Velocity estimation error rates using triple-PRT pulsing

Measurement of high-speed targets at large distances with a pulsed Doppler radar is restricted by the so-called Doppler dilemma. The use of a non-uniform pulsing scheme provides a means of extending the unambiguous velocity range significantly without reducing the operational radius of the radar. Possibly the most common choice is the staggered-PRT (pulse repetition time) technique where two pulse intervals of unequal length are used alternatingly. By using non-uniform pulsing schemes it is possible to increase the range of measurable velocities to cover most weather phenomena, e.g. from -50 m/s to 50 m/s, without compromising the operational radius of the radar or any of the needed (polarimetric) products. In this work, we have chosen to use a triple-PRT pulsing scheme where patterns of three unequal pulse intervals are used sequentially. This pulsing scheme was chosen over the simpler staggered-PRT method for two reasons: (1) it performs better for signals contaminated by ground clutter and (2) it is less prone to erroneous velocity estimates for wide unambiguous velocity ranges. Here we present error rates for velocity and spectrum width estimation using simulated weather signals with various spectral widths and signal-to-noise ratios. In estimating the moments, we use a unique technique based on comparing estimated autocorrelation functions with model functions.