2.2.3 | Minimaxi
Minimaxi is a fixed-value threshold selection method to find the optimal threshold for the root mean square error against the ideal procedure, which can be expressed as:
\(\text{\ \ \ \ \ \ \ \ \ \ \ λ}=\left\{\par \begin{matrix}\sigma(0.3936+0.1829(\frac{\text{lnN}}{ln2})),\ \ \&N>32\\ 0,\ \text{\ \ \ }\ \text{\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ N}\leq 32\\ \end{matrix}\right.\ \) (4)
Here, N is the signal length and \(\sigma\) is the signal variance.