Re: LF: LF EbNaut test from JN80 on 137370

[Andy Talbot <andy.g4jnt@gmail.com>]

And, once again I managed to find a keyboard shortcut by mistake that sent the message too soon............

I always think only in terms of sin and cos.

Real = = Inphase = COS

Imag =  Quadrature = SIN

So , to shift up in frequency, Fout = A + B

From basic trig identities:

I term   =   COS(A+B) = COS(A).COS(B) - SIN(A).SIN(B)

Q term  =   SIN(A+B) = SIN(A).COS(B) + COS(A).SIN(B)

with  A or B replaced by the input samples

Andy  G4JNT

On 13 December 2015 at 12:59, Andy Talbot <andy.g4jnt@gmail.com> wrote:

The terms "leading" and "lagging" are the most confusing terms I've ever come across in AC.  A waveform on a scope, to the right of another one looks as if its leading it.   But it is later in time - so its lagging.

See my point ?

On 13 December 2015 at 12:33, Wolfgang Büscher <dl4yhf@freenet.de> wrote:

Just to confirm (answer a bit late)..

In a file recorded by SpectraVue (the software which came along with the SDR-IQ),  with the VFO tuned to 77490 Hz, recodring DCF77 on 77500 Hz, produces a 'positive' 10 Hz baseband frequency in the I+Q wave file recorded by that software.
And, as expectable for "I" ~ cos(omega*t) and "Q" ~ sin(omega*t), the phase of the signal's "I" component *leads* the "Q" component by 90 degrees. So, "all ok" with the files recorded by whatever software.

Besides that, it's advisable *not* to tune the SDR exactly to the frequency of the EbNaut signal. Detune by a few Hz to avoid having the "0 Hz" component (after the downconversion, either digital or analog) inside the decoder's passband.

Now back to local EbNaut testing...

Have a nice sunday an a merry xmas season everyone,
   Wolf .

Am 12.12.2015 um 13:12 schrieb Andy Talbot:

Hit send too soon on the last one

On 12 December 2015 at 12:06, Andy Talbot <andy.g4jnt@gmail.com> wrote:

I've just added a fine tune to the PC prog to shift teh input frequeny by any amount specified in a text box

           

 

  {  ipI& and ipQ& are the input samples from the receiver  }

 

            LOtheta = Val(Tuning.Text) / Fsample  

            LocalOsc = LocalOsc + LOtheta             'Both are normalised frequencies (Fsample = 1)

            LocalOsc = LocalOsc - Int(LocalOsc)      'Fractional part only

            theta = 2 * pi * LocalOsc

  

            IPData(DataPointer&, 0) = ipI& * Cos(theta) - ipQ& * Sin(theta)       '    I

            IPData(DataPointer&, 1) = ipQ& * Cos(theta) + ipI& * Sin(theta)      '   Q

The effect of that is for any positive freqeuncy in Hz intereed into Tuning.text , the spectrum plot shift positive, and conversely the display shifts wown for any -ve value entered

Andy  G4JNT

On 10 December 2015 at 20:34, Paul Nicholson <vlf0403@abelian.org> wrote:

In vlfrx-tools I use

 I = s(t) * cos( 2pi * Fo * t);

 Q = s(t) * sin( 2pi * Fo * t);

where s(t) is the incoming signal and Fo is the
local oscillator frequency.   Is that right, or
should there be a minus sign in there somewhere?

If s(t) is VO1NA at 137370.00000 and I set Fo to
137369.9, then when I put I and Q into EbNaut, it
needs an offset of +0.1 Hz to decode as you would
expect with the LO being 0.1Hz on the low side.

For more general mixing and frequency shift,
to select the upper sideband of the mixer output
I use

   I - j * Q

is that the right way round?   When I take I/Q data
from an RTL2832 I find I have to invert Q to get the
correct sideband, ie I + j * Q.   That's the only
independent source of I/Q signal that I have available
to try.

--
Paul Nicholson
--