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

72291Re: Softrock Dynamic Range and MDS Redux

Expand Messages
  • warrenallgyer
    Feb 17, 2013
    • 0 Attachment
      Hi Nick

      It all makes perfect sense if I can accept the lack of logic in having readable signals below the noise floor.

      However the ARRL Handbook says this in Section 25.5.1:

      "Several methods are used to determine receiver sensitivity. The modulation mode often determines the best choice. One of the most common sensitivity measurements is minimum discernible signal (MDS) or noise floor, which is suitable for CW and SSB re- ceivers. The minimum discernible signal is defined as that which will produce the same audio-output power as the internally gener- ated receiver noise. Hence, the term "noise floor." "

      So I guess we are saying the Handbook is just flat out wrong on this point in equating MDS to noise floor?

      Warren Allgyer - W8TOD


      --- In softrock40@yahoogroups.com, "g3vnc" <g3vnc@...> wrote:
      >
      > Warren:
      >
      > --- In softrock40@yahoogroups.com, "warrenallgyer" <allgyer@> wrote:
      >
      > > I am with you so far and no argument. So now I need to extend that to calculate dynamic range. Would you agree dynamic range equals Overload Signal Level - MDS + 3 dB for a given resolution bandwidth
      > >
      >
      > Hmm. It all depends how you define Dynamic Range and Overload Signal Level!
      >
      > The ARRL Handbook 2011 para 12.4.3 discusses different measures of dynamic range.
      >
      > Another good reference is Wes Hayward's "Solid State Design for the Radio Amateur" page 113.
      >
      > Try and get hold of those or similar. Lots of stuff on the internet about this.
      >
      > To me the most useful thing to measure is 2 tone Dynamic Range, defined by Hayward as
      >
      > Dynamic Range = 2(Pi-MDS)/3
      >
      > where Pi is the input intercept. Note that MDS is a negative number.
      >
      > To measure Pi you need two signal generators, a hybrid combiner and a spectrum analyser (or an SDR).
      >
      > If Pi is 0dBm, and MDS is -150dBm, then DR is 100dB.
      > If Pi is -10dBm, and MDS is -150dBm, then DR is 93dB.
      >
      > But you must specify the bandwidth.
      >
      > > If so then I am still troubled by the fact I can rather easily copy CW at levels lower than MDS
      >
      > Don't be. It is possible to copy CW well below the MDS. The brain is a wonderful thing. Mine's full of filters!
      >
      > But this is a subjective effect. Some people are better at it.
      >
      > To compare results with others you've got to measure something objectively. Hence MDS.
      >
      > >
      > > Finally, what is the "noise floor"? The ARRL says MDS = Noise Floor.
      >
      > The standard definition of MDS is that signal level which causes a 3dB increase in the output power in a given resolution bandwidth.
      >
      > 3dB is twice the power, so half the output is the noise power and the other half is the signal power. So if the input signal at MDS is -150dBm and the noise power is -150dBm in 500Hz, then the total output power is -147dBm.
      >
      > So if you define Noise Floor as the noise power in the resolution bandwidth at a given frequency then in my example MDS = Noise Floor = -150dBm.
      >
      > HTH
      >
      > Nick G3VNC
      >
    • Show all 39 messages in this topic