[HamWAN PSDR] Testing performance of OFDM systems / radios

[Bart Kus]

Does anyone know how to do this?  I have a bunch of ideas written down 
at the Radio Analysis 
<https://www.hamwan.org/t/tiki-index.php?page=Radio+Analysis&structure=HamWAN> 
page, but honestly it feels like stabs in the dark. I don't have any 
true noise generators with adjustable bandwidth (donate! :)), so my 
simulated wideband interference signal consists of a bunch of peaks 
instead (FM'ing a carrier with a sine wave). Any tips or help appreciated.

Also seeking 1W+ 20dB+ 6GHz+ attenuators, and splitters/combiners (with 
good isolation) that can work @ 6GHz+.  I'd like to simulate a full cell 
site (3 PtMP sectors + some PtP links) within a totally confined coaxial 
system, and inject arbitrary noise into it.  Given the cost of these 
microwave attenuators + splitters, it might be easier to set it up with 
real antennas instead.  But that has problems of environmental noise and 
multipath / echoes.

Linked from the Radio Analysis page is the RouterBoard Metal 5SHPn 
<https://www.hamwan.org/t/tiki-index.php?page=RouterBoard+Metal+5SHPn&structure=HamWAN> 
page.  It has the result of 1 performance test so far, dealing with 
amplifier clipping leading to increased sideband energy.  This is of 
great importance in designing sectorized systems.  When these unexpected 
sideband emissions leak into adjacent antennas, you can shoot yourself 
in the foot.  So I'm trying to characterize the tolerance of the radios 
to nearby or in-band interference.  Doing this in TDMA/OFDM space is not 
trivial.

--Bart

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.hamwan.net/pipermail/psdr/attachments/20130218/f516a3c2/attachment.html>