Kai Klaas wrote:
-------------------------------
Curves which differ not much from straight line can be well approximated by a n-th order power-series polynomial:

"y = a0 + a1 * x + a2 * (x**2) + ... + an * (x**n)
n-th order polynomial is fully determined by these n+1 parameters a0, a1, ..., an."


But there is no deterministic way to find out the 'n'. is any? infact Mr. Taylor was right stating its a deadly techneque. it totaly depends on the curve how near we can find wiestrass polynomial.


for example say one collects the data for a transistor in the active region between Ib and Ic (Vce fixed) a good curve fit will result in an straight line and its almost error free fit but as we all know, above the active region in the saturation region its totaly differant curve ending with large value of error.

more there is no way you cant test your result because for that you have to have values on the undefined points. and if you have those values you dont have to extrapolate.


abhishek