XPost: comp.dsp, sci.crypt, sci.electronics.design   
   From: eric.jacobsen@ieee.org   
      
   On 12/20/2009 3:42 PM, Steve Pope wrote:   
   > Eric Jacobsen  wrote:   
   >   
   >> On 12/20/2009 10:32 AM, Steve Pope wrote:   
   >   
   >>> On Dec 20, 2:36 am, Eric Jacobsen   wrote:   
   >   
   >>>> The point was really that even from an advanced FEC standpoint an input   
   >>>> BER of 1 in 10 isn't practical to work with for the described   
   >>>> application.   Yielding half the bandwidth to FEC overhead is actually   
   >>>> practical, and using R = 1/2 coding over satellite channels is quite   
   >>>> common.   Using something like an R = 1/6 capacity-approaching code to   
   >>>> be able to handle such low input error rates is, I think, not practical.   
   >   
   >>> Last I checked, such a channel is within the operating range of a rate   
   >>> 1/3 binary convolutional code...   
   >   
   >> BICO capacity for R = 1/2 is at about 0.177dB Eb/No, and for R = 1/3   
   >> it's about -0.357dB Eb/No.   
   >   
   > Agree   
   >   
   >> Not a lot of difference there from a   
   >> capacity perspective.   
   >   
   > Correct; as one gets into lower code rates, going to even lower rates   
   > tends to give you no additional normalized coding gain (i.e. the   
   > additional coding performs no better than a repetition code).   
   >   
   >> For uncoded, i.e., raw BER of 1e-1 happens at about -1dB Eb/No   
   >   
   > Actually it's about -1.7 dB, but let's say it's -1 for the   
   > sake of argument.   
   >   
   > (or about   
   > -4dB (equivalent raw Eb/No) at R = 1/2, and -5.77dB (equivalent raw   
   > Eb/No) for R = 1/3.   
   >   
   > I'm not sure what this sentence means.   
   >   
   > A rate 1/2 coded system operating at an Eb/No of +2 dB has the   
   > same raw BER as an uncoded system operating at an Eb/No of -1 dB.   
   >   
   > A rate 1/3 coded system operating at an Eb/No of +3.77 dB has   
   > the same raw BER as an uncoded system operating at -1 dB.   
   >   
   > (Unless I'm confused, which has happened before...)   
      
   Doh!  I think I went the wrong way with the 3db and 4.77dB differences.   
     I get stuff like that backwards all the time.   
      
   >> So I'm skeptical.  In a practical system, especially a practical   
   >> satellite system, I think it'd be very difficult to operate with an   
   >> input BER of 1e-1.   
   >   
   > I'm not too skeptical.  I would posit that GSM phones in their   
   > basic 2G mode operate under conditions this bad, and 802.11 systems   
   > at 1 mbps might also.   
   >   
   > Steve   
      
   I'm less skeptical now.  ;)   
   --   
   Eric Jacobsen   
   Minister of Algorithms   
   Abineau Communications   
   http://www.abineau.com   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)