On the influence of segment length in PW-tsm Doppler systems

Two different signal processing techniques can be used for blood velocity measurements with Pulsed Wave (PW) Doppler. These are based on measurement of phase shift and time shift, respectively. The first technique is based on a correlation measurement scheme, which features inherent velocity aliasing. The second technique, which is based on the location of the peak in the cross-correlation function of consecutive received signal segments, is in principle free of this problem. However, as noise are present and finite signal segments are used, only an estimate of the cross-correlation function is available. Thus, as detection of velocities greater than the aliasing velocity will require searching for a peak in a function with several peaks, the method is prone to detection of incorrect peaks. This paper investigates the effect of the length of the signal segments to be cross-correlated on the probability of incorrect detection.