k-Sigma transform make signal ISO-Independent???

according to Eqn. (9) in the paper, variance over f(x) is x*/k+sigma/(k*k) = u*+sigma/(k*k) .
it seems this variance dependents on k and sigma or sigma/(k*k).
obviously k and sigma both are dependent on ISO.
for sigma/(k*k), because k = ga, sigma = g*g*sigmaD+sigmaR,
so sigma/(k*k) = sigmaD/(a*a) + sigmaR/(a*a*g*g), which is also dependent on g, aka. ISO.

i also tried to calculate sigma/(k*k) over different iso value, using "class KSigma" in the source code with estimated parameters provided in the code: K_coeff=[0.0005995267, 0.00868861],
B_coeff=[7.11772e-7, 6.514934e-4, 0.11492713],
and find it is not Constant.

so How to understand k-Sigma transform can make signal ISO-Independent?

Start with Eq. (9) in the cited paper and the `KSigma` class in the source, using the K and B coefficients listed in the issue. Reproduce the calculation of sigma/(k*k) across ISO values, then compare the implementation and assumptions with the paper. Done means explaining the apparent ISO dependence and identifying whether the claim, parameters, or implementation needs correction.

Written by the indexing model from the issue text.