[USRP-users] [Discuss-gnuradio] Antenna compatibility problem

Marcus Müller marcus.mueller at ettus.com
Tue Mar 22 07:54:33 EDT 2016


Hi SangHyuk Kim,

this question is unrelated to GNU Radio; I'm taking the freedom to reply
to usrp-users at lists.ettus.com, too, and ask you to continue discussion
there!

On 03/22/2016 12:39 PM, SangHyuk Kim wrote:
> Hi,
>
> I'm using CBX daughter board which supports from 1.2GHz to 6.0GHz with
> ANT500 (from 75MHz to 1GHz).
Reading https://greatscottgadgets.com/ant500/ reveals that ANT500 is
simply a telescopic monopole antenna. You adjust its length to roughly
one quarter of the wavelength of the frequency you're interested in!
>
> Actually, I didn't know this antenna coverage. 
>
> I have two question.
>
> i) Is there any problem when I use ANT500 with CBX ?
No; it's just an antenna; it should be roughly 50 Ohm in impedance,
which is the right impedance for all Ettus daughterboards.
> I have tested QAM and QPSK performance using OFDM using those devices
> and the results are poor(very unstable, too many bit error in very
> short range).
To what length did you extend the antenna? At what frequency where you
operating? What was your signal bandwidth? Did you make sure there were
no strong interferers?
> So, I doubt this is antenna compatibility with CBX.
Good news is that I don't think there's anything wrong with using that
antenna :)
>
> ii) Which antenna is best for CBX 40MHz(1.2GHz-6.0GHz) and USRP N210 ?
Obviously, that depends on what you want to do with your antenna.
(Frequency, directed link or broadcast, preference for selectiveness or
wideband behaviour, importance of matching, size, cost, ruggedness…)
> There are three products:
> 1) VERT400 (144MHz, 400MHz, 1200MHz Triband)
> 2) VERT900 (824-960MHz, 1710-1990MHz Dualband)
> 3) VERT2450 (2.4-2.5 and 4.9-5.9GHz Dualband)
There are millions of antenna types out there, and it all depends on
what you want to do.
As mentioned in context of other problems: As you are an engineer, you
should apply your communication knowledge to first model your problem
and then find a suitable solution.
Without knowing what you want to do, no one can help you, I'm afraid.
And the only person knowing what you really want to do is you, unless
you explain that sufficiently.

Best regards,
Marcus




More information about the USRP-users mailing list