Quotient Sync Filter Better Than Lock In

Aug 01, 2009 42 Replies

See the way I look at it is, if you use summation convergence of its simple root along two processing line, then isn't the signal sampling needed reduce by the square factor thereof?, and since there are four edges, you can reduce the sampling by a further 4 times. or have the same sampling time, and achieve an extremely high signal definition.

and since any transition time in the clocking signal =3D 0 ; the result of instantiating two processing lines into 1 coherent perfect (limit "iff" (sync_difference >transition_max) is true) instantaneous transition square wave.

do you see what I mean Bret?

this may be the case @the cost of a few extra processing protocols

The error squared term is probably what kills the quotient sync filter for high noise levels.

At noise levels < the signal it should work.

After rectification or multipling both signals by the ref, the denominator and maybe the numerator can be smoothed just enough to always be a certain amount above zero.

After division the quotient of the signal will be mostly DC. Little more than the noise needs to be smoothed after that.

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required