Heart rate time series (HRTS) signal has been studied using Auto-Regressive (AR) based power spectrum. Twenty-five controls and twenty-five MI patients were taken for this study. From HRTS signal, 480 epochs of 32 seconds overlapping by a 31 seconds are extracted. Power spectrum is computed epoch-by-epoch and frequency domain HRTS signal parameters are derived. These parameters show the discrimination between the controls and MI patients. The statistical results are shown in the table below. From this study it can be concluded that the analysis of HRTS signal may lead to detection of Ml in advance.