Adaptive time-frequency signal analysis and its case study in biomedical ecgwaveform analysis

Traditional time-frequency (TF) signal representations are not appropriate for parametric analysis of most real world signals. In this study, we describe adaptive time-frequency distribution (TFD) as a robust approach to TF signal decomposition for non-stationary signal processing. This approach has not been utilized for TF signal analysis other than feature extraction. We introduce T wave alternans (TWA) analysis as a new application for this approach. In order to test the robustness of the proposed technique, we challenge it under conditions of non-stationary dynamics that are expected with real world TWA. The results of the numerical simulation support the effectiveness of this approach for TWA estimation.