Heart rhythm is extrinsically modulated by the autonomic nervous system and recently, the Tone-Entropy (TE) measurement was reported as a measure of autonomic balance and activity in time domain HRV analysis. Current algorithm for T-E measurement describes only beat-to-beat or influence of a heart beat on a train of succeeding beats on a single scale. Therefore, conventional T-E analysis has often not been able to discern various physiological conditions using heart rate variability (HRV) signal. In this study, we will present a mathematical framework to define multi-scale T-E analysis, apply this in differentiating physiological and synthetic RR time series. Finally, we compare the performance of proposed parameters with conventional T-E measurements.