Application of Genetic Algorithms in Parameter Estimation of Periodic Signals

Abstract

A genetic algorithm is used to estimate the parameters of sine and multiharmonic acquired waveforms. This new approach uses genetic algorithms (GA) to find the fundamental frequency of the signal. The signal frequency is then used in a multiple linear regression algorithm to determine the amplitudes and phases of each harmonic, as well as the signal DC offset. This procedure takes advantage of the robustness of genetic algorithms and overcomes the converge problems of traditional least-squares minimization procedures.