Velocity is one of the main navigation parameters of moving objects. However some systems of position estimation using radio wave measurements cannot provide velocity data due to limitation of their performance. In this paper a velocity measurement method for the DS-CDMA radio navigation system is proposed, which does not require full synchronization of reference stations carrier frequencies. The article presents basics of FDOA (frequency difference of arrival) velocity measurements together with application of this method to an experimental radio navigation system called AEGIR and with some suggestions about the possibility to implement such FDOA measurements in other kinds of asynchronous DS-CDMA radio networks. The main part of this paper present results of performance evaluation of the proposed method, based on laboratory measurements

In the paper, the variation of the intensity of the geomagnetic field force is analysed in time and space. For the research, the data from measurements of the intensity of the geomagnetic field force at four airports (Kaunas, Klaip˙eda, Palanga andVilnius) and 6 geomagnetic field repeat stations aswell as the data from Belsk Magnetometric Observatory (Poland) were used. For the data analysis, the theory of covariance functions was applied. The estimates of the cross-covariance functions of the measured intensity of the geomagnetic field force or the estimates of auto-covariance functions of single data were calculated according to the random functions created from the force intensity measurement data arrays. The estimates of covariance functions were calculated upon varying the quantization interval on the time scale and applying the software created using Matlab package of procedures. The impact of radars of airports on the intensity of geomagnetic field variation and on changes of their covariance functions was established.

In recent years, deep learning and especially deep neural networks (DNN) have obtained amazing performance on a variety of problems, in particular in classification or pattern recognition. Among many kinds of DNNs, the convolutional neural networks (CNN) are most commonly used. However, due to their complexity, there are many problems related but not limited to optimizing network parameters, avoiding overfitting and ensuring good generalization abilities. Therefore, a number of methods have been proposed by the researchers to deal with these problems. In this paper, we present the results of applying different, recently developed methods to improve deep neural network training and operating. We decided to focus on the most popular CNN structures, namely on VGG based neural networks: VGG16, VGG11 and proposed by us VGG8. The tests were conducted on a real and very important problem of skin cancer detection. A publicly available dataset of skin lesions was used as a benchmark. We analyzed the influence of applying: dropout, batch normalization, model ensembling, and transfer learning. Moreover, the influence of the type of activation function was checked. In order to increase the objectivity of the results, each of the tested models was trained 6 times and their results were averaged. In addition, in order to mitigate the impact of the selection of learning, test and validation sets, k-fold validation was applied.