Improved Clutter Suppression Algorithm for Atmospheric Target Detection Using Polarimetric Doppler Radar

When using dual-orthogonal frequency modulation continuous wave (FMCW) polarimetric radar, the polarimetric range-Doppler spectra of the atmospheric targets are contaminated by noise, clutter, and artificial signals that disturb the analysis of atmospheric target parameters, such as reflectivity, mean Doppler velocity, and Doppler width. This paper describes an improved clutter suppression method that is able to detect most atmospheric targets while suppressing clutter, noise, and artificial signals as well as possible. By using logical decisions that are based on the spectral linear depolarization ratios (SLDRs), a binary mask matrix is constructed, and the characteristics of this original mask matrix are improved via mathematical morphology methods. To suppress clutter, noise, and artificial signals and retain atmospheric targets, we multiply the range-Doppler spectra by the improved mask matrix. The raw atmospheric data are acquired using the Polarimetric Agile Radar S-And X-band (PARSAX) radar, which is able to measure the backscattering matrix of atmospheric targets in one sweep. After calibration, the range-Doppler polarimetric spectra are processed using the proposed clutter suppression method, and the atmospheric targets are successfully detected. Compared with the unprocessed data and the data processed by noise clipping and double SLDR filtering, the data processed by the proposed method show improvement in the ability to determine the atmospheric targets in the range-Doppler spectrogram, the reflectivity, the mean Doppler velocity, and the Doppler width.