Introduction: Studies show that changes in the average of sleep duration has unpleasant effects on physiological and cognitive functions of individuals. On the other hand, recording and analysis of biological signals are one of the safest methods for studying emotional responses, that participant is rarely able to fake them voluntarily. Given that the effects of sleep on the physiological parameters of individuals in response to emotion is not considered sufficiently, this study has attempted to provide a comprehensive evaluation on signal analysis methods, to examine the differences in autonomic parameters, including the electrocardiogram and galvanic skin response (GSR). Materials and Methods: Autonomic signals of 47 students while watching the happy and sad images were collected and analyzed. For this purpose, a wide range of linear and nonlinear characteristics were extracted from the signals and the differences between the two groups were evaluated by means of the non-parametric Mann–Whitney test. Results: 22 of the participants, had normal sleep (mean age 21.32 ± 1.67 years) while 25 subjects had insufficient sleep (mean age 21.88 ± 1.64 years). The results of present study show that physiological responses to emotional stimuli depend on the duration of sleep. The highest percentage of discrimination devote to the non-linear characteristics of GSR signal in response to the sad images. Of these, the Poincare indices revealed significant differences between the two groups. The results of linear analysis also indicated better performance of GSR signal in illustrating the difference between emotional responses of two groups with different amount of sleep. Conclusion: There is a significant relation between the amount of sleep and emotional behavior of the participants. Analysis of autonomic parameters enable us to investigate this relationship. However, the results are depended on the type of stimulation and the methodology applied for data analysis.