The Improved Multi-Innovation Parameter Estimation Algorithms for the Sine Signal Modeling with the Single-Frequency

Abstract

The problem of modellig for sine signals with a single-frequency is addressed in this paper. In term of the amplitude, the phase and the frequency of the sine signal and the nonlinear relation between the model output and the characteristic parameters, the parameter estimation methods are presented by means of the discrete measured data. In order to trace the information variation with the time-varying, the shift window data are used to develop a novel algorithm to achieve parameter estimates for the single-frequency sine signal. By using these data, the multi-innovation stochastic gradient method and an improved multi-innovation stochastic gradient method are devised for estimating the signal parameters. At last, an example is given for the purpose of testing and analyzing the performance of our propsed methods