Re[2]: [WM]: Normalized Correlation

["Yuri V. Klimets" <klimets@uni-wuppertal.de>]

JOR> To make m-sequences orthogonal the missing all zero
JOR> state needs to be included. This is not hard to do in software but 
JOR> in hardware a shift register would lock in an all zero state.
     ^^^^^^^^^^^^^^^^^^^^^^^^^^

     There is pretty simple hardware implementation of LFSR with all
     states - you just have to add the wide N-OR gate that has all
     inputs connected to cells of shift register (the MSB is
     connected through the inverter) and output connected
     to the XOR gate in LFSR. Here is the sketch:
                    _____
          /-------o|     |
         / /-------|  OR |
        / / /------|     o---\
       / / / /-----|     |   |
      / / / / /----|_____|   |
      | | | | |              |
     -----------     /-------/
     |1|0|0|0|0|<-------------------\
     -----------     \     _____    |
      |     |         \___|     |   |
       \     \____________| XOR |---/
        \_________________|     |
                          |_____|


This scheme will insert 00000 state after 10000 state preserving all
others states in the sequence.
                          
                          
Yours sincerely,

                 Dr. Yuri Klimets




______________________________________________________________________________

Watermarking Mailing List - http://www.watermarkingworld.org/ml.html
To unsubscribe send email to "majordomo@watermarkingworld.org" with
"unsubscribe watermarking YOURMAIL" in the body.
______________________________________________________________________________