Tom Azlin N4ZPT
n4zpt at cox.net
Sat Sep 4 16:48:55 CDT 2010
On the RFSpace site is a discussion of using one of their SDRs to see
pulsars but do not know the antenna size or aux equipment needed. I
think was looking at known pulsar so could sync the SDR capture frames
using the known pulsar rep rate.
I am working on getting my SDR-14 software and drivers running since I
just gave my SDR-IQ to the VWS EME project. The SDR-14 can down sample
easily at the first and 2nd IF. And we can save away either narrow band
I/Q or the FFT words. Also looking for my QS1R which can do the same
On 9/4/2010 9:27 AM, Robert E. Seastrom wrote:
> hal feinstein<hfeinstein at cox.net> writes:
>> I know there's a desire to do amazing things but everyone needs to
>> remember the 40ft'er is very modest equipment compared to what the
>> pros are using. Think of it as a 75watt cw station with a dipole
>> antenna. Its like a training ship to learn the fundamentals.
>> There's a lot of basics to learn before you head out into deeper
> understood, and thanks for putting it in those terms.
>> I think the high jump for us would be to use the 40ft'er to observe
>> one of the stronger pulsars. Pulsars are weak, real weak -- you've
>> got to integrate. Pulsar signals suffer frequency dispersion due to
>> their travel through space filled with very sparse interstellar
>> particles. The signal arrives dispersed so its tricky to receive.
>> Dispersion compensating filter can be constructed and placed in the
>> receiver so you can integrate. There's dsp ways to do this too.
> the last point is what i was thinking of when i was talking about
> having reasonably studly amounts of storage.
> if you wanted raw data to play with in dsp-land, how much oversampling
> do you need to get the resolution you want (in other words, do you
> need to know what the waveform looks like, or just an indication that
> there's "something there" or what? shannon and nyquist need not apply
> here; we're not trying to decode communications...) also, how many
> bits of resolution?
> what's the passband again? if we made a second if stage (is this
> practical to do without killing what we're looking for due to noise?)
> to baseband an 80 mhz passband, and took 8 bit samples (likely too
> little resolution) at 160 megasamples/second, that exceeds the
> throughput of any hard drive you can buy today, and when you start
> talking about striped arrays with either parity or mirrors, the
> problem gets even more annoying.
> as frank said, so many questions...
More information about the Tacos