Page 1 of 1

GRAPHS

Posted: Wed Jul 06, 2016 5:48 pm
by WU2J
I wish to have a graph monitor a signal in percent (%) so I can compare
gains on different antennas> How does one do that?

PS Currently I can see many APs and monitor their strengths in dBm. I also
found a way to change antennas and see their continuous red strengths
in dbm. This graph starts on the right and works it way to the left.

I'd like to see their strengths start on the left and move to the right plus
be in percentage terms.
Byron

GRAPHS

Posted: Wed Jul 06, 2016 6:34 pm
by pferland
Vistumbler does not support using more than one network card at a time. Currently the only way to have more than one antenna active is to have more than one instance of Vistumbler running.

One Instance uses Antenna A on Card A while the second instance uses Antenna B on Card B.

You would then have to have both graphs running and compare the two.

Unless I am not understanding your question correctly.

-Phil

GRAPHS

Posted: Thu Jul 07, 2016 8:31 am
by WU2J
You do not understand the question. I only have one network card (Alfa). I physically disconnect and change
antennas for comparisons between them. This is done hundreds of times by Andrew McNeil in his YouTube videos (with various antennas) using Vistumbler. I wish to duplicate his procedures with the same graphs

GRAPHS

Posted: Thu Jul 07, 2016 10:57 am
by arizonajon
Connect your Wi-FI sniffing setup to your computer with the first test antenna attached.
Start up Vistumbler but do not push the "Scan APs" button yet.
Go to "View>Graph>Use RSSI in graphs" and uncheck if it's currently checked. That will make sure you get the data as a percentage.
Click the "Graph 1" button.
Select "Scan APs". Highlight in the list the AP that you want to use as the reference for your antenna measurements. The signal strength data you expect should start to plot on the graph.
Once you've collected sufficient data points, toggle "Stop Scan".
Connect antenna #2 to your wi-fi sniffing setup.
Select "Scan APs".
Rinse and repeat.

GRAPHS

Posted: Thu Jul 07, 2016 5:29 pm
by WU2J
Thank you. Will do all tomorrow.
Byron, Florida USA

GRAPHS

Posted: Fri Jul 08, 2016 7:35 pm
by WU2J
Ok, fine, thank you. Now, how can I shorten the time interval between points
so data points are very close together?

GRAPHS

Posted: Fri Jul 08, 2016 8:26 pm
by pferland
So, the intervals that the signal are captured at is based on the Refresh Loop variable. You can find it in the Settings->Misc Settings menu. It is base on milliseconds and is defaulted to 1 second. However, It is really not recommended to set it less than one second since it sit tends to take about 1 second for the process loop to run once you get more than a few hundred APs in your list. On my laptop, I had 35 active AP's and 65 total and was averaging around 250ms for a loop run. This can be seen on the main UI under the label "Actual loop time".

Let me know if you have any other questions.

-Phil

GRAPHS

Posted: Sat Jul 09, 2016 2:37 pm
by WU2J
Thank you very much for your instructions, Phil. Perhaps a picture in the beginning
would have been more helpful. Here is one of many Andrew McNeil's YouTube videos.
Kindly look at the Vistumbler presentations from 7:45 to 8:56 minutes and you
will understand what I desire to see. One second would be okay if the spacing
was close together.

https://www.youtube.com/watch?v=4t6OliwhJ5g

Thank you very much.
Byron

GRAPHS

Posted: Sat Jul 09, 2016 2:56 pm
by pferland
The graph that Andrew McNeil is using is "Graph 2" (aka the Bar Graph). In that video he is using a quite old version of Vistumbler and I am not sure how he got it to be tied to the left hand side. I will have to ask ACalcutt how it is done, it also may have been changed to only show from right to left in later versions.

If it is showing gaps of 0 signal strength in the Signal map, then you might have Graph Dead Time enabled. To disable that go to View->Graph->Graph Dead Time.

If that is still not what you are looking for, would you be able to take a screenshot of what you would like to see?

GRAPHS

Posted: Sat Jul 09, 2016 8:12 pm
by WU2J
McNeil's graphs/presentation is exactly what electronic engrs and antenna
designers want to view and analyze. That's why he has such a following
worldwide. Vistumbler is a very good program. It does have a competitor
called Network Stumbler which appears to be off by -15 dBm on Receive
Strength plus it lacks your desired graphs. Thanks for your help, Paul, I
think you pretty much answer things and I am more comfortable with the
program now.
Cheers,

Byron

GRAPHS

Posted: Wed Jul 13, 2016 6:24 pm
by arizonajon
Hi Byron -
I too viewed Mr. McNeil's presentation. Neat antenna design!
If what you mean by "Network Stumbler" is NetStumbler, I used to run that over a decade ago. It was a very nice software program and (I believe) worked directly with the WLAN card drivers. It could do some pretty neat things, like get the card driver's own estimate of RSS. Alas, the number and complexity of WLAN interfaces has grown enormously since 15 years ago when I first started wardriving with my Orinoco Gold Wi-Fi card. I had to give up on NetStumbler because it had not been updated since 2004 (?) at version 0.4.0, and wouldn't work correctly with any of many cards (or USB wi-fi dongle) I'd purchased since about 2008.
Because of the vastly increased number of and increased complexity of WLAN interfaces, Microsoft, from Windows Vista forward, established a common developer interface, called netsh, for access to the WLAN device(s). You can see it yourself if you start a command line instance and run "netsh". Sadly, one of the things MS chose to do was to make the default signal level reporting in percent, not in dBm. So what you get there is technically an RSSI, not an RSS. And as a career-long RF engineer, ham radio operator, and wireless telecom systems engineer, that's no way to monitor a wireless system. It should be in dBm and as estimated by the receiver itself, not munged up and neutered by middleware.
Vistumbler uses the netsh command as the way to grab the WLAN information. It's as fast as Windows will process it. I've got Vistumbler running on several Windows PCs and with different horsepowers and the loop time varies quite a bit, as Phil points out.
With the netsh command, most write ups on it (for instance, http://stackoverflow.com/questions/1579 ... o-rssi-dbm) suggest that 1% step is equal to 0.5 dB. Sadly, that can't be the case, as I've measured myself in the lab with calibrated attenuators. And, according to Mr. McNeil's own numbers, he shows the change from a "5.5-6 dBi dipole antenna" to his "9 dBi biquad antenna" as a jump from "60-65%" to "85%". So, his two antennas vary in gain by about 3.5 to 3 dB. However, using the numbers that are available via the netsh command, he is seeing a 12.5 to 10 dB improvement, which clearly isn't possible with the published 1% = 0.5 dB estimate. Indeed, in my own testing, netsh is not linear at all and the % value that is native is only of mild interest, and mostly useless for anything more than putting a wet thumb in the air to feel what way the wind's blowing. Not completely useless, mind you, but not very useful. It's really too bad that Microsoft didn't have a couple of real RF engineers specifying the design for netsh! %^)
If you do have access to an old Wi-Fi card, supported by NetStumbler, and perhaps an old PC running XP, then NetStumbler is the way to go. From what I remember of NetStumbler, the RF dBm level reported was actually useful and represented real-world values reasonably well.

Cheers and 73 - Jon N7UV

GRAPHS

Posted: Thu Jul 14, 2016 4:43 pm
by WU2J
Hi Jon,
Thanks for the email and info (just got back). We have similar backgrounds. Yeah, I meant Net Stumbler!
As a member of the G-QRP Club Int'l, I wrote quite a few design articles (SSB Xcvr 17m, 10W Linear Amp, Ant. articles, Resistance bridge, etc) for their magazine. Also developed my own mini transmission line transformers from 50 ohms down to 5.6 and up to 200 ohms for HF Ant. Matching. Jerry Sevick (Bell Labs Engr) was a good friend of mine. So was the famous Doug DeMaw.
Anyway, I recently built 3dB, 4dB, & 6 dB pads between some RP-SMA connectors (2.4GHz) using copper clad PC boards to pack'em in tiny. I use the 3 dB pad between the WiFi antennas and the Alfa card (in and out) to check gains on Vistumbler using a little ~2 dBi Gnd Plane vertical as a ref. to get some dBd figures. I also have two of Andrew McNeils ant. (BiQuad Yagi and Long Range BiQuad Yagi) He's quite a guy! Compared coax with USB cables ~13 to 20 ft long as feedlines to the antenna. A ~13 ft RG-8X was 'right on' with a loss of 3 dBat 2.4 GHz compared to a 20ft length of 28/22 AWG USB extension cable.
Experimenting now with larger size wires in USB cables. I'm using an old Dell D610 PC (Xp) with the Alfa card as a setup plus a telescoping alum pole with a screw mount on top. Have two different QTHs at which to play. One doesn't have internet avail. so I'm trying to connect to Walmart 3200 ft away. Got a tree problem, too. I did many HF Ant designs and gain measurements as a hobby in the '91-'98 timeframe. Built individual ant. meas. rcvrs that used VXO's for stability. Retired here in FL since '91....
Byron WU2J (ex-WB2HAL)

GRAPHS

Posted: Thu Jul 14, 2016 5:03 pm
by WU2J
Sorry 'bout that. This forum is for Vistumbler. I have NetStumbler and compared
to Vistumbler's dBm figures, NetStumbler is down -15dBm (difference) on all
APs. So, my question was, which program has the most accurate dBm figures?
Byron

GRAPHS

Posted: Thu Jul 14, 2016 6:40 pm
by arizonajon
Hi Byron -

Wow, sounds like you've been deep in this for a while!

Since Vistumbler uses the native Windows netsh, that's what's discussed in the link
http://stackoverflow.com/questions/1579 ... o-rssi-dbm
0.5% = 1 dB, which, like I said above, is just wrong. No better than a warm fuzzy.

For NetworkStumbler, if it's a WLAN card that's natively supported by NetStumbler, then you should get the WLAN card's own estimate of RSS, which is actually getting to the "S/dBm" meter in the receiver.

In any event, your mileage may vary %^). You may want to use some of your homebrewed pads to provide a "calibration" for either program. If putting in a 6 dB pad changes the measured level from -nn.n dB to -nn.n - 6 dB, then adding an antenna with 6 dB additional gain should bring the reported value back up to around -nn.n dB. Not precise, but better rule of thumb.

Cheers and 73 - Jon N7UV

GRAPHS

Posted: Mon Aug 01, 2016 5:44 pm
by ACalcutt
Hi Jon,

I just wanted to point out that in native wifi mode, which has been the default for a while, the rssi value should be the WLAN cards own reading of RSSI. You are correct that when using netsh mode (native wifi mode unchecked) it is a estimate, but in native wifi mode it should be a real value from the wifi card.

From the native wifi api documentation here (https://msdn.microsoft.com/en-us/librar ... s.85).aspx)

Code: Select all

lRssi
The received signal strength indicator (RSSI) value, in units of decibels referenced to 1.0 milliwatts (dBm), as detected by the wireless LAN interface driver for the AP or peer station.
The trouble is its up to the manufacturer as to what to return, so it isn't always consistent between manufactures with different drivers

GRAPHS

Posted: Wed Aug 03, 2016 10:01 pm
by arizonajon
well knock me down and call me cornpone - i guess i missed that somewhere along the way, Andrew! I will make sure that my mobile setup does have that function checked and will also go back and measure my RSS vs what I'd expect, and see how well my Wi-Fi dongle is doing vs what it's reporting!

Cheers - Jon N7UV

GRAPHS

Posted: Sun Aug 07, 2016 1:41 pm
by WU2J
Hello again,
Vistumbler is fine and I've established good graphs with important information...now I can't find a
method to print it so I can label the different items tested.

Byron :|

GRAPHS

Posted: Sun Aug 07, 2016 3:04 pm
by pferland
Using the wifidb graph under the experimental menu might work, it creates a png image from the AP data. Or you can get the data from the raw access db file.

GRAPHS

Posted: Mon Aug 08, 2016 1:17 pm
by WU2J
I can't find graph under Experiment Menu?? And, where is the "Raw Access dB file?
Thanks,
Byron

GRAPHS

Posted: Mon Aug 08, 2016 4:37 pm
by pferland
Oops sorry about that it is under the WifiDB Menu called Graph Selected AP Signal to Image.
Untitled.png
Untitled.png (16.35 KiB) Viewed 14951 times

GRAPHS

Posted: Wed Aug 10, 2016 10:27 am
by WU2J
I tried that and the 'hour glass' flashed a second but nothing happens. That's
a shame we can't print a graph but I certainly appreciate all your time and
help. Byron :D

GRAPHS

Posted: Wed Aug 10, 2016 11:13 am
by pferland
Did you have an access point selected in the list when you selected the menu item? Will give it a look later today myself.

GRAPHS

Posted: Thu Aug 11, 2016 2:50 pm
by WU2J
I have the Graph with several patterns done over a period of time as I measured and compared
different antennas. I only want to print the Graph. I have taken some photos but they are
not clearly focused.

GRAPHS

Posted: Thu Aug 11, 2016 6:58 pm
by pferland
The WiFiDB Graph should work, I tested it out today. Select the AP in Vistumbler, go to the WiFiDB Menu and select the Graph item and it should open a browser page to https://api.wifidb.net/wifi/ with GET parameters that are the details of the AP that you selected.

Andrew can correct me, but If I remember correctly, the way that Andrew is using AutoIt to draw to the screen is not easily savable to an image file.

Or you can use the Snipping tool in windows to grab a screenshot of the graph in vistumbler or use the PrintScreen key and save that to an image editor like MS Paint.

GRAPHS

Posted: Sun Aug 14, 2016 2:55 am
by WU2J
Thanks! Another problem I have vs. what I see on YouTube for Andrew McNeil is that when I have
a list of APs and see the weakest with signals down -100dBm, and click on any of them and use
Graph 2, none EVER GO BELOW 35% ON THE GRAPH. Is there a method to correct the dynamic
range of the display so that such a weak AP is near zero signal level?
Also, what is the 2.4GHz Experimental graph?
Byron

GRAPHS

Posted: Sun Aug 14, 2016 1:29 pm
by WU2J
I just downloaded another wireless card on another computer and see "No Adapter Found" This is a new
Mediatek RT2870 wireless LAN card and apparently it is not approved yet by Vistumbler? I posted under
NO ADAPTER FOUND too. Clicked on UPDATE MANFUACTURERS under EXTRA but that didn't help. :roll:
Byron

GRAPHS

Posted: Sun Aug 14, 2016 1:48 pm
by pferland
I am not sure how Andrew detects the network adapters. I will have to ask him when he gets back later today.

The Update Manufactures is just for the Access Point MAC Address match up to the manufacture that made it.

As for the Experimental 2.4Ghz graph is a graph that shows the Channels that are in use by the Access Points near by you.

It looks like Graph1 and Graph2 is not in % but in db, I wonder if Andrew can make it selectable as to if it will graph % or DB. Will ask him later.

If I remember correctly, its not possible to go higher than -30db (close to 100%) and lower than -90db (close to 0%) with WiFi.

GRAPHS

Posted: Sun Aug 14, 2016 3:25 pm
by WU2J
I typed in the Adapter model info and re-downloaded Vistumber...no problem now, got it.
Also under Vistumbler Settings I found the dBm limits and, yes, it does work to at least
-15dBm (where I set it and also set to -100Dbm= don't know it this will take me to the
bottom of a graph on a weak sig/ ALL IS FINE FOR THIS JR. MEMBER!

GRAPHS

Posted: Sun Aug 14, 2016 3:26 pm
by WU2J
YES, THE -100 DBM WORKS TOO- GREAT!
BYRON WU2J

GRAPHS

Posted: Sun Aug 14, 2016 3:27 pm
by WU2J
:D :) :? :lol:
:lol: BYRON

GRAPHS

Posted: Sun Aug 14, 2016 9:14 pm
by ACalcutt
Hi Byron,

The default graphs are in dBm, but using signal percent is still supported. There is an option, uncheck (View --> Graph --> Use RSSI) in graphs and the graph will change to be based on the signal percent value.

RSSI changes are finer and are the value read from the wifi card when using native wifi mode, so it is better to use in the graph.... but you can still use signal percent value if you want.

Also, If you are using the graph at all, you may also want to check (View --> Graph --> Graph Dead Time)

-Andrew

GRAPHS

Posted: Tue Aug 16, 2016 12:15 pm
by WU2J
Yeah, I figured that out some time ago and use BOTH at various times for
different reasons. I have 3, 4, and a 6 dB attenuator so I can compare/correlate
dBm at 10 dbm. Thanks for a useful tool, Andrew!

GRAPHS

Posted: Tue Aug 16, 2016 1:28 pm
by WU2J
Oops, what is "NATIVE wifi mode"??

GRAPHS

Posted: Tue Aug 16, 2016 5:46 pm
by pferland
Native WiFi Mode uses the Native Wif API in windows instead of using the NetSH commands and scraping the output that Vistumbler originally did to get the Wireless APs. You can still use the NetSH mode, but it s slower than the API. To toggle it check or uncheck Options->Use Native Wifi.

Andrew really needs to make some really basic documentation on what each menu item does and an overview of the settings windows. :wink: :lol: