[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [linrad] Question about time delay compensation

how I am intending to proceed is the following:
Given that what must be compensated for is a fixed delay caused by the hardware
of the sound card, I will place a slider in a "Calibration" panel, where the user can
finely adjust the time delay, and in real time see the effect of his adjustment on the
spectrogram. This will need a steady carrier as input, and the adjustment will be
done for the maximum rejection of the unwanted sideband. The spectrogram can
be adjusted for more than 130 dB of dynamic range, so the visual adjustment
shouldn't be too difficult. This will require from me to compute the h(k) coefficients
of the allpass filter also in almost real time, but that should be possible.

The paper you couldn't load is on:


probably you will see this link underlined only up to the apostrophe following PIMRC.
If this is the case, just copy manually with the mouse all of the row, up to the final .pdf
and paste it into the address line of the browser. In this way it should work.

73 Alberto I2PHD

Robert McGwier wrote:


I possibly was not careful enough when I wrote. I suggested
that you find the amount of correction in the frequency domain
and that it be applied in an interpolated or upsampled version
of the signal where you only apply the upsampling filter
(again in the time domain) at the correct phase. The upsampling
filter can and should be dynamic since it can correct for amplitude
and frequency dependent distortions determined in step 1. My approach
would be to do the interpolation in the time domain as specified
in the papers you list. I cannot get the last paper to load and
since you say it is the most interesting, I would like to get
a copy of it.

Good luck,