Parameter estimation algorithms utilizing implicit phase

In a previous paper, it has been shown that time domain signal parameters such as energy center of mass, time constant, and time delay can be estimated from the FFT of the signal. The estimation techniques use phase in an implicit manner, hence the name "implicit phase". The lack of an inverse FFT can lead to a 50 percent reduction in computation time. In the present paper we minimize the error variance of the implicit phase algorithms for parameter estimation by weighting the waveform DFT before putting it through the algorithms. Two techniques for designing the weighting sequences are described. One method assigns values of one or zero to the weights. The other method finds the best real valued weights.