A simple non-linear shift-variant process/filter for reducing gaussian noise in impact signals

This paper presents a non-linear shift-variant process/filter that can be used on non-repeatable, time variant impact response signals to reduce the Gaussian noise introduced by the instrumentation. The process can improve the quality of the data in terms of both the time and frequency domains. The purpose of the paper is to show that, just as operations in the frequency domain can improve the quality of the signal in the time domain, this process's operations in the time domain may improve, if the noise is Gaussian, the quality of the signal in the frequency domain. The improvements in the signal quality in many cases can increase the accuracy of transfer function analysis.