Evaluation of convective cloud microphysics in numerical weather predictionmodel with dual-wavelength polarimetric radar observations

<p>The representation of microphysical processes in numerical weather prediction models remains a main source of uncertainty until today. To evaluate the influence of cloud microphysics parameterizations on numerical weather prediction, a convection permitting regional weather model setup has been established using 5 different microphysics schemes of varying complexity (double-moment, spectral bin, particle property prediction (P3)). A polarimetric radar forward operator (CR-SIM) has been applied to simulate radar signals consistent with the simulated particles. The performance of the microphysics schemes is analyzed through a statistical comparison of the simulated radar signals to radar measurements on a dataset of 30 convection days.</p> <p>The observational data basis is provided by two polarimetric research radar systems in the area of Munich, Germany, at C- and Ka-band frequencies and a complementary third polarimetric C-band radar operated by the German Weather Service. By measuring at two different frequencies, the<br />dual-wavelength ratio is derived that facilitates the investigation of the particle size evolution. Polarimetric radars provide in-cloud information about hydrometeor type and asphericity by measuring, e.g., the differential reflectivity ZDR.</p> <p>Within the DFG Priority Programme 2115 PROM, we compare the simulated polarimetric and dual-wavelength radar signals with radar observations of convective clouds. Deviations are found between the schemes and observations in ice and liquid phase, related to the treatment of particle size distributions. Apart from the P3 scheme, simulated reflectivities in the ice phase are too high. Dual-wavelength signatures demonstrate issues of most schemes to correctly represent ice particle size distributions. Comparison of polarimetric radar signatures reveal issues of all schemes except the spectral bin scheme to correctly represent rain particle size distributions. The polarimetric information is further exploited by applying a hydrometeor classification algorithm to obtain dominant hydrometeor classes. By comparing the simulated and observed distribution of hydrometeors, as well as the frequency, intensity and area of high impact weather situations (e.g., hail or heavy convective precipitation), the influence of cloud microphysics on the ability to correctly predict high impact weather situations is examined.</p>

Convolutional Neural Networks for Classification of Drones Using Radars

The ability to classify drones using radar signals is a problem of great interest. In this paper, we apply convolutional neural networks (CNNs) to the Short-Time Fourier Transform (STFT) spectrograms of the simulated radar signals reflected from the drones. The drones vary in many ways that impact the STFT spectrograms, including blade length and blade rotation rates. Some of these physical parameters are captured in the Martin and Mulgrew model which was used to produce the datasets. We examine the data under X-band and W-band radar simulation scenarios and show that a CNN approach leads to an F1 score of 0.816±0.011 when trained on data with a signal-to-noise ratio (SNR) of 10 dB. The neural network which was trained on data from an X-band radar with 2 kHz pulse repetition frequency was shown to perform better than the CNN trained on the aforementioned W-band radar. It remained robust to the drone blade pitch and its performance varied directly in a linear fashion with the SNR.

Specific Radar Recognition Based on Characteristics of Emitted Radio Waveforms Using Convolutional Neural Networks

With the increasing complexity of the electromagnetic environment and continuous development of radar technology we can expect a large number of modern radars using agile waveforms to appear on the battlefield in the near future. Effectively identifying these radar signals in electronic warfare systems only by relying on traditional recognition models poses a serious challenge. In response to the above problem, this paper proposes a recognition method of emitted radar signals with agile waveforms based on the convolutional neural network (CNN). These signals are measured in the electronic recognition receivers and processed into digital data, after which they undergo recognition. The implementation of this system is presented in a simulation environment with the help of a signal generator that has the ability to make changes in signal signatures earlier recognized and written in the emitter database. This article contains a description of the software’s components, learning subsystem and signal generator. The problem of teaching neural networks with the use of the graphics processing units and the way of choosing the learning coefficients are also outlined. The correctness of the CNN operation was tested using a simulation environment that verified the operation’s effectiveness in a noisy environment and in conditions where many radar signals that interfere with each other are present. The effectiveness results of the applied solutions and the possibilities of developing the method of learning and processing algorithms are presented by means of tables and appropriate figures. The experimental results demonstrate that the proposed method can effectively solve the problem of recognizing raw radar signals with agile time waveforms, and achieve correct probability of recognition at the level of 92–99%.