The problem of precipitation detection using Frequency Modulated Continuous Wave (FMCW) radar under strong sectorial interference is addressed. The effect of such strong interferences in the case of an FMCW scanning radar is presented. Three signal-processing pipelines (two reflectivity-based and one Doppler-based) are proposed. The performances of all these pipelines are analyzed and compared. The morphology-based pipeline performs better for higher signal-to-noise ratios (> −15dB), whereas the entropy-based pipeline performs better in the case of lower SNRs (< −15dB). On the other hand, the circular variance-based masking technique is computationally very efficient. The proposed techniques are applied to simulated and real X-band fast-scanning radar data.

@en

The problem of the limited accuracy of precipitation Doppler spectrum moments estimation measured by fast azimuthally scanning weather radars is addressed. A novel approach for the Doppler moment estimation based on maximum likelihood estimation is proposed. A simplified semianalytical parametric model for the precipitation power spectral density (PSD) as a function of the velocity parameters of the scatterers and the finite radar observation time is derived for typical precipitation-like weather conditions. An inverse problem for estimating the Doppler moments from measurements of the PSD is formulated and solved. It is demonstrated that the variance of the estimation of the Doppler moments approaches the Cramer Rao Lower Bound (CRB) when the observation time approaches infinity. The performance of the proposed approach is compared with some classical techniques and another realization of the maximum likelihood approach based on simulated and experimental data. The results indicate the superiority of the proposed approach, especially for short observation time. Furthermore, a scanning strategy to accurately estimate the Doppler moments based on the true velocity dispersion of the scatterers is provided with the help of the proposed approach.@en

The challenge of processing incoherent sets of radar echoes to estimate the Doppler moments and noise standard deviation from precipitation with a fast-scanning radar is addressed. The recently proposed maximum likelihood parametric Doppler spectrum estimator (PSE) is extended to accommodate the noise standard deviation along with the Doppler moments in the parametric model of the Doppler power spectral density (PSD). Its performance in estimating the Doppler moments and the noise standard deviation is compared with another maximum likelihood approach. The proposed approach has a smaller bias in the estimation of the noise standard deviation for a wide range of spectral widths on simulated data. The algorithm is verified on experimental data from a fast-scanning weather radar. @en

The challenge of avoiding aliasing in the Doppler spectrum for precipitation is addressed. A novel integrative signal processing approach has been proposed to address the research gaps from various disciplines. The proposed approach consists of several steps. First, an aperiodic way of sampling the echoes (aperiodic sampling refers to aperiodic pulse train in the context of radar echoes in slow time) has been proposed by which the maximum unambiguous Doppler frequency (velocity) is enhanced. Second, the Doppler spectrum moment estimation is performed with the help of a parametric form of its covariance. The performance of the moment estimation is assessed by the bias and the variance in the estimated counterparts. The theoretical variance for the parameter estimation is also derived. An aperiodic pulse train design recommendation has been proposed for adequately and unambiguously estimating the Doppler moments for one extended target (like precipitation). Finally, a spectrum reconstruction technique is implemented after the moment estimation on simulated radar echo samples for a realistic precipitation-like event. The comparison with the other approaches proves its superiority for parameter estimation and Doppler spectrum reconstruction.

@en

The problem of aliasing in precipitation Doppler spectrum with uniform pulse repetition time is addressed. This work focuses on de-aliasing such Doppler spectra using non-uniform sampling techniques, namely, Log-periodic and Periodic non-uniform sampling. These techniques reduce the ambiguous main lobes caused by aliasing (by going beyond the observable frequency limit) into ambiguous sidelobes that are distinguishable from the original spectra. The SNR is further enhanced by using an Iterative Adaptive Approach (IAA) algorithm. The performance of Doppler moment estimation is presented after applying the IAA algorithm on simulated precipitation-like radar echoes. However, the ambiguous sidelobe suppression is highly dependent on the spectral width of the Doppler spectra.

@en

The theory of radiation from an open-ended circular crosssection waveguide is extended by including the excitation of all possible higher-order waveguide modes at the waveguide-to- free-space boundary. The theoretical expressions are formulated using spectral-domain techniques. The validation is performed using the existing commercial tool FEKO with the Method of Moments (MoM) solver for a cylindrical waveguide with the fundamental mode excitation (TE11) and the corresponding higher-order mode, which has the same azimuthal variation as of the fundamental mode, which is (TE12). @en

The challenge of reconstructing the Doppler spectrum of a precipitation-like event observed by a fast-scanning weather radar is addressed. A novel method is proposed where the echo sequence in time is assumed to be a complex Gaussian process with a known covariance structure. It is a two-step approach where the first step is the estimation of the hyperparameters of the covariance function with a maximum likelihood approach, and the second step is the reconstruction of the spectrum directly in the time or spectral domain. The proposed approach is applied to simulated data for hyper-parameter estimation performance analysis and real radar data for the complete Doppler spectrum reconstruction. @en

At the airports, fast-scanning phased array radars are used to detect point-like targets, e.g., birds and drones. However, these radars do not possess Doppler processing and thus are not used for observation of the weather targets like precipitation events. A series of steps need to be performed, such as beamforming in elevation, target masking for weather-like targets, and Doppler processing to complete the signal processing chain for this kind of radar for weather-like targets. This paper focuses only on beamforming in elevation which is the first step in the direction of 3D wind field estimation for a fast-scanning phased array radar. The reiterative Minimum Mean Square Error (rMMSE) beamforming technique is proposed in this paper and has been used on real radar observations. The performance is analysed for the first time when noise is under or overestimated. The rMMSE beamformer has been compared with the Fourier (FR) and Capon (CP) beamformers.@en

The performance and limitations of the Doppler processing of the scattered signals from extended meteorological objects (precipitation) are analysed in the case of radar with fast azimuthal scanning. The classical method of the Discrete Fourier Transform (DFT) has been applied to simulated weather radar signals to estimate the Doppler velocity spectrum and characterise it with the mean Doppler velocity and the Doppler spectrum width. The accuracy and resolution of these estimations have been analysed as a function of the scanning radar rotation speed. Finally, the performances of the 2D wind field retrieval are analysed in relation to the accuracy and resolution of Doppler spectra estimations. The wind field retrieval has been done using the classical velocity azimuthal display (VAD) retrieval technique that gives an overall/average estimate of the wind field over an observation region. A few possible approaches for improving the accuracy and resolution of a fast scanning weather radar Doppler signal processing are proposed and analysed based on simulated scanning radar data.@en