Computationally effective algorithm for robust weighted averaging

One of the greatest disadvantages of the weighted signal averaging method is its sensitivity to the presence of noise and outliers in data and the need to estimate the noise variance in all signal cycles. The robust weighted averaging method based on the (epsilon)-insensitive loss function is free of these disadvantages, but has a very high computational burden and requires a choice of the insensitivity parameter (epsilon). In this study, a new computationally effective algorithm for robust weighted averaging with automatic adjustment of the insensitivity parameter is introduced.