two questions a) stress testing b) really low gain antennas

another couple of questions.

stress tests

I've been able to find a couple of commercial systems built to stress-test wi-fi networks, but are there any freeware products to do it? You know, something that would allow a single laptop to behave like a dozen (or a gross).

antennas

what can be done about getting a really low gain antenna? something between an Access Point with the antenna removed, and the standard single omni antenna the Access Point comes with?

thanks again

Reply to
cscus00
Loading thread data ...

Simulating a client radio is simply a matter of generating MAC addresses on the fly. A special program isn't really necessary as simply using ifconfig to change the MAC address on the fly is sufficient. See:

formatting link
dig out "Wellenreiter" for the entire system.
formatting link
formatting link
sample code.

The Perl code fragment works but takes too long to change the MAC address because it waits for the ARP table to expire. Send the inetd process a kill -1 after each change and it will go faster. I do the same thing with a simple shell script call to ifconfig in a loop (beause I'm a rotten Perl and C++ programmist).

I also take great care not to run this on a live system with potentially irate users. The limiting factor is the ability of the access point(s) to build a large enough table of MAC addresses heard and port numbers. Cheap junk will do 32 MAC addresses and crap out. Proxim/Orinoco access points will go up to 2000. If dual radio, do NOT assume that both radios share the same table.

I'm restricted by NDA from disclosing any results, but I would be very interested in the results of your testing.

Well, that's a new one. What are you trying to accomplish?

I just did a quick test with Netstumbler. Difference between an alleged 2dBi rubber ducky antenna and with both antennas terminated by

50 ohms is only about 20dB. My WRT54Gv1.1 apparently leaks quite a bit. 20dB would be a reduction in range to 1/10 of the normal (with antennas) configuration.

Try:

- 50 ohm dummy load. With the antenna removed or terminated, the circuit board leaks quite a bit. Few access points are properly shielded.

- XX dB attenuator driving stock rubber ducky antenna. Finding an attenuator with R-SMA or R-TNC connectors will be almost impossible. I had to make my own or use adapters.

- Leaky coax cable (Radiax).

- Helical antenna (1/4 wave wound into a spiral).

- 1/4 wave radiator inside a sealed paint can. Punch some holes in it to leak RF.

- Put access point inside an old microwave oven.

Reply to
Jeff Liebermann

I still had my test setup together so I tried it. I juggled the router for maximum signal strength to my laptop. The 3.8" paper clip was about 3dB below the alleged 2dBi rubber ducky antenna. That gives the paper clip a gain of about -1dBi or worse. However, the accuracy of my desktop test range and sloppy measureing tools leaves much to be desired.

Reply to
Jeff Liebermann

I use a 1dbi paper clip sometimes. But that's an estimated gain figure.

Reply to
Rôgêr

I seem to be getting some interference with this method. But it goes away when I turn the microwave off. Coincidence?

Reply to
Rôgêr

8*) It has to do with the RF-activated shielding in the microwave. By using RF-activated shielding, they extend the useful lifetime of the shielding, as it's only working when the microwave is on. Put a 2.4GHz portable phone in the microwave to keep the sheilding active (note, you must keep the phone in an active call). 8*)
Reply to
William P. N. Smith

Cabling-Design.com Forums website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.