM part
is a little bit trickier. And that was the reasons that lead me to post
here.

My aim is the 80m band.

Back to work.
Cheers.
César

From: Tauno Voipio on 9 Apr 2010 10:34
On 9.4.10 2:22 , csilva1975 wrote:
> Hello all.
>
> Sorry for my absence, but my time is becoming scarce these weeks.
> Good news: the DSP chip has arrived, will be programming it soon to try out
> my algorithm.
>
> Vladimir:
>
>>> Yup you guessed it. I will upconvert it with a ne602 controlled by the
> pwm
>>> output of the DSP/MCU.
>>
>> OMG. I can only imagine what the output spectrum is going to be.
>
> I forgot to mention that the pwm output will be 50% duty cycle, so I
> don't expect much interference. I will have signals spread at harmonics
> of my pwm frequency, but they are spaced and can be easily filtered.
>
>> What is the final goal that you try to accomplish?
>
> Well … I remember a while ago when I first saw a simple and effective use
> of practical DSP. I hooked up a ne602 to the sound card of the computer and
> using software (that I can't remember the name) I could listen to AM
> broadcasting, SSB, change the filters etc.
>
> Then I started to look for better ways of doing this. The objective *was*:
>
> Sample the output of a 455 Khz ceramic filter (there will be a mixer before
> the filter). Process the signal and output the demodulated signal.
>
> After I started simulating the process, it has occurred to me that I could
> easily use the same platform with little modification to transmit.
>
> AM and SSB are relatively easy implementation for my knowledge, the FM part
> is a little bit trickier. And that was the reasons that lead me to post
> here.
>
> My aim is the 80m band.
>
> Back to work.
> Cheers.
> César
>

Make a phasing exciter. It is pretty easy to use modern digital
components for creating a pair of 90 degree offset carrier signals
for the 3.5 MHz band, using a pair of D flip-flops connected as
a ring counter. Also, modern CMOS bus switches make good balanced
modulators.

For FM, create it as PM, by using one balanced modulator for the
only sideband pair, and leak the carrier via the other modulator,
by a suitable DC offset. You can also create a pair of angle-
modulated I and Q signals for the balanced modulators, for a more
exact phase modulation. Please remember that for AC modualting
signal, FM and PM are essentially the same, except that phase is
the integral of frequency.

--

Tauno Voipio, OH2UG
tauno voipio (at) iki fi


From: csilva1975 on 9 Apr 2010 19:38
>
>Make a phasing exciter. It is pretty easy to use modern digital
>components for creating a pair of 90 degree offset carrier signals
>for the 3.5 MHz band, using a pair of D flip-flops connected as
>a ring counter. Also, modern CMOS bus switches make good balanced
>modulators.
>
>For FM, create it as PM, by using one balanced modulator for the
>only sideband pair, and leak the carrier via the other modulator,
>by a suitable DC offset. You can also create a pair of angle-
>modulated I and Q signals for the balanced modulators, for a more
>exact phase modulation. Please remember that for AC modualting
>signal, FM and PM are essentially the same, except that phase is
>the integral of frequency.
>
>--
>
>Tauno Voipio, OH2UG
>tauno voipio (at) iki fi
>
Hello again.

I was thinking of using something like this:

s(t)=sin((2 * pi * 6540 + (m(t) * 2500)) / Fs)

where m(t) is the modulating signal ranging from -1 to 0.99996948 (Q1.15
format). This whould give a 6540 hz signal for m=0, 4040 Hz for m=-1 and
9040 Hz for m=1.
It's pretty easy to implement that in DSP as long as we have a good sine
generator/oscillator.

It's impossible for me to implement a IQ output (and then mix it with 90
degree carriers) because I don't have enough I/O pins since I will not use
a DAC chip but a R2R ladder insted.

Cesar

From: Tauno Voipio on 10 Apr 2010 11:59
csilva1975 wrote:
>> Make a phasing exciter. It is pretty easy to use modern digital
>> components for creating a pair of 90 degree offset carrier signals
>> for the 3.5 MHz band, using a pair of D flip-flops connected as
>> a ring counter. Also, modern CMOS bus switches make good balanced
>> modulators.
>>
>> For FM, create it as PM, by using one balanced modulator for the
>> only sideband pair, and leak the carrier via the other modulator,
>> by a suitable DC offset. You can also create a pair of angle-
>> modulated I and Q signals for the balanced modulators, for a more
>> exact phase modulation. Please remember that for AC modualting
>> signal, FM and PM are essentially the same, except that phase is
>> the integral of frequency.
>>
>> --
>>
>> Tauno Voipio, OH2UG
>> tauno voipio (at) iki fi
>>
> Hello again.
>
> I was thinking of using something like this:
>
> s(t)=sin((2 * pi * 6540 + (m(t) * 2500)) / Fs)
>
> where m(t) is the modulating signal ranging from -1 to 0.99996948 (Q1.15
> format). This whould give a 6540 hz signal for m=0, 4040 Hz for m=-1 and
> 9040 Hz for m=1.
> It's pretty easy to implement that in DSP as long as we have a good sine
> generator/oscillator.
>
> It's impossible for me to implement a IQ output (and then mix it with 90
> degree carriers) because I don't have enough I/O pins since I will not use
> a DAC chip but a R2R ladder insted.


Please get a decent DAC, if you are intending to come
on the band (though I think that you're far enough away
on 80 m band ...).

Forget the 6 kHz subcarrier - it will be very difficult to
get rid of the spurious responses of the subcarrier and
mirrored sidebands in the up-conversion phases.

--

-Tauno
First  |  Prev  | 
Pages: 1 2 3 4
Prev: Xilinx Pipelined/Streaming FFT Architecure?
Next: Shannon's upper bound for AWGN channel, but what about Multipath?