Abstract: In this paper, based on the feature of ensemble empirical mode decomposition(EEMD) which is very suitable for analyzing non-stationary and nonlinear signals, the pulse rate variability (PRV) signal is decomposed into multiple intrinsic mode function (IMF) components by EEMD, to calculate the energy of each IMF component respectively, and high-frequency energy(EHF), low-frequency energy(ELF) and very low-frequency energy(EVLF) of PRV were reconstructed from the energy of the IMF components according to the characteristics of each IMF component spectrum. This method is compared with the traditional AR power spectrum estimation method through the experiment. The results show that the correlation coefficient of each frequency domain parameter corresponding to PRV obtained by the two methods is higher than 0.96, which indicates that the method can truly reflect the sympathetic nerve and vagus nerve activity.