Loading ...
Sorry, an error occurred while loading the content.

Drift and Width of WSPR signals

Expand Messages
  • Joe Taylor
    I ve been away for a few days, and I see there has been some discussion of what the WSPR Drift parameter means. I will also take this opportunity to
    Message 1 of 9 , Apr 28, 2008
    • 0 Attachment
      I've been away for a few days, and I see there has been some
      discussion of what the WSPR "Drift" parameter means. I will
      also take this opportunity to introduce a new parameter W
      (for "Width") made available in WSPR v0.7.

      You can think of "Drift" as measuring the slope of the
      slightly fuzzy horizontal line that represents each received
      signal on the WSPR waterfall display. Zero drift means the
      line is horizontal -- constant frequency -- as it should be.
      Positive drift corresponds to a line sloping upward from
      left to right.

      As described in the "Quick Start Guide for WSPR v0.6",
      version 0.6 measured and reported drift in units of Hz per
      transmission -- that is, total upward or downward change
      across the full width of the fuzzy line.

      As you can read in the "Quick Start Guide for WSPR v0.7" at
      http://physics.princeton.edu/pulsar/K1JT/WSPR_Quick_Start.TXT
      v0.7 measures and reports drift in units of Hz/minute. This
      means that the numbers will be smaller by a factor 1/2. An
      advantage of the new scale is that nonzero reported drifts
      will nearly always be significant when compared with the
      measurement uncertainty.

      Of course, if only one measurement were available it would
      be impossible to tell whether the transmitter or receiver is
      drifting. However, in practice (and with most equipment
      being used with WSPR) it will nearly always be the
      transmitter that is less stable. Note that it is very easy
      to identify a drifting receiver: its effect will be to make
      nearly all received signals appear to drift in the same
      direction and by the same amount.

      WSPR v0.7 provides a new measurement labeled "W" for Width.
      This quantity is the spectral width of the received signal
      after all tone steps have been removed. It is possible to
      do this because the tight synchronization and strong FEC in
      every decoded WSPR signal means that the waveform generated
      at the transmitter is known exactly. This same waveform,
      reconstructed at the receiver, is numerically mixed or "beat
      against" the received signal. The result should be exactly
      the "zero beat" condition, except for small errors in the
      measured frequency, uncompensated drift errors, and phase-
      and Doppler shifts produced by the propagation path.
      Usefulness of the width measurement remains to be
      demonstrated, but I am imagining that it will usually be
      dominated by path effects and therefore will be interesting
      from a propagation-study point of view. Measured values of
      W have small values (less than 0.7 Hz, say) most of the
      time, but on some paths -- trans-polar paths,
      trans-equatorial nighttime paths, ...? -- may be larger. We
      will see...

      Your comments and observations will be welcome!

      -- 73, Joe, K1JT
    • Joe Taylor
      GM all, Last night a few of us ran our WSPR systems on 160 m, dial frequency 1.8086 MHz. I copied the following unique messages, most of them many times: F1VS
      Message 2 of 9 , Apr 29, 2008
      • 0 Attachment
        GM all,

        Last night a few of us ran our WSPR systems on 160 m, dial frequency
        1.8086 MHz. I copied the following unique messages, most of them many
        times:

        F1VS JN04 40
        F6IRF JN35 40
        K9FH EN62 30
        KC2QII FN20 30
        W1BW FN42 30
        W1BW FN42 37
        WD4KPD FM15 30
        WD4KPD FM15 37
        WD4KPD FM15 40

        My 3 Watts were copied by each of the above stations and also by VE1VDM
        (FN85), K6MBY (CN88), WP4SM (FK68), K0OG (EM47), W8LIW (EN81), and
        KC2RXS (FN32). Other stations also had good success.

        We should get a good turnout on 160 m again tonight... perhaps at a
        higher frequency, say 1.8366 dial. Watch the WSPRnet page for details,
        and join us if you can!

        -- 73, Joe, K1JT (FN20qi)
      • Ian Wade
        From: Joe Taylor Date: Tue, 29 Apr 2008 Time: 07:56:05 ... Hi Joe, This looks *very* interesting! Quick question: Where is the WSPRnet
        Message 3 of 9 , Apr 29, 2008
        • 0 Attachment
          From: Joe Taylor <joe@...>
          Date: Tue, 29 Apr 2008 Time: 07:56:05

          >GM all,
          >
          >Last night a few of us ran our WSPR systems on 160 m, dial frequency
          >1.8086 MHz. I copied the following unique messages, most of them many
          >times:
          >
          >F1VS JN04 40
          >F6IRF JN35 40
          >K9FH EN62 30
          >KC2QII FN20 30
          >W1BW FN42 30
          >W1BW FN42 37
          >WD4KPD FM15 30
          >WD4KPD FM15 37
          >WD4KPD FM15 40
          >
          >My 3 Watts were copied by each of the above stations and also by VE1VDM
          >(FN85), K6MBY (CN88), WP4SM (FK68), K0OG (EM47), W8LIW (EN81), and
          >KC2RXS (FN32). Other stations also had good success.
          >
          >We should get a good turnout on 160 m again tonight... perhaps at a
          >higher frequency, say 1.8366 dial. Watch the WSPRnet page for details,
          >and join us if you can!
          >
          > -- 73, Joe, K1JT (FN20qi)


          Hi Joe,

          This looks *very* interesting!

          Quick question: Where is the WSPRnet page?

          --
          73
          Ian, G3NRW
        • wgtayloruk
          Hi Joe & All, ... A higher frequency such as this would be much better for UK participation as the UK allocation is from 1.810 MHz. I received last night for a
          Message 4 of 9 , Apr 29, 2008
          • 0 Attachment
            Hi Joe & All,

            --- In wsjtgroup@yahoogroups.com, Joe Taylor <joe@...> wrote:

            > We should get a good turnout on 160 m again tonight... perhaps at a
            > higher frequency, say 1.8366 dial. Watch the WSPRnet page for
            > details,
            ---

            A higher frequency such as this would be much better for UK
            participation as the UK allocation is from 1.810 MHz.

            I received last night for a short while & logged F1VS, F6IRF & DF1VB,
            all many times.

            Gary - G4WGT.
          • Ian Wade
            From: wgtayloruk Date: Tue, 29 Apr 2008 Time: 18:00:56 ... Gary, I wrote a similar comment in a reply earlier today, then scrubbed
            Message 5 of 9 , Apr 29, 2008
            • 0 Attachment
              From: wgtayloruk <wgtaylor@...>
              Date: Tue, 29 Apr 2008 Time: 18:00:56

              >Hi Joe & All,
              >
              >--- In wsjtgroup@yahoogroups.com, Joe Taylor <joe@...> wrote:
              >
              >> We should get a good turnout on 160 m again tonight... perhaps at a
              >> higher frequency, say 1.8366 dial. Watch the WSPRnet page for
              >> details,
              >---
              >
              >A higher frequency such as this would be much better for UK
              >participation as the UK allocation is from 1.810 MHz.
              >

              Gary,

              I wrote a similar comment in a reply earlier today, then scrubbed it
              when I realized that with a dial frequency of 1808.6 kHz, the actual
              transmitted centre frequency would be 1.5 kHz higher, at 1810.1 kHz.
              Only 100 Hz inside the band, but inside nevertheless -- provided you
              remember to use USB!

              --
              73
              Ian, G3NRW
            • john flinn
              I would also favor 1.8366 dial freq. It is well above the DX window and low enough to be away from SSB gang. Some of the CW ops on 160 have been complaining
              Message 6 of 9 , Apr 29, 2008
              • 0 Attachment
                I would also favor 1.8366 dial freq. It is well above the DX window and low
                enough to be away from SSB gang. Some of the CW ops on 160 have been
                complaining about the digital stations in the lower 15 khz of 160 meters.
                Will be QRV this evening especially during grey line to Eupopa.
                73 John W9SE



                >From: "wgtayloruk" <wgtaylor@...>
                >To: wsjtgroup@yahoogroups.com
                >Subject: [wsjtgroup] Re: WSPR on 160 m
                >Date: Tue, 29 Apr 2008 18:00:56 -0000
                >
                >Hi Joe & All,
                >
                >--- In wsjtgroup@yahoogroups.com, Joe Taylor <joe@...> wrote:
                >
                > > We should get a good turnout on 160 m again tonight... perhaps at a
                > > higher frequency, say 1.8366 dial. Watch the WSPRnet page for
                > > details,
                >---
                >
                >A higher frequency such as this would be much better for UK
                >participation as the UK allocation is from 1.810 MHz.
                >
                >I received last night for a short while & logged F1VS, F6IRF & DF1VB,
                >all many times.
                >
                >Gary - G4WGT.
                >
                >
                >
                >
                >
              • Joe Taylor
                A good turnout of WSPR activity on 160m last night. I copied 25 unique messages on 1.8366 MHz (dial frequency, USB), from 15 different stations. Best DX was
                Message 7 of 9 , Apr 30, 2008
                • 0 Attachment
                  A good turnout of WSPR activity on 160m last night. I copied 25 unique
                  messages on 1.8366 MHz (dial frequency, USB), from 15 different stations.

                  Best DX was F6IRF (copied four times between 0130 and 0414 UTC) and F1VS
                  (copied 11 times between 0130 and 0446 UTC). Conditions at my end were
                  quiet, but the transatlantic path was not producing strong signals. The
                  French stations were both running 10 W, and their SNR here varied from
                  -25 to -29 dB.

                  -- 73, Joe, K1JT
                • Joe Taylor
                  Dear WSPR Friends, WSPR v0.7 (r752) seems to be stable and free of significant bugs, so I am working now on the possible future evolution of the program. I
                  Message 8 of 9 , May 1, 2008
                  • 0 Attachment
                    Dear WSPR Friends,

                    WSPR v0.7 (r752) seems to be stable and free of significant
                    bugs, so I am working now on the possible future evolution
                    of the program. I hope to be ready to share some of my
                    ideas with you in the reasonably near future. Present plans
                    involve an optional "reply mode", as discussed here earlier,
                    and eventually an optional "QSO mode" as well.

                    In the meantime, I have a request that should help to make a
                    smooth transition to the next version of WSPR easier.

                    Since the first WSPR version, all transmitted messages have
                    consisted of a callsign, a grid locator, and an integer
                    power level in dBm. For example, my standard message is
                    something like:

                    K1JT FN20 30

                    The power-level indicator is coded into 7 bits, and
                    presently all values from -64 to +63 are permitted.

                    In the future, WSPR will allow you to select a dBm level
                    only from a drop-down list with the entries

                    0 3 7 10 13 17 20 23 27 30 33 37 40 43 47 50 53 57 60

                    The values correspond to all power levels from 1 mW to 1 kW,
                    in a 1 2 5 10 20 50 ... sequence. You will need to select
                    the closest one to the power level you are actually using.

                    My request is the following. Even though in WSPR v0.7 you
                    are allowed to specify power levels like 28 or 35 dBm --
                    numbers that will not be in the drop-down list -- it would
                    be helpful if you start NOW to use ONLY numbers in the list.
                    The reason is that the in-between numbers WILL NOT DECODE
                    CORRECTLY in the next WSPR release. (Similarly, some new
                    message formats generated by WSPR v0.8 will be not be
                    correctly decoded by v0.7.)

                    As soon as everyone has upgraded to the new release these
                    things will no longer be confusing; but for a short time,
                    when versions 0.7 and 0.8 are both used on the air, some
                    confusion will be generated.

                    It will probably be helpful, also, to have a well advertised
                    "changeover date" after which everyone should switch to
                    using v0.8.

                    Your comments and suggestions will be appreciated.

                    With best wishes,

                    -- 73, Joe, K1JT
                  • Joe Taylor
                    To all WSPR ers: A number of us are on 80m this evening (NA time) -- dial frequency 3.5926 MHZ, upper sideband as always. Come join us for this overnight
                    Message 9 of 9 , May 1, 2008
                    • 0 Attachment
                      To all WSPR'ers:

                      A number of us are on 80m this evening (NA time) -- dial frequency
                      3.5926 MHZ, upper sideband as always. Come join us for this overnight
                      party!

                      -- 73, Joe, K1JT
                    Your message has been successfully submitted and would be delivered to recipients shortly.