[GRASS-user] LOS, Fresnel Zones, Multi-Line wave propagtion and GHz communications

[gary]

Funny you folks brought up SPLAT!  A slightly more user friendly version 
of the program (free but not open source):
http://www.cplus.org/rmw/english1.html

You can read about my experience with the program here under wine:
http://forum.winehq.org/viewtopic.php?t=580

SPLAT has the drawback that it can only use the 3 arc second DEMS. Radio 
Mobile can use (in theory) DEMs down to 1/3 are second. The drawback to 
both programs is the fixed array size. For splat, the size is fixed at 
3600x3600. Radio Mobile can vary the size, but no larger than 2000x2000. 
These programs are old, dating back to the days when computers were very 
limited in RAM and speed.

One nice thing about both programs is they read back the height at the 
transmitter location. Since the DEM doesn't show the peak altitude but I 
suppose an average over the region, the altitude used by the program is 
often less than real life. I find a get more accurate results by finding 
the altitude used by the software, then adding enough to it to reach the 
known altitude of the peak.

When I get the chance, I'll do some comparisons between the programs.

Incidentally, one real annoyance with r.los is that it can't accept 
lat/lon.

Here is an example of using GRASS for a similar study:
> http://www.gis.unbc.ca/courses/geog499/projects/2004/vera_lindsay/GrassDocumentation.html




Dylan Beaudette wrote:
> On Thursday 17 April 2008, Tom Russo wrote:
>> On Thu, Apr 17, 2008 at 05:36:18PM +0200, we recorded a bogon-computron 
> collision of the <peter.hopfgartner at r3-gis.com> flavor, containing:
>>> Dylan Beaudette wrote:
>>>> There is also an open source application called 'splat' :
>>>> http://www.qsl.net/kd2bd/splat.html
>>>>
>>>> I do not know if splat can account for near-surface conditions (i.e.
>>>> Fresnel Zone) or higher elevation atmospheric parameters (ionosphere).
>>>>
>>>> It would be nice to have a couple of more robust LOS-like modules in
>>>> GRASS for this type of work.
>>> from the description on the web page it seems that Fresnel Zones are in
>>> SPLAT, at least in recent versions.
>>>
>>> I will have a look at it, tomorrow.
>> SPLAT is pretty cool, and if all you want is to compute some RF coverage
>> maps it works well as a standalone application.  It would make an
>> interesting project for GRASS integration.
>>
>> As distributed, SPLAT only works when you give it digital elevation models
>> in a very specific file format, and it produces PPM output (not
>> georeferenced) to display its results.
>>
>> I've often thought that it would be cool to adapt it to a GRASS module that
>> can work with any GRASS elevation data and produce a GRASS raster of RF
>> loss and radio coverage instead of an ungeoreferenced image file.  SPLAT
>> itself is a fairly simple program once you get digging in it (the math is
>> not simple, but the code structure is), but the integration project has
>> always been too daunting for me.
> 
> I wonder if the core code used in SPLAT would scale better than the current 
> r.los algorithm. If that is the case, it might be a nice candidate for an 
> r.los replacement for GRASS 7...
> 
> 
>