[hpsdr] Direction Finder using Dual-Mercury Diversity Operation

[Joe Martin K5SO]

Hi Abhi,

Multiple ADCs are necessary, as it is the phase difference(s) between  
the separate input rf streams that is critical to DF.  One rf input  
stream won't do it.  If one were to design around Hermes it would be  
necessary to add additional ADCs somewhere.  The single FPGA and  
single ethernet interface would work fine for all receivers, in fact  
Hermes currently supports multiple receivers but, if configured as  
multiple Rx, all the receivers in Hermes use the output from the  
single ADC and therefore there is no spatially derived phase  
distinction available in the IQ streams from those multiple  
receivers,  as there is only a single, common, rf source for them all;  
i.e., the one ADC.

You mentioned an add-on board as a possible way to interface with the  
existing Hermes.  Such an add-on board would need to include  
additional antenna inputs and ADCs, at a minimum.  The outputs from  
the additional ADCs would have to be routed somehow to the FPGA on  
Hermes.  This could in principle be done, as I noted before, either  
via Atlas bus lines or with ribbon cabling if there is some kind of  
header available on Hermes that could be used (perhaps there isn't).   
It seems to me that neither of those choices is optimum as there is  
probably good reason for the ADCs on Hermes and Mercury being  
physically close to the FPGA.  But it might work, I don't know.

I think a self-contained (new) board design would be the cleanest  
approach if there is demand for an efficient single-board DF rig.   
However, as I also mentioned, the current selection of HPSDR boards  
provides the base setup for a direction finding configuration already;  
namely, using multiple coherent Mercury receivers.  So we already have  
in hand the essential hardware necessary to demonstrate the DF  
feasibility.  If/when the steering program is fully operational, it  
may generate sufficient interest in DF activity to warrant the group's  
serious attention toward developing a single-board DF project.   That  
is, unless there is already sufficient interest in the group!

In any case, we can certainly mull over the possibilities ourselves  
and even put some single-board designs together if we need something  
new to occupy our time!

73,  Joe K5SO


On Sep 22, 2011, at 12:32 PM, Abhi Arunoday wrote:

> ***** High Performance Software Defined Radio Discussion List *****
>
> Hi Joe,
>
> I would use hermes only if the multiple receivers would run inside  
> the one FPGA, use the same clock, ADC and the ethernet interface  
> which are all available on Hermes, i'm not sure if this will work  
> for diversity reception or DF though?
>
> I believe multiple ADCs would be needed, therefore, 2x or 3x mercury  
> cards can be used as is the case now or all the ADCs could use the  
> same FPGA and ethernet interface, this would need to go on a new PCB,
>
> I'm OK with all options,  it seems if we need multiple ADCs then   
> Mercury boards can be used but at a higher cost than using one FPGA  
> and multiple ADCs.
>
> Abhi
>
>
> _______________________________________________
> HPSDR Discussion List
> To post msg: hpsdr at openhpsdr.org
> Subscription help: http://lists.openhpsdr.org/listinfo.cgi/hpsdr-openhpsdr.org
> HPSDR web page: http://openhpsdr.org
> Archives: http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/


 1316719153.0