Criteria for comparative analysis of a computer steganography software are offered. Substantiation and choice of software for researches are carried out. Pilot studies of the chosen software by the offered criteria are conducted and on the basis of the obtained results analysis of the most effective software of a computer steganography for concealment of information is defined.

It's proposed a finite-step algorithm with strong stability for solving incorrectly formulated two-point boundary problems for ordinary differential equations arising in the irregular waveguide theory.

Using mathematical modeling the angle-frequency characteristics of the beams and the accuracy of direction finding of the useful signal in an adaptive multibeam antenna arrays (AMLAR) with spatial compensation of noise and sum-difference processing are investigated. It's shown that the required number of generated spatial channels for AMLAR is approximately twice as large as in MLAR without adaptive processing. The nature of dependence of the maximum displacement of the total and zero difference channel of the beam on the deviation of the frequency of the received signal from the reference, for which the vectors of the weighting coefficients are formed at spatial noise compensation and digital diagrams. It's shown that the broadband AMLAR is actually limited by the deviation of the frequency at which there are significant changes in the amplitude diagrams of the direction of the receiving elements.

The methods of transformation of information from one spectral range to another based on Fabry– Perot microresonators are offered. The methods allow to minimize environmental temperature fluctuations influence on characteristics of one or two Fabry–Perot microresonators, which are transforming information from one spectral range to another. In the first case, minimization is performed when the initial temperature point of the microresonator corresponds to the maximum of the probing radiation intensity change depended on temperature. In the second case, minimization is performed when only one of two microresonators has a fixed shift of the initial temperature point.

A research direction in the field of controllable technological systems based on construction of artificial neural network models for functioning of intellectual real-time decision support systems is proposed.

It's made an analysis of the temperature effect on the frequency instability of a generator made on the basis of a metal waveguide. In addition, the change in the electrophysical properties of the active element and the structural elements of the stabilizing resonator is estimated. Recommendations for improvement of the temperature instability of the frequency are given.

In this paper it's obtained a generalized system of statistical inequalities under additional conditions of regularity of the statistical experiment, which is a generalization of the Cramer–Rao inequality. The resulting system of inequalities allows to find the lower bounds of arbitrary even error moments of estimates of unknown parameters. It's found relations which allow one to approximate and numerically calculate the distribution density of the estimation error with a limited set of cumulant coefficients.

The task of optimizing of the compensators of interfering radiations in radars with fast linear scanning is relevant. The main problem of autocompensators with a gradient self-tuning method is the large dynamic errors in the formation of weight coefficients. To reduce the impact of dynamic errors on compensation efficiency, the bandwidth extension of closed self-tuning circuits is used. However, in this case, the contribution of fluctuation errors to the total errors increases. There are optimal values of autocompensator parameters, at which the total errors are minimal. The article obtained expressions for the optimal parameters of a dualchannel auto-compensator for interfering radiations with a gradient self-tuning method when suppressing the signals of two radiation sources. For simplicity, the case of self-tuning chains with the first order of astatism is considered, and the side lobes of the main radiation pattern are described by harmonic approximation. The obtained results are confirmed by the simulation modeling of a two-channel auto-compensator.

The results of modeling of electron transfer processes in a three-dimensional semiconductor structure containing a single layer of graphene using the Monte-Carlo method are presented. The use of graphene, which has a high mobility of charge carriers, high thermal conductivity and a number of other positive properties, is promising for the creation of new semiconductor devices with good output characteristics . As a result of modeling, the dependences of the velocity, average energy, mobility, diffusion coefficient on the structure length and electric field intensity in a semiconductor structure containing a graphene layer and a region of a 4H-SiC silicon carbide material are obtained.

Evaluating of the power spectral density (SPM) of signal are usually performed through procedures using classical methods spectral analysis based on the discrete Fourier transform. This approach to spectral analysis is computationally efficient and provides acceptable results. However, in spite of these advantages, a classical method has a number of fundamental limitations. The most important of these is the limitation of the frequency resolution. For the solution of a problem of increase in the frequency resolution, in practice such methods as MUSIC, EV, Berg, MD assessment etc. In article presents classical methods of estimation power spectrum density, method the minimum error variance (MD estimate) and developed combined method.

Electromagnetic radiation reflection and transmission characteristics of composite materials obtained by forming into a plate a mixture based on powdered titanium dioxide and a gypsum binder were studied, depending on the thickness of these materials. According to the research results, approaches to the practical use of studied materials in electromagnetic shielding systems were developed.

The article is devoted to the condition of R-regularity (Error Bound Property) in problems of mathematical programming. This condition plays an important role in analyzing the convergence of numerical optimization algorithms and it is a fairly general condition of regularity (constraint qualification) in problems of mathematical programming. The article obtains new sufficient conditions for the presence of R-regularity in problems of mathematical programming.

The elementary model of the superconducting state of a thin film in the nonuniform external magnetic field is developed. In the diffusive limit of the superconductivity microscopic theory, the problem of the critical amplitude of the periodic external magnetic field has formulated and solved. In the frame of Ginzburg – Landau model, the critical current problem for the thin film in the weak periodic external magnetic field has solved. The critical current temperature dependencies of the superconductor/insulator/ferromagnet structure have clarified.

Assessment is performed of the permissible nonuniformity of irradiation and temperature scatter as per the wafer area during the rapid thermal treatment, not causing the thermal stresses in it, resultant in the plastic melting or silicon disruption. It was shown, that formation of platinum silicide at Т≤ 810 °С does not cause the negative phenomena in silicon, and it should be performed in the nitrogen environment, whose flooding into the the sealed chamber is done after creation in it of vacuum of 10 –2 mm, mercury column.

The article investigates a low-temperature plasma of the magnetron discharge of a device used for the synthesis of dielectric films by reactive cathode sputtering. The aim of the study is to determine the temperature characteristics of plasma particles and a sputtered substance, as well as the mechanism for the formation of a chemical bond between sputtered atoms and active gas molecules. A study of the composition and energy parameters of the plasma, as well as the chemical composition of the particles obtained by sputtering, was carried out by a spectroscopic method. The quantitative composition was determined by a mass spectrometer to determine the composition of the sputtered particles.

The efficiency of threshold decoding of non-binary self-orthogonal convolutional codes in the communication channel with additive white Gaussian noise with the correction of t (t ≥ 1) symbolic errors is considered. The characters or elements of the code sequence (CS) used by the self-triggered convolutional code with the code transmission rate equaled R = k₀/n₀ = 1/2 are the elements of the final field GF(aⁱ), i ≥ 2. The number of erroneous code symbols and their positions in the received CS on the length of the code constraint equaled n_a = (m + 1)∙n₀ binary symbols is determined by the decoder based on the structure of the generated syndrome sequence (SS) S(x) = {S₀, S₁, …, S_N}, depending on the structure of errors in the communication channel and the structure of the generating polynomials: m is maximum degree of the generating polynomials. It has been established that while preserving the merits of the threshold decoding algorithm, an increase in i (i ≥ 2) times the multiplicity of corrected errors is ensured by binary self-orthogonal codes with an equal transfer rate of codes.

The application of visible and infrared images integration in tracking target of the homing head (HH) to improve the shooting efficiency of anti-aircraft missile system (AAMS) type «Strela-10M2» is substantiated. A variant to reduce the firing cycle of AAMS by eliminating the fact of switching operating modes is presented. It's shown the dependence of likelihood of fire mission execution by a subunit armed with AAMS as a function of the image quality of the accompanied target characterized by a standard deviation of pixel brightness in the tracking monitor of the HH.

The possibility of implementing a remote software update mean for connected to a common network devices controlled by microcontrollers is considered. Requirements for the mean are presented. The structural scheme of this mean for ARM microcontrollers using the CAN-protocol is developed.