I'm trying to implement something called the "Cascade Algorithm" which is
based upon the Dilation Equation, used when needed to approximate a
Wavelet Scaling function or mother function.

Everything is well apart from a small snag, I've been studying for a while
now and still have no idea how to implement it?!!

it goes something like this;

f(t) = c(0)f(2t) + c(1)f(2t - 1) + c(2)f(2t - 2) + c(3)f(2t - 3)

where 
c(0) = 0.6830127
c(1) = 1.1830127
c(2) = 0.3169873
c(3) = -0.1830127 (daubechie co-efficents, 4-tap)

I've tried many times on google for something that can explain this to me,
without glossing over the initial stages of "how to prime" the algorithm,
but with little luck.  I think the recursion is confusing me a tad, it
seems to say that to produce a low scale rendering of a wavelet, you need
to know what the high scale version looks like first, which is odd
considering what I've already researched seems to suggest that the low
scale rendering is the start point for the high scale rendering!?!?!


Could somebody explain it to me please, in a "cascade algorithm for
dummies" type way? I'd be very greatful (once again!)