Re: [Discuss-gnuradio] Designing a good receiver

[Mike Clark]

I'm using the RFX2400 tuned to 2.42e9. And I do always use attenuators
(10db on each device) when cabled :)

One thing I was thinking about after reading over the chapter
suggested by Brian is that I only had an antenna on the TX/RX side
with a terminator on the RX2 port. I'll test with an antenna on both
ports and report back. If you have any other advice though, I'd love
to hear it.

Mike

On Thu, Jun 2, 2011 at 8:10 AM, Marcus D. Leech <address@hidden> wrote:
> On 06/02/2011 07:10 AM, Mike Clark wrote:
>> This comes from a discussion over on the USRP-users mailing list, but
>> I felt it would be more appropriate to post here. Over on that list,
>> someone was asking about problems they were having consistently
>> receiving data on their USRP. My background is in CS, not RF comms, so
>> please forgive me as I'm sure the answer will be fairly standard :)
>>
>> Anyways, the question I have is, is there a general procedure one can
>> follow to design a decent receiver in gnuradio? For example, I have a
>> project that I'm using for experimentation where my receiver looks
>> like this: USRP Source -> GMSK Demod -> Packet Decoder -> File Sink.
>> This works well when I have the USRPs cabled together and even when I
>> have antennas with line of sight (I haven't checked max distance).
>> When I don't have line of sight, however, I stop receiving packets.
>> Are there any other gnuradio blocks I can add in to my setup which
>> will help get better performance when there is no line of sight? I
>> haven't tested it, but my guess is that when there is no line of
>> sight, there must be a frequency shift or something of that nature
>> happening which ruins the connection.
>>
>> Mike
>>
>>
> There are a few things that could be a problem:
>
>    o multipath
>    o received signal strength
>
> What type of daughtercards are you using, and when you cable them
> together, I hope that you're
>  using an attenuator?  Otherwise, you risk damaging the LNA on the
> receive side, which will cause
>  it to be either partially or completely deaf.
>
> Frequency offset *is* a problem, but is usually not sensitive to path
> loss.  In most *real* radios, you
>  need to determine the frequency offset, as seen by the receiver and
> compensate.  This is more, or less,
>  of a problem depending on the signal bandwidth.  A wideband signal is
> less sensitive to frequency
>  offsets, because the 'error' is a smaller fraction of the total
> bandwidth.  OFDM is the exception, since
>  in effect there are dozens of individual carriers, of relatively
> narrow bandwidth.
>
>
>
>
> --
> Principal Investigator
> Shirleys Bay Radio Astronomy Consortium
> http://www.sbrac.org
>
>
>
> _______________________________________________
> Discuss-gnuradio mailing list
> address@hidden
> https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
>