Our results are consistent with previous findings. It has shown that the similarity measures of HRV signals were affected by the psychological state. The results showed that the lowest corrEn values were achieved for the HRV signals of pKYM and dKYM states. The lower corrEn might be attributed to the lower stress levels and lower sympathetic reactions in the dKYM. In contrast, the highest corrEn has belonged to the pCCM and SNB. The highest CSD was attained for pKYM and the lowest was obtained for the SNB. The fewer variations of the similarity measures were perceived across the lower kernel sizes.

The critical role of parameter selection in classification performances has been also shown. Selection of both kernel size and sigma parameters affected the classification accuracy. In addition, our results showed the benefit of feature level fusion strategies in HRV classification. Previously, the role of feature level fusion in increasing the performance of the classification of psychological data has been shown (Goshvarpour et al., 2017a; 2017b).

By combining both similarity indices derived from corrEn and CSD, valuable information of HRV signals is achieved which led to superior classification results. The CSD was unable to show the nonlinear similarity of the signals, which is accessible with corrEn. Consequently, the information of these two features is complimentary. Although the high classification performance (100%) was achieved by each feature, division, and weighted sum fusion strategies obtained a higher number of maximum accuracies using more k and ? parameters. Totally, the best value for k and ? parameters is 1 and 0.05, respectively.

The classification of HRV signals during mediation has not been documented sufficiently. Previously, a system was proposed to classify HRV signals of Samadhi and non-Samadhi groups based on time and frequency based features (Phongsuphap and Pongsupap, 2011). The accuracy of 94.8% was attained using Fisher discriminant analysis. In another study, Lyapunov exponent and entropy were calculated to characterize HRV dynamics during CCM and pCCM (Goshvarpour et al., 2012). Three machine learning systems were evaluated including Quadratic, fisher, and k nearest neighbor. Quadratic classifier achieved the maximum accuracy of 92.31%.

To sum up, the results of this study established that new combined similarity index of HRV signals could be served as an appropriate quantity to correctly discriminate different psychological states.