Kismet Wireless

Kismet Forums


Posted by:Dutch
Subject:Looking for example .cvs files
Date:22:43:08 12/12/2006

> Hi All,
> I'm building a system similar to Wigle but focused on a single (UK) city. The system will finally utilize public transport with built in wardriving kit. I assume this kit will be running Kismet. (This is the shortened version, if people are interested in this project please feel free to get in touch)
> The online system to combine all this data will need to parse a standard output into the db. Seems like Kismet .csv and netstumbler should cover most bases, do people agree?
> We're still in an early phase and so I've been using KisMac currently rather than Linux kit and Kismet. Seems like KisMac doesn't output a Kismet .csv though, am I being dumb or is that the case?
> Has anyone got access, or know where to find some example Kismet .csv wardrives to help us look at this stuff and confirm our future plans.
> Thanks for any help,
> Dan Efergan
> dan [at]

Setup linux and Kismet on a pc, or boot a Linux liveCD which has Kismet, such as Auditor or Back-Track, then run Kismet with a GPS attached and the csv dump enabled.

That way you got your own data to test with, and know better how Kismet functions.
KisMac and Kismet are two different programs, without a common codebase.

IMHO using Kismet's CSV is not the ideal way┬┤if you are going to do any kind of plotting and interpolation of locationdata based on the various logs you enter into the system, since you only got 3 location coordinates from that : GPSMinlat/lon, GPSMaxlat/lon, and GPSBestlat/lon.
Parse the .GPS files, which are in an XML format, and contains all locations a BSSID has been observed during a run. That way your interpolation routines for plotting the BSSID will have more data to work with.

Same thing goes for the NS files, do not use the summary exported text, but parse the binary .NS1 files, since they contain location data for all the locations NS observed the BSSID. The summary just contains the location where the highest signal strength was received, and that data is only 100% valid when used with a Hermes chipset based WiFi card. With cards using NDIS 5.1, Signal and Noise data are not to be trusted in NetStumbler.


Reply to this message