V.A. Petrov and L.V. Baranova
Kharkiv National University of Radio Engineering and Electronics,
14, Lenin Ave, Kharkiv, 61166, Ukraine
Source Structure and Directivity Pattern of Secondary Radiation by the Long-Distance Tropospheric Wave Scattering
The structure of secondary radiation sources at the scattering of waves into finite spatial angle is considered. It is shown that these sources are analogous to linear antenna systems of running wave. The secondary radiation diagrams were calculated according to random disturbances and adjust with experimental data.

Back
O.A. Remayeva, Ye.V. Remayev, and G.K. Snurnikova
1Kharkov National University of Radio Engineering and Electronics,
14, Lenin Ave, Kharkov, 61166, Ukraine
Investigation of Joint Influence of the Earth Troposphere and Ionosphere upon the Accuracy of Corrections to GPS Measurement Results

Investigation into the simultaneous action of the Earth troposphere and ionosphere on precision of GPS measurements are carried out. It is shown that the dissimilarities of complete correction from the sum of tropospheric and ionospheric corrections can reach 0.15...4.95 m in the range of zenith angles of 70 o...90o. This effect is connected with variations of trajectory form of electromagnetic signal at differ approaches to examination of its propagation peculiarities.

Back
N.N. Beletskii, S.A. Borysenko, and V M. Yakovenko
A. Usikov Institute of Radio Physics and Electronics,
National Academy of Sciences of Ukraine
12, Academician Proskura Str., Kharkiv 61085, Ukraine
Magnetoresistance and Spin Polarization of Electron Current in Magnetic Tunneling Junctions
The influence of the bias voltage on the magnetoresistance and spin polarization of the electron current of the ferromagnetic metal-insulator-ferromagnetic metal (FM/I/FM) tunneling junction has been studied theoretically a two-band model of free electrons in ferromagnetic electrodes. It is shown that the value and sign of the magnetoresistance of the FM/I/FM tunneling junction depend essentially both on the height and the width of the potential barrier and on the value of the bias voltage. It has been found that the height and the width of the potential barrier can be chosen in such a way that the value of the magnetoresistance can be both decreased and increased as the bias voltage increases. It has been determined that the magnetoresistance changes the sign and oscillates with increasing bias voltage. The change of the sign of the magnetoresistance takes place due to the fact that the tunneling transmission coefficient of the majority electrons is the largest in the case of the antiparallel magnetization of the ferromagnetic electrodes and not parallel one. It has been shown that the value and sign of the spin polarization of the electron current can be changed with reversing the ferromagnetic electrode magnetizations.

Back
V.V. Lukin1, S.K. Abramov1, N.N. Ponomarenko1, Benoit Vozel2, and Kacem Chehdi2
1Department of Transmitters, Receivers and Signal Processing, National Aerospace University, Chkalova Str. 17, 61070 Kharkiv, Ukraine, 2University of Rennes I, 6, Rue de Kerampont, 22 305 Lannion cedex, BP 80518, FRANCE,
Methods for Blind Evaluation of Noise Variance in Multichannel Optical and Radar Images
A priori knowledge of noise type and statistical characteristics (at least, noise variance) is a pre-requisite of successful solving many practical tasks of image processing like filtering (denoising), edge detection, segmentation, etc. This relates to both gray-scale and multichannel (multi- and hyper-spectral) images. However, in practical situations noise characteristics depend upon many factors and it is a common practice to determine noise type and variance just for images at hand. It is desirable to perform this operation automatically especially if one deals with multichannel images. Thus, below we consider several approaches and methods for blind evaluation of noise variance. Their accuracy and applicability are analyzed for images with different properties, in particular, various percentage of texture regions. Besides, a set of noise variance values is tested.

Back
V. Golikov and O. Lebedeva
Universidad Autonoma del Carmen, Ciudad del Carmen, Campeche, Mexico
Efficiency and Sensitivity of CFAR Detection in Clutter with Unknown Covariance Matrix

We conduct a simulation analysis for assessing the constant false alarm rate (CFAR) behavior of three coherent radar detectors in the presence of correlated Gaussian clutter with unknown covariance. We establish the conditions a detector must fulfill in order to ensure the CFAR property. We discuss several detectors with quasi CFAR property. The analysis of CFAR property and the performance analysis, which has been carried out in the presence of correlated Gaussian clutter with unknown covariance, shows that the proposed detectors exhibit a quite acceptable loss with respect to optimum Neyman-Pearson detector.

Back
Jose Velazquez-Lopez, Juan Carlos Sanchez-Garcia and Hector Perez-Meana
National Polytechnic Institute of Mexico, Mexico
Adaptive Filters with Codified Error LMS Algorithm

In this paper a modification of the Least Mean Square (LMS) adaptive algorithm is presented based on a modification of the codification of the error signal. This modification allows having an algorithm easier to implement in a digital adaptive filter, besides that a greater convergence speed is obtained. A comparison of the proposed modification with the conventional LMS algorithm and some of its known variants is provided to show the advantages of proposed algorithm.

Back
V.A. Tikhonov, K.V. Netrebenko, I.V. Savchenko
Kharkov National University of Radio Engineering and Electronics,
14, Lenin Ave, Kharkiv, 61166, Ukraine
The Wiener Third-Order-Statistics-Based Compensator of the Non-Gaussian Noise

The enhancement problem of the Gaussian signals against a background of non-Gaussian correlated noises is solved. To solve it the Wiener filters synthesized with the third order moment functions is offered. The equations for linear parameters estimation are determined. Examples of enhancement are given the Gaussian signals against a background of the non-Gaussian correlated noises using the reference channel are discussed.

Back