An Extended Frequency Splitting Method for Ultrasound Color Flow Imaging

Signal-to-noise ratio (SNR) can be improved by chirp excitation signal in ultrasound color flow imaging. However, the performance of autocorrelation estimator, which relies on a narrowband approximation, will be degraded by the broadband characteristics of compressed chirp signals. Frequency splitting (FS) method has been proposed to deal with this problem, where broadband signal is divided into several narrow bands, and then the velocity of each subband is estimated and finally averaged to have improved estimation. Nevertheless, Nyquist limit of each subband is different and velocity range will be reduced by the lowest limit. This paper proposes an extended frequency splitting method, which can detect alias in high frequency subbands and also theoretically extend the final Nyquist limit by 3.3 times.