Kismet Wireless

Kismet Forums


Posted by:rayv5n
Subject:KISMET/GPSD/GooleEarth incorrect/inconsistent waypoint mapping help!
Date:18:15:26 10/01/2007

Hi Dutch and thanks!

GPSD,O=GSA 1168452209.000 0.005 41.419442 -121.971248 9.84 19.20 14.70 0.0000 0.051 0.000 ? ? ? 3

That last 3 shows that I'm in 3D mode and I forgot to mention that I have collected my data both from inside the building and the parking lot. Same results either way. My location is in a valley and we happen to be on a flat plane so I don't think its refraction. I've done some googling and read through the gpsd readme but I'm not finding 'datum|data' set and am still unclear. Although I'm not a php coder I will turn my attention on kismet earth. I've read that some articles about data generated by kismet/netstumbler can be counter to what google earth and gpsvisualizer need to see, ie., N and W instead of S and E.


> > Which albeit some errors created the Kismet*.kml file. I then ran Google Earth and open the kml, oh joy! a was soon zooming down to my location gliding towards those AP marked waypoint with ease! There was just one problem, the waypoint representing my APs where 30 meters from where they should be. Instead of hovering over the building they were clear across the parking lot.
> You do realize that the GPS data shows where YOU are when you detect the networks, not where the networks are ?
> If you were in the building, and not on the parking lot, it can be due to several reasons :
> Wrong datum set on the GPS. Normally it should be set to WGS84.
> A fix with only 3 sats visible, aka a 2D fix. Less precise and can fluctuate a lot compared to a fix with 4 or more sats visible (a 3D fix).
> Multipath reflections of the GPS signal, can happen in urban canyons or indoors.
> Errors in your setup due to missing dependencies for the PHP parsing of the php script.
> You need to check all the above, and fix anything wrong, in order for the plotting of the logs to be valid.
> Just for fun, try to use kismet's included gpsmap utility to plot the same log files. If they show the same distance error, then it is most certainly items 1,2 or 3 above.
> Dutch

Reply to this message