Parameter estimation algorithm for the exponential signal by the enhanced DFT approach

Abstract

Based on enhanced interpolation DFT, a novel parameter estimation algorithm for the exponential signal is presented. The proposed two-step solution consists of a preprocessing unit which constructs a new signal sequence by continuously cycle shifting sample points and summing up N buffered exponential signal sample sequences, then an interpolation DFT engine to obtain accurate parameter estimation of the exponential signal based on the signal sequence generated by the preprocessing unit. Compared to previous works, owing to the combination of the pretreatment and the interpolation DFT analysis, the tedious iteration for spectral leakage elimination can be removed, and remarkable improvements are achieved in terms of estimation accuracy and speed. The simulation results verify the effectiveness of the proposed algorithm.