Modified Constrained Notch Fourier Transform (MCNFT) for Sinusoidal Signals in Noise and Its Performance

Adaptive Fourier analysis of sinusoidal signals in noise is of essential importance in many engineering fields. So far, many adaptive algorithms have been developed for this purpose. In particular, a filter bank based algorithm called constrained notch Fourier transform of its cost-efficiency and easily controllable performance. However, its performance deteriorates when the signal frequencies are not uniformly spaced. This paper proposes, at first, a new structure for the CNFT, referred to as modified CNFT (MCNFT), to compensate the performance degeneration of the CNFT for noisy sinusoidal signals with non-uniformly spaced frequencies. Next, a detailed performance analysis for the MCNFT is conducted. Closed form expression of steady-state mean square error (MSE) for the discrete Fourier coefficients (DFCs) is derived. Extensive simulations are presented to demonstrate the improved performance of the MCNFT and the validity of the analytical results.