[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Linrad] Re: Update on SDR-IQ + Linrad + MAP65

Hi Ed,

Many thanks for your note!  I'll let you know when there's something 
ready to test.  You should be able to run it on a single (fast) computer.

Let's be sure to work on 2m this weekend!  If you call CQ, I'll find you...

     -- Joe, K1JT

Edward Cole wrote:
> At 07:54 AM 11/13/2008, you wrote:
>> Yesterday I spent some time playing with an SDR-IQ (from
>> RFSpace) on both HF and 2m EME.  It is an impressive little
>> box: very simple, it just does what it is supposed to do.
>> It works fine with Linrad, as well as with the SpectraVue
>> software by AE4JY.  Running the SDR-IQ under Linrad, I found
>> that 2m JT65 EME signals are easy to see on the waterfall.
>> I'm sure that none of this will surprise any reader of this
>> list who has already gained some experience with the SDR-IQ.
>> I have worked out, on paper, a scheme that should allow
>> MAP65 to work with an unmodified SDR-IQ running under
>> Linrad, while minimizing necessary changes to the MAP65
>> program code.  The scheme involves choosing suitable
>> decimation parameters CIC2, CIC5, and RCF for the SDR-IQ
>> (these are parameters you can set from within Linrad) and
>> suitable lengths for the FFTs used in MAP65.
>> For your interest, here's a table showing some parameter
>> values in the present MAP65 (used with 96 kHz soundcard
>> input) and in the proposed "MAP65-IQ" as it would be
>> configured for the SDR-IQ:
>> Parameter                      Soundcard      SDR-IQ
>>                                 (MAP65)        (MAP65-IQ)
>> -----------------------------------------------------------
>> f0  RF sample rate (Hz)          -            66,666,667
>> nd  Decimation factor            -            714
>>      CIC2 CIC5 RCF                -            7 17 6
>> f1  Baseband sample rate (Hz)  96000.0        93370.68
>> n0  FFT for sync search        32768          32768
>> w0  Bin width (Hz)             2.930          2.849
>> n1  FFT full transmission      5,376,000      5,378,240
>>      Factors                    7 5^3 3 2^11   7^5 5 2^6
>> t1  Time for complex FFT (s)   1.39           0.66
>> n2  Inverse FFT, each signal   77175          79380
>>      Factors                    7^3 5^2 3^2    7^2 5 3^4 2^2
>> t2  Time for complex FFT (s)   0.00607        0.00324
>> f2  Final sample rate (Hz)     1378.125       1378.102
>> Both systems provide usable bandwidths of 85-90 kHz -- more
>> than enough to cover the relevant JT65 sub-bands for EME on
>> 144, 432, and 1296 MHz.
>> For each one-minute block of received data, MAP65 does 279
>> FFTs of length 32768 to get power spectra for all required
>> half-JT65-symbol intervals.  The resulting (32768 x 279) 2-D
>> spectra are searched for JT65 sync signals and shorthand
>> messages.  The slightly different bin widths w0 in standard
>> MAP65 and MAP65-IQ can be easily accommodated in the software.
>> MAP65 then does an FFT over the full span of received data
>> (about 52 seconds), with the complex data arrays zero-padded
>> to length n1.  Inverse FFTs of length n2 are then done with
>> data around each frequency where a sync tone was found.
>> This process yields downsampled complex data with effective
>> sample rate f2=f1*n2/n1.  As you can see in the table above,
>> the parameter values given in the MAP65-IQ column yield an
>> f2 with almost the same value as in the present version of
>> MAP65.  This condition is desirable because it requires a
>> minimum of changes to the remaining downstream program code.
>> As shown in the table for both MAP65 and MAP65-IQ, the
>> selected values of n1 and n2  can be factored into a number
>> of small primes.  This condition is necessary for the
>> efficient computation of FFTs.  The FFT lengths for MAP65-IQ
>> are actually "better" than those in the present MAP65, in
>> that their FFTs run nearly twice as fast (see values of t1,
>> t2 in the table).  Of course, since MAP65 operates on
>> dual-polarization data while MAP65-IQ processes a single
>> polarization, MAP65-IQ already has less than half the
>> computational load.  MAP65 is fast enough on a modern
>> computer.  MAP65-IQ will be several times faster.
>> MAP65-IQ should be an excellent tool for anyone using JT65
>> with a single-polarization (or switchable polarization)
>> receiving system on 144 and 432 MHz.  It should also make a
>> superb JT65 system for 1296 MHz, where circular-polarization
>> EME is the norm.
>> I hope to find time to work on code for MAP65-IQ over the
>> next few months.
>>         -- 73, Joe, K1JT
> Joe,
> Thanks very much for working on this.  I will try this out with my 
> SDR-IQ and Linrad.  I guess you are still running Linrad on a 
> separate computer from MAP65 (or not).  I have not gotten into the 
> interfacing of Linrad to MAP65 since I was waiting for the SDR-IQ solution.
> Is there things to do to make Linrad talk to MAP65 in the above 
> procedure?  I can also run Linrad on the same 2.33 GHz computer If I 
> can find out how they port together.
> ***********************************************************
> 73, Ed - KL7UW              BP40iq, 6m - 3cm
> 144-EME: FT-847, mgf-1801, 4x-xp20, 185w
> 1296-EME: DEMI-Xvtr, 0.30 dBNF, 4.9m dish, 60W (not QRV, yet)
> http://www.kl7uw.com       AK VHF-Up Group
> NA Rep. for DUBUS: dubususa@xxxxxxxxxxx
> *********************************************************** 
> > 

You received this message because you are subscribed to the Google Groups "Linrad" group.
To post to this group, send email to linrad@xxxxxxxxxxxxxxxx
To unsubscribe from this group, send email to linrad+unsubscribe@xxxxxxxxxxxxxxxx
For more options, visit this group at http://groups.google.com/group/linrad?hl=en