AMSAT-BR QO-100 FT8 QRPp Experiment

    Until yesterday only one station was able to reach -23 dB (EbN0 = 3 dB), but today one station was able to reach -24 dB and several stations are being able to reach -23 and -22 dB. I am puzzled to what could cause this. No changes have been made to the RX hardware or software setup.

    Hi Edson,

    I did achieve -24 finally, and today was the war of milliwatting from R3ICQ / PY4ZBZ abd RA9DA to challenge the 1st position :)

    Indeed, far from being a specialist, I assume that finally the Minimum Discernable Signal from the RX stations stays around -22/21, since this is the easy limit to reach all day long.

    I suppose that either by noise level drop on the transponder, or noise drop on the local RX station area for some reason, sometimes the MDS also improves for a couple of dB ?

    I guess this mioght be very tricky to go further down, right ?


    My rig is MKU2424 + 1m dish + 5.5 turns helix.

    Power out at MKU2424 : more or less 33 dBm (chinese wattmeter toleranced:/) + 33 Db attenuator pile-up = 0 dBm + 3 dB cable loss = -3 dBm at antenna.


    Finally, the RX station makes the whole job, and this were the performance is:)


    73, Pat

    Hallo ...


    some more have reached the -24dB .... but my receiver (85cm Dish, ADALM PLUTO) shows sometimes better signal levels of some stations ...


    anyone have an idea about the difference in decoding levels... in WSJTX ???

    Maybe it is interesting to hear some ideas about... and maybe it is helpful to get more about the monitoring station for this test...

    which hardware is used?


    Tnx and 73 de dg0opk

    It is difficult to know for sure what could cause the differences in the SNR. Variantions in the uplink and downlink radio paths would probably the the main candidates.


    The setup at the AMSAT-BR RX is the following:


    As refletor I have a 1.2m offset dish. I use SDR Console to correct the LNBF frequency drift, demodulate and route the demodulated audio samples (16-bit @ 48 kHz) over the network to a Linux PC using UDP datagrams. At the Linux PC I receive the samples using netcat and route the samples via a shell pipe to a command-line process (ft8d) I have developed to write a WAV file in sync with the 15 seconds FT8 time windows. Once the WAV file is complete, the process calls the FT8 decoder (the original wsjt-x jt9 command-line process). The decoder returns the spots, if any, to ft8d which then filters the CQ spots and forwards them to the AMSAT-BR web site.


    73, Edson PY2SDR

    Halo Edson and all others....


    many thanks for your information about the station setup... that is very interesting ....


    But i have no idea about the different decoding levels and the performance of WSJTX also on different days ....normaly the same signals on the transponder must reach the same SNR level if you have enough ground noise of the transponder.


    Maybe there are some other influences in the signal path of the propagation and maybe also in the satellite hardware?


    BTW: an very interesting test ....


    73 for now

    Michael dg0opk

    Hello Michael,


    I am very puzzled about why some days are different than others. There are many variables involved and not having a way to instrument them, it may be hard to know for sure what could be the cause for the differences we have been observing.


    It would be good not know the antenna RF input power levels and antenna gains that the participating stations are using.


    73, Edson PY2SDR

    I have exchanged a couple of emails with Joe K1JT about the QRPp experiment and the -24 dB limit for the FT8 decoder. The following are Joe's remarks.


    ###########################################################################

    By convention all SNRs reported by the software package WSJT-X are

    measured in a standard reference noise bandwidth of 2500 Hz.

    Detailed measurements of FT8 sensitivity in various simulated

    propagation conditions are summarized in a graph posted on the WSJT web site here:


    http://physics.princeton.edu/p…T8_Decode_Probability.png


    This plot was made by generating 1000 simulated FT8 signals at each

    specified SNR, subjecting them to simulated propagation by nine standard

    ITU propagation models, and counting the number of resulting decodes for

    each case.


    Propagation to the geostationary satellite QO-100 ahould closely

    approximate the AWGN (additive white Gaussian noise) path, for which the

    leftmost curve in the plot shows a 50% probability of decoding success

    at SNR = -20.8 dB and a very small (less than 1%) probability at SNR =

    -23 dB. The probability of decoding an FT8 signal with true SNR less

    than -22 dB is extremely small.


    Why is it, then, that the decoder sometimes reports SNRs as low -24 dB?

    And why do we never see reported SNR -25 dB or lower?


    All physical measurements have associated uncertainties. In principle

    we could estimate and report an uncertainty along with each measured

    value -- for example, something like SNR = -18 ± 2 dB. If WSJT-X were

    intended as a measurement tool, we would do it this way; but for its

    intended purpose such reports would be cumbersome, distracting, and

    irrelevant.


    Reported SNRs lower than -22 dB are nearly always underestimates of the

    true value. To avoid unwarranted focus of attention on very small

    reported SNRs, we arbitrarily clamp reported FT8 SNRs at -24 dB.

    ###########################################################################




    Hmmmmmmm... You are giving me ideas!


    Would you like to do some tests with WSPR?

    Yes, with pleasure.

    FT8 operating guide by ZL2IFB (available via hyperlink from WSJT-X on-line manual) provides a chart with performance comparison, which shows the best achiveable theorical SNR.

    WSPR supposed to overshoot our FT8 -24dB (-21 :) )



    Let us know if you would find some fun in arranging something at the end of the FT8 challenge.


    For some reason, I have no decode at all on QO-100 FT8. I assume this is a frequency drift issue...



    73, Pat

    The problem is known and lies in the operation of the program. In the directory C:\users\user\AppData\Local\WSJT-X\ there is an file: "ALL.TXT".

    This file contains all the calls and messages previously received.

    The first time the program decodes a signal it must necessarily receive the whole message correctly. Let's say the signal is at -20 dB. Subsequently, the program is able to receive a lower signal from the same station since the part of the message that possibly did not receive well reconstructs it thanks to the local archive contained, in fact, in the file: "ALL.TXT"

    Example:

    200324_115645 10489.540 Rx FT8 -20 0.9 1401 CQ PY4ZBZ GH70

    Subsequently PY4ZBZ reduces its signal by 3 dB but the program correctly receives only the blue part of the message.

    200324_115645 10489.540 Rx FT8 -24 0.9 1401 CQ PY4ZBZ GH70

    Here the program goes fishing in the archive and tries to reconstruct the missing part of the message. The same was used 30 years ago with QSOs via EME. You knew in advance who was on the other side and it was enough to receive only a few characters to consider the QSO valid.

    Delete the file "ALL.TXT" and you will see the -24, -23 etc disappear

    That does not apply to the experiment since I am not using the decoder with deep search enabled.

    Forgive me but maybe there is a basic misunderstanding.

    I don't know the details of your experiment. I don't know how you organized it. I am talking about JT digital systems in general and how they try to decode a message. I tried and tried again to delete that blessed archive file and the various programs, not only that of K1Jt but also the derivatives, become more "deaf" without that file. It is an experiment that anyone can replicate. And if after deleting that file, disconnect the Internet as well, you will have other surprises.

    Quote from I0LYL

    Forgive me but maybe there is a basic misunderstanding.

    I don't know the details of your experiment. I don't know how you organized it. I am talking about JT digital systems in general and how they try to decode a message. I tried and tried again to delete that blessed archive file and the various programs, not only that of K1Jt but also the derivatives, become more "deaf" without that file. It is an experiment that anyone can replicate. And if after deleting that file, disconnect the Internet as well, you will have other surprises.

    Lucio, I think you haven't read what I wrote, I do not use wsjt-x or any derivative. Perhaps there is a basic misunderstanding indeed.

    Just used the free time and nice weather to test if i can reach the QRPP receiver in PY with very small equipment.


    Using my old BU500 (@ 450mW) and a 4-turn helical roughly aimed to QO100 gives -13/-11dB SNR... much to strong!


    Then i used a old WiFi omnidirectional antenna > -16dB …. still to strong to participate…


    Next time i try with a dummyload ;)


    BTW: no problem with the TX frequency drift of the BU500 in direct sunlight :)


    73s DB8TF