[postgis-users] Radcom GIS

[Simon Greener]

Greg,

Did you get my email expressing interest in helping you?

regards
Simon
On Fri, 25 May 2012 18:59:21 +1000, Greg Evans <ggpevans at yahoo.com.au> wrote:

>> MAny thanks to those who can help. I have attached a word based doc to ease the reading.
>
>   
> Hello, my name is Greg and in years past I use to work as a Radio Communication Engineer.
>   
> My background is in Electrical and Electronic engineering, not computer science and I am a little stuck. I was wondering if >someone can help answer my Comsci questions. The problem is described below.
>   
> As an engineer doing Radio Communications the task is to determine coverage and other Radio Communication attributes >such as intermods and other things such as harmonic interference.
>   
> In the process of doing the job, you start with a transmitter located at an x-y point.
> Then you give that x-y point a transmitting height and power level.
>   
> This Tx power level and height, enables distinct points in a closed geographical rectangle or circle, to have its field/strength (F/>S) determined at each point.  
>   
> To determine field strength (F/S), you use different power levels, along with different Tx heights and different computer >propagation models to find the F/S. These F/S results are graphically displayed (over maps)
>   
> Once the field strength is determined, the second step is called Carrier-To-interference (C/I). If you have two sites close to >each other, if they are on the same frequency, the two signals can interfere, degrading either or both of the signals. But >sometimes they don’t. eg. Place your new TX site behind a mountain. The mountain shields the two systems from interfering >with each other.
>   
> Anyway, for C/I calculations, you can pick an x-y point (in a region) and subtract the field strength (from each TX source). Eg. >At an x-y point between the two Tx’s, you subtract F/S1 – F/S2. or Vice Versa.
>   
> Again computer propagation models are needed to determine F/S, which then allows for the determination of C/I.
>   
> If the C/I is bad, you have to re-site the new Tx site location and start again.
>   
> It has been a while since doing radio communications but I remember that the maps had DEM’s associated with it. Ie. >Different maps for different DEM’s.
>   
> I am not the expert when it comes to GIS systems and geographic systems.
>   
> From memory you site a TX point, then the program took the DEM (value) across each point (in a region specified by a map) >and then for each point, use it as a distance parameter into Radio propagation models, which determines the field strength (F/>S) at each point.
>   
> I cant remember whether the smaller DEM gives different field strength results compared to a larger DEM being used.
>   
> From memory I think you get different field strengths for different DEM Maps. I am not sure if this is correct. What I do know >is that a smaller DEM (for a given area) takes longer to compute than using a larger DEM. The smaller DEM might give the >same F/S results but takes ages to compute. I am not sure if this is correct.
>   
> Ie. The smaller the DEM, the smaller the “distance parameter” between two points, which takes longer to compute all the F/S >at each point.
>   
> Further, for a smaller DEM, this means over a given area, there is more Field strength points to calculate. But either model >yields good results, only if the propagation model is correct. The size of the DEM determines time to compute.
>   
> The larger the DEM model, the less points there are in a given area, meaning it takes less time to compute compared to smaller >DEM’s, but the results at a large DEM, may give false information. The resolution of parameters is poor.
>   
> This is the crux of the problem.
>   
> I have been reading about the GRASS GIS systems and PostGIS. I believe I can create a better computerised model than that >being used at regulators based on GIS concepts.
>   
> Unfortunately I cant test these systems (on a computer) at the moment. But to save time I am asking some questions. If you >can help, I would really appreciate it.
>   
> I don’t understand Vectors and Rasters w.r.t GIS systems at the moment, I will learn it later.
>   
> What I intend to do is create a new Radio Communications Tool that is web based.
>   
> Ie. get all the Maps and layers and display them in a web-page served by a Web based server. Then write models that will >take a specified point eg. A transmitter, along with a DEM to figure out the Field Strength and Carrier-to-Interference all done >through a web-browser.
>   
> I call this application a Web-RADCOM-GIS tool. (WRG - Web based radio communications)
>   
> This would be good for groups of Radio Communications engineers.
>   
> Instead of paying $100,000 for software crippled by dongles and other means, the idea is that the radio communications >engineers (or anyone else interested in radio communications, eg HAM radio people). can see F/S results, displayed in a >browser window, without any money laid down.
>   
> The limit of users is limitless.
>   
> Instead, the idea is that they can log into a mapserver using a browser, see different layers and do different vector (and raster) >operations to determine radio calculations (eg. F/S and C/I).
>   
> This new webserver GIS system, could even be extended to planning mobile Base Station sites. The better the modelling tool, >the better the mobile coverage and interference modelling, meaning less RF spectrum is wasted. The better the Mobile Phone >models get done, the less of mobile base stations need to be deployed, meaning less RF energy bumping around near >buildings. In other words, making more out of less.
>   
> The less base stations, the less rent paid for building housing the base station antennas. For Example.
>   
> Anyway having a web-based GIS system allows for heaps of group-work. Instead of buying the DEM maps and installing >them on each PC (in the group), all they have to do is reference different DEM maps through a MAP server, inside a browser.
>   
> I have seen this in action. It takes ages to change the Maps on heaps of PC. And similarly if the computer propagation models >change, it is effortless to change.
>   
> I intent to make this paid for or free for use, this way anyone on the planet can model and investigate different Radio >Communication attributes.
>   
> Moreover, more everyday people get a chance to see Radio communications. It gives uses who need a frequency allocation (a >channel to operate on) a chance to design them using the tool. Then all the FCC has to do is check their results.
>   
> Previously in Australia the regulators do all the calculations and frequency allocation.
>   
> Anyway going back to the RADCOM tool. I saw a screen shot showing height profiles (in a GRASS window). This is exactly >what is needed to determine Field Strength.
>   
> I am not sure if I am right, I figure for web page based system, one can use PHP to generate SQL statements, which puts data >into either a postgreSQL system or postGIS systems. Then similarly, use PHP and SQL to extract data from the PostGRE or >PostGIS and display it on a web-page in a browser.
>   
> Even better save the results (as a dataset) and print the F/S or C/I results. Going a step further overlay the results over a >census collection district, for example.
>   
> The crux of this, leaves me wondering. To make this Radcom tool faster, you should in theory change the way the system >computes. Specifically, I am looking at changing either or both, {PostGIS) or (postgre sql) to be re-done in assembly >language.
>   
> Assembly is orders of magnitude faster. I am unsure whether the performance increase comes from coding PostGIS or >PostgreSQL in assembly.
>   
> Your advice on whether to code PostGIS or PostGRE in assembly this is keenly sought.
>   
> The second thing is that in theory for a web based app, I would have to redo all the menu (and window forms) in PHP? Could >you please advise of any books that are good on these topics.
>   
> Thankyou for your time and help on these matters.
>   
>   
>




-- 
Holder of "2011 Oracle Spatial Excellence Award for Education and Research."
SpatialDB Advice and Design, Solutions Architecture and Programming,
Oracle Database 10g Administrator Certified Associate; Oracle Database 10g SQL Certified Professional
Oracle Spatial, SQL Server, PostGIS, MySQL, ArcSDE, Manifold GIS, FME, Radius Topology and Studio Specialist.
39 Cliff View Drive, Allens Rivulet, 7150, Tasmania, Australia.
Website: www.spatialdbadvisor.com
Email: simon at spatialdbadvisor.com
Voice: +61 362 396397
Mobile: +61 418 396391
Skype: sggreener
Longitude: 147.20515 (147° 12' 18" E)
Latitude: -43.01530 (43° 00' 55" S)
GeoHash: r22em9r98wg
NAC:W80CK 7SWP3
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/postgis-users/attachments/20120622/9180e582/attachment.html>