The aim of this article is to create a novel filtering algorithm for the emitted electromagnetic signals of corona discharges, using the special time-domain features of the continuous wavelet transform. The first part of this article therefore focuses on the time-domain examination of these signals to find characteristic features that are compatible with the wavelet transform. The second part of this article utilizes these features to create and validate the filtering algorithm. This is an important step toward creating an accurate three-dimensional partial discharge location approximation system that could contribute to preventing and finding partial-discharge-related faults. In the presented research, needle-plane corona discharge sources with different geometries are tested at both ac and dc voltages.