Improvement on accuracy of adaptive frequency estimation by direct sampling control

摘要

As one of the applications for digital signal processing based on the time domain control, we have considered an adaptive frequency estimation algorithm for sinusoidal signals in additive noises. The proposed algorithms were implemented by the oversampling processing to control the time domain of the signal. Therefore, there is a problem that it has additions. This paper introduces an adaptive frequency estimation algorithm with a small amount of calculation, where the sampling frequency is directly controlled. This method using an FIR notch filter is based on a gradient algorithm to make the output signal zero. To improve the estimations accuracy we use an IIR bandpass filter, which can reduce the noise components. Simulations are presented to make clear the performance of the proposed method, and it leads that it can estimate the input frequency accurately and the computational complexity is very low.