Threshold analysis in wavelet-based denoising

The hard threshold t = cσ is efficient in wavelet threshold-based nonlinear filtering. In general, the optimal constant c would vary with the signal and the added noise. The nearly optimal choice of c by minimising F(c), which is an equivalent function to the mean square error (MSE) of the recovered signal, is discussed. Experiment shows that F(c) is consistent with the MSE.