The purpose of this study is to investigate the potential of the ensemble empirical mode decomposition (EEMD) to extract cardiogenic oscillations from inductive plethysmography signals in order to measure cardiac stroke volume. First, a simple cardio-respiratory model is used to simulate cardiac, respiratory, and cardio-respiratory signals. Second, application of empirical mode decomposition (EMD) to simulated cardio-respiratory signals demonstrates that the mode mixing phenomenon affects the extraction performance and hence also the cardiac stroke volume measurement. Stroke volume is measured as the amplitude of extracted cardiogenic oscillations, and it is compared to the stroke volume of simulated cardiac activity. Finally, we show that the EEMD leads to mode mixing removal.