<p>The decomposition of a multicomponent non-stationary signal is helpful in obtaining its time-frequency distribution (TFD). In this paper, a novel empirical mode decomposition (EMD) like eigenvalue decomposition of Hankel matrix (EVDHM) technique is proposed, which extracts the mono-component signal iteratively. In each iteration, EVDHM is performed on the residue signal to obtain elementary components (ECs). Later, the most dominant mono-component is extracted from the obtained ECs using a novel parameter proposed namely, envelope of the normalized instantaneous amplitude modulated bandwidth. The separability conditions for the two sinusoidal multicomponent signal are studied for the proposed decomposition technique using the average correlation measure. Finally, the proposed EMD-like EVDHM's performance for signal decomposition is compared with iterative EVDHM and EMD technique for synthetic and real-world speech signal. Signal to reconstruction error (SRE) ratio is used as a performance measure and the proposed technique is found to be giving highest SRE value for all the considered synthetic signals when compared with the baselines. Further, the proposed technique is also found to be providing highest-resolution TFD for speech signal among all the techniques studied in this paper.</p>