A Digital Filter for the State Estimation of Impulsive Noise Under the Existence of a Background Noise and its Experiment in the Work Environment.

In the actual work environment, a specific objective noise, like machine and industrial noises, shows impulsive fluctuation form. Furthermore, it often occurs that a specific noise is partially or completely contaminated by background noise. In that case, the fluctuation wave form of the specific noise has to be estimated moment by moment in order to evaluate and/or improve the work environment. In this study, a digital filter for estimating an impulsively fluctuating specific noise is proposed. More specifically, after introducing a generalized time series regressive model of the specific noise, a method for estimating the impulsive noise based on an observation contaminated by the background noise is theoretically derived. Furthermore, the proposed method is applied to the actual task of estimating an industrial impact noise.