An error distribution based nonlinear companding method for analog behavioral modeling via wavelet approximation

In this paper, we develop a wavelet collocation method with nonlinear auto-companding for behavioral modeling of analog circuits. The companding function is automatically constructed according to the initial error distribution obtained through approximating the input-output function of the circuit blocks by conventional wavelet collocation method. Compared with the published modeling approaches, the proposed nonlinear auto-companding method works more efficiently not only in controlling the error distribution but also in reducing the modeling errors. A switched-current filter example is employed to demonstrate the promising features of the proposed method.