On Fri, 16 Jul 2004, Shawn Yates wrote:

> I use manchster for everything I do.  Here is the general idea I used
> for my receiver.

<snip>

Shawn,

Your approach works well for polling. All you need to know is when the bit
stream begins, how wide a typical bit is, and how to recover if you sample
two halves in the same state. A little more challenging is an interrupt
driven Manchester Decoder. And even more challenging is one that can
dynamically adapt to the varying frequencies that one often encounters
with (cheap) RFID devices. Even more challenging is to get the Manchester
Decoder overhead below 2% of the total processing capability of the CPU.
You may be wondering how I know this is so challenging :). I'm not at
liberty to share the code, but I think Phil's suggestion of drawing the
Manchester Waveform and following all of the transitions is a good one.
You should be able to design a simple state machine capable of recognizing
all of the state transitions.

Scott

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics