if it does then when will a clock be useful in decoding bipolar signals
'I was thinking' it just see if it detects a -voltage & a +voltage and do "save " the bits accordingly
Network Engineering Stack Exchange is a question and answer site for network engineers. It only takes a minute to sign up.
Sign up to join this communityif it does then when will a clock be useful in decoding bipolar signals
'I was thinking' it just see if it detects a -voltage & a +voltage and do "save " the bits accordingly
All serial signals require some kind of clock - either in a separate channel or (more commonly) embedded into the signal.
The basic problem is that the receiver loses track of the transmitter's symbol stepping when there are too many identical symbols in sequence (=the line doesn't change at all for a while). Whether you use electrical, single-ended, differential, optical or wireless signaling doesn't matter.
E.g. when the sender transmits ten zeros 0000000000
, the receiver may read that back as nine zeros or eleven zeros. The actual number of bits after which a resynchronization is required varies with PLL technology, but it's finite.
A simple COM link commonly sends a fixed 0
in front of (start bit) and a fixed 1
after each user byte (stop bit).
Early Ethernet uses Manchester code which runs a clock at twice the data rate and modulates it with the user data. Fast and Gigabit Ethernet use 4b/5b and 8b/10b respectively which ensure resynchronization after 5 bits at most. 10G Ethernet uses 64b/66b, forcing resync after 66 bits.
-+
for 1 or +-
for 0. Simple but wasteful.