Identification of human individuals within a group of 39 persons using micro-Doppler (μ-D) features has been investigated. Deep convolutional neural networks with two different training procedures have been used to perform classification. Visualization of the inner network layers revealed the sections of the input image most relevant when determining the class label of the target. A convolutional block attention module is added to provide a weighted feature vector in the channel and feature dimension, highlighting the relevant μ-D feature-filled areas in the image and improving classification performance.

A simple electromagnetic model of wind-turbine's main structural elements as the linear wired structures is developed to simulate the temporal patterns of observed radar return Doppler spectra (micro-Doppler). Using the model, the micro-Doppler for different combinations of the turbines rotation frequency, radar pulse repetition frequency, and duration of the Doppler measurement interval are analyzed. The model is validated using the PARSAX radar experimental data. The model ability to reproduce the observed Doppler spectra main features can be used for development of signal-processing algorithms to suppress the wind-turbines clutter in modern Doppler radars.