[kismac] Re: KisMAC 0.05e

  • From: Brad Knowles <brad.knowles@xxxxxxxxx>
  • To: kismac@xxxxxxxxxxxxx
  • Date: Wed, 10 Sep 2003 16:30:45 +0200

At 5:20 PM +0200 2003/09/05, Michael Rossberg wrote:
>                    internally kismac uses now a better exception
>  handling, so that a lot of crashes should can be avoided.
>  have fun and give me some feedback please.

        I don't have a GPS device yet, so none of the mapping stuff helps 
me -- yet.

        However, I have been doing some testing with my two new Airport 
Extreme base stations, plus my two old Airport "Snow" base stations. 
I'm going to try to attach a couple of PNGs that demonstrate some 
problems I'm having here.

        If you look at the network table, you can see that I have four 
base stations, each serving the same network (four different BSSIDs). 
You can also see current and max signal strength.  The two base 
stations with BSSIDs starting with 00:03:93:EF are the new "extreme" 
model, and the other two (with BSSIDs starting with 00:30:65:24) are 
the old "snow" model.

        However, when you look at the traffic chart, you don't see any 
BSSIDs, so you can't easily determine which colors belong to which 
station.  By looking at the table, you should be able to figure out 
which is which, except that during the entire five minute period, 
stations 00:03:93:EF:28:1B and 00:30:65:24:6D:AD were both showing 
max signal strength of 33, with the latter switching to a max signal 
strength of 72 moments before I clicked on the "Stop" button.  Of 
course, this new peak also doesn't show up on the chart.

        By looking at the chart, you can tell that one station was 
obviously stronger than the other (red versus orange), but because 
you don't have the BSSIDs you don't know which is which.  Moreover, 
while the chart was being recorded, the orange and red colors were 
flipping back and forth regularly, as one station would record a new 
current signal strength that was slightly higher than the current 
signal strength for the other.

        I recommend putting the BSSIDs on the chart as well as in the 
table.  I also recommend leaving the colors in the chart stable, 
based on readability of neighboring colors.  Don't change color 
assignments of a base station based on signal strength.

        I suggest that you choose color combinations that are equally 
spaced around the color wheel, based on the number of stations being 
measured.  This is usually easiest with an HSV color model, and 
dividing 360 by the number of stations and then using that as an 
index into the Hue axis.  Make sure to alternate color assignments 
for adjacent base stations, to maximize contrast and readability of 
the chart.

        In addition, max signal strength is clearly not really useful. 
We need either an average signal strength (in which case you'd also 
need to calculate a standard deviation), or a median signal strength 
(which is less subject to being thrown off by unusually high or low 
readings, and doesn't need an additional number like standard 
deviation in order to be useful).

        Also, I would recommend that you do not clear the traffic chart 
when minimizing the window after data collection has stopped.  It is 
useful to be able to go back and see what traffic used to look like 
for the previous measurements, without necessarily being forced to 
take a screen dump after collecting the data and before doing 
anything else.

        I'm hoping to use KisMac as an integral tool in preparing charts 
and diagrams of wireless signal strength here in the house (which I 
can then integrate into a 3D model produced with SketchUp), using 
various wireless NICs (built-in Airport card in my PowerBook G4, 
Lucent WaveLAN Gold, Cisco AiroNet 350, Senao 2511CD-PLUS, and 
possibly others) as well as with different base stations and base 
station configurations (Snow, Extreme, Extreme+external antennas, 
etc...), across all the floors in the house.  I want to use this as 
the basis for an article I want to write on wireless networking, and 
which I believe I can get professionally published.

        But before I can do that, I need to get some more help from KisMac.

Brad Knowles, <brad.knowles@xxxxxxxxx>

"They that can give up essential liberty to obtain a little temporary
safety deserve neither liberty nor safety."
     -Benjamin Franklin, Historical Review of Pennsylvania.

GCS/IT d+(-) s:+(++)>: a C++(+++)$ UMBSHI++++$ P+>++ L+ !E-(---) W+++(--) N+
!w--- O- M++ V PS++(+++) PE- Y+(++) PGP>+++ t+(+++) 5++(+++) X++(+++) R+(+++)
tv+(+++) b+(++++) DI+(++++) D+(++) G+(++++) e++>++++ h--- r---(+++)* z(+++)

-- Binary/unsupported file stripped by Ecartis --
-- Type: image/png
-- File: KisMac-Table.png

-- Binary/unsupported file stripped by Ecartis --
-- Type: image/png
-- File: KisMac-Chart.png

Other related posts: