D.I. Lekhovytskiy, D.S. Rachkov, A.V. Semeniaka, D.V. Atamanskiy, A.A. Pushkov

We investigate the peculiarities of pulse-pair-method-based (PPM) estimation of meteorological formations (MF) mean radial velocity - on phases of estimates of complex correlation (covariance) coefficients of samples of reflections from MF out of two adjacent intervals of sounding, and the PPM modifications for mode with staggering of these intervals. The accuracy of these methods is numerically evaluated for C-band pulsed Doppler weather radar (PDWR), and its dependences on the parameters of reflections from MF, the training sample size, and the average interval of sounding are determined. It is shown, that in PDWR with wavelength cm and constant interval of sounding ms ( ms) the range of radial velocity unambiguous measurement is approximately 2 (4) times less than commonly required one m/s, what is practically inadmissible. The well-known and new ways to eliminate this PPM drawback by means of staggering of pulse repetition intervals are analyzed. For such sounding mode, there are proven: - staggering order (the number of distinct from each other intervals of sounding); - the principles of sounding pulses placement on the time axis; - the algorithms for velocity measurement on the correlation (covariance) coefficients estimated by the finite-size training samples; - rational organization for estimation procedure of MF-s radial velocity. It is shown, that for pulse bursts of arbitrary size the reasonable staggering order does not exceed the value . For chosen staggering order, it is rational the maximum near (adjacent) arrangement of intervals of sounding, at which the correlation coefficients phases are evaluated, whose modified differences are used for MF-s radial velocity estimation. Among the considered well-known algorithms and their modifications, the most accurate velocity estimation algorithms are those based on differences of phases of averaged estimates of correlation coefficients of reflections samples separated by the equal time intervals. For mode with staggering of pulse repetition intervals, we prove the modification of traditional PPM, which allows to measure unambiguously MF-s velocity within the range being greater than commonly required. It is shown, that expanding the unambiguity range increases the level of velocity measurement errors, which can exceed the acceptable one (1 m/s) at average interval of sounding ms. The error level is within acceptable limits at ms, when the unambiguously measured distance is still sufficiently large (150 km). There are proven the rational recurrent procedures and schemes for MF-s mean radial velocity estimation that take into account successive on time (distance) and azimuth receipt of samples of reflections from MF in PDWR