[hpsdr] Multiple Independent Receivers & Mercury

[Georg Prinz]

Hello Frank,

you are right, that there are different definitions of multiple
independent receivers. 
Unfortunately I don't know if there is an accepted or IEEE defined
nomenklature existing. 
Mostly it is understood to talk about a multi-channel-receiver. This is
the version used in base stations. The whole system-bandwidth is
digitized and the channel selection is done via multi channel receiver
based on FFTs.

The other kind of "multi-receivers" is a system used in smart antennas.
You have N inputs with A/D-converters with digital down-converters which
produce baseband samples simultaneously. They are passed simultaneously
to a N-point complex FFT to perform a spatial FFT. That means the
outputs are separated in direction and not in frequency.

We should make a kind of brainstorming to define, what we need as hams.

Georg
dl2kp




On Do, 2009-03-19 at 08:57 -0400, Frank Brickle wrote:
> ***** High Performance Software Defined Radio Discussion List *****
> 
> 2009/3/19 Steven Doyle <steve_doyle at o2.co.uk>
>  
>         As I understand it a single Mecury is quite capable supporting
>         multiple independent receivers. I believe Phil said it could
>         do 6, and only limited by our USB bandwith. The only reason
>         that we currently have a single max 192KHz was so that we
>         could use existing software like PowerSDR.
> 
> There are too many definitions of "multiple independent receivers"
> floating around here.
> 
> DttSP sdr-core is essentially the DSP code running inside PowerSDR. A
> single sdr-core instance is capable of running a fairly large number
> of independent receivers inside itself. I haven't checked "fairly
> large" recently, but it's probably around 16 or so without strain. The
> limitations on this method are (1) all of the receivers have to be
> tuned within the same basic passband -- in the case you describe,
> 192kHz; and (2) all of the RX outputs get routed to the same audio
> output, so that they can only be distinguished by their pan position
> in the stereo output.
> 
> The number of receivers enabled in PowerSDR is a consequence of the
> way PowerSDR is designed and built.
> 
> It is also possible to run *multiple sdr-core instances* (and thus
> multiple truly independent receivers) on a suitable OS (Linux, Mac OS
> X, BSD).  Here, each sdr-core needs its own IF input stream (although
> they can be fanned out from a single source). The number of such
> independent processes is also not small. I've run 6 simultaneous
> sdr-cores on a modest machine, although at 48kHz per RX. On a
> reasonably  capable machine, you could afford to be somewhat more
> ambitious. Here, each of the sdr-cores also routes its output
> independently. Note that each of these independent processes can have
> its own block of multiple internal receivers, as described in the
> paragraph above.
> 
> Neither of these scenarios has the slightest dependence on Mercury per
> se.
> 
> In both these scenarios, the limitation observed in PowerSDR is not on
> Mercury or the DSP: it's on the specific application (PowerSDR).
> 
> java-sdr and sdr-shell do not have the same limitations.
> 
> 73
> Frank
> AB2KT
> 
> 
> 
> -- 
> Take her and cut her out in little stars,/And she will make the face
> of heaven so fine/That all the world will be in love with night... --
> Wm. Shakespeare, ad lib
> _______________________________________________
> HPSDR Discussion List
> To post msg: hpsdr at hpsdr.org
> Subscription help: http://lists.hpsdr.org/listinfo.cgi/hpsdr-hpsdr.org
> HPSDR web page: http://hpsdr.org
> Archives: http://lists.hpsdr.org/pipermail/hpsdr-hpsdr.org/


 1237491068.0