On 20 August 2015 at 15:42, Jordan Kagan <[email protected]> wrote: > > So in short I am wondering why one wouldn't just choose the center frequency > to be the FM channel frequency or a frequency in the range of interest.
A great question and one that causes confusion for many people. It's discussed in the HackRF FAQ, but from a different angle (i.e. "how do I solve problem X?" "offset the centre frequency"). If you start gqrx or any other tool that shows an FFT of the received data, you will notice a large spike in the centre of the display. This is the 0Hz component of the received signal, also known as the DC offset or DC bias. It's a side effect of the quadrature sampling technique and is just something that we have to live with. This is discussed here: https://github.com/mossmann/hackrf/wiki/FAQ#what-is-this-big-spike-in-the-center-of-my-received-spectrum The following FAQ question is about how to deal with it: https://github.com/mossmann/hackrf/wiki/FAQ#how-do-i-deal-with-the-dc-offset The tl;dr of that question is: offset the centre frequency of the radio and then shift it back in software. This is what those flowgraphs are doing. I hope that answers the question, although if you would like more detail, I'm sure someone on this list can answer it better than I can. Thanks, Dominic _______________________________________________ HackRF-dev mailing list [email protected] https://pairlist9.pair.net/mailman/listinfo/hackrf-dev
