卷 15, 编号 2 (2017)

The article describes the problem deficit of the quantity of ensembles of orthogonal signals for stochastic using. The task of the article is to develop a methodology for estimating the quantity of ensembles of multiphase orthogonal signals variants obtained, described by eigenvectors of bidiagonal Hermitian matrices. The aim of the article is to determine the maximum quantity such signals. The use of obtained ensembles of multiphase orthogonal signals is possible not only in the radio channel, but also in the optical channel. There is a device, which can transmit an Mary symbol with a pseudo-random reconstruction of the intensity of the glow of a photocell. In the article authors have calculated the quantity of ensembles of new multiphase orthogonal signals for Hermitian matrices of various orders, have presented models of multiphase orthogonal signals in 3D space. Authors have got some formulas for calculating the quantity of unique signals, taking into account the appearance (absence) of inverse signals and graphical relation of quantity of unique ensembles of multiphase orthogonal signals and pseudorandom chaotic sequences from orders of Hermitian matrices. Comparative analysis of the above signals revealed the advantage of the first.

In this article, the main algorithms of fuzziness clusterization for determination nodes of demand in mobile radio networks are compared. Identification of the most suitable algorithm for dynamic finding of nodes of demand in the conditions of abnormal zones of spatial-time changes of traffic transmitting (change of loading) is necessary for fast redistribution of a final radio-frequency resource of base stations, communication channels and computing powers of the provider. Each algorithm of an accurate and fuzzy clustering is the unique mathematical tool capable to analyze traffic on the modern telecommunication. At the same time, all algorithms have applicability boundaries in real implementation - the calculation speed, feature of a choice of boundaries of fuzziness, a choice of a metrics of belonging to some cluster and the whole range of heuristic criteria for the solution of real physical tasks. Comparing of two algorithms of a fuzziness clustering revealed that Fuzzy C-Means works quicker in the conditions of the increasing traffic, than Gustafsson-Kessel's algorithm. Simulation was made for two cases: increase of number of subscribers and increase of quantity of clusters in case of invariable number of subscribers.

Using functional transformation, we propose an approach allowing to convert a self-similar input packet stream of multiservice traffic and to obtain a stream with the properties of the elementary stream and the variation coefficient equal to one. A model is developed that converts a self-similar input packet stream that distributes the intervals between packets distributed according to the law the Pareto in the flow is distributed between packets that describe the exponential distribution. A functional connection between the Hurst exponent and the coefficient of variation, which determines the structural complexity of the input packet flow, possessing self-similar properties, is found. A functional relationship between the coefficient of variation and the Hurst index is found using the cubic regression equation. Approximation of the dependence of the coefficient of variation on the Hurst index of the input flow is presented.

Multi-core fiber has been one of the main innovations in fiber optics in the last decade. Reported work on multi-core fiber has been focused on increasing the transmission capacity of optical communication links by exploiting space-division multiplexing. Additionally, multi-core fiber presents a strong potential in optical beamforming networks. The use of multi-core fiber can increase the compactness of the broadband antenna array controller. This is of utmost importance in platforms where size and weight are critical parameters such as communications satellites and airplanes. Here, an optical beamforming architecture that exploits the space-division capacity of multi-core fiber to implement compact optical beamforming networks in hybrid networks is proposed, being a new application field for multi-core fiber. The experimental demonstration of this system using a 4-core multi-core fiber that controls a four-element antenna array is reported. An analysis of the impact of multi-core fiber on the performance of antenna arrays is presented. The analysis indicates that the main limitation comes from the relatively high insertion loss in the multi-core fiber fan-in and fan-out devices, which leads to angle dependent losses which can be mitigated by using fixed optical attenuators or a photonic lantern to reduce multi-core fiber insertion loss. The crosstalk requirements are also experimentally evaluated for the proposed multi-core fiber-based architecture. The potential signal impairment in the beamforming network is analytically evaluated, being of special importance when multi-core fiber with a large number of cores is considered. Finally, the optimization of the proposed multi-core fiber-based beamforming network is addressed targeting the scalability to large arrays in data transmission hybrid networks.

Importance of an investigation of specific aspects of ultrashort pulse propagation and voltage peak values localization along printed circuit board (PCB) multiconductor bus with different approaches (with fixed selection of an ultrashort pulse duration and with using an optimization) is highlighted. Trapezoidal ultrashort pulse propagation along the conductors of real PCB multiconductor bus was simulated. With the usage of genetic algorithms (GAs) an optimization of the whole ultrashort pulse duration and severally of the rise, top and fall durations was made by criteria of peak voltage maximization in the PCB bus. A signal maximum which is revealed with the fixed selection of the ultrashort pulse duration by 20% exceeds the steady state level. A crosstalk minimum value which is 39,8% of the steady state level is revealed and localized. A signal maximum which is revealed with the using of the optimization by 38% exceeds the steady state level. It is shown that significant signal excess is observed with decreasing of ultrashort pulse duration both for maximum and minimum. Also, localization of voltage peak values is inconstant: they appear in different parts of PCB bus including different PCB layers.

The problem of improving the reliability of biometric authentication procedures of users of computer systems using small training sample for configuring of recognition machines was considered. In the present article is shown that the Bayes multidimensional correlation functionals can be modified to using the identically correlated in modulus biometric data. The correlation coefficients of biometric data have a significant error at a small test samples. This impedes their use when configuring the classical quadratic form and Bayesian networks. The authors suggested using the method of symmetrization of biometric data, ie, replace biometric data by other multidimensional data with the same entropy. After symmetrization of correlation matrix it will have the same coefficients of pair correlations outside the diagonal. It is proved that requirements to biometric data volume are significantly reduced while using the mentioned method, error of calculation results of correlation coefficients gets smaller. The effect of increasing the stability of calculations is observed for any biometric data. As a consequence, the configuring of the classical quadratic forms and of the Bayesian maximum likelihood networks become more resistant tasks. The data that allowing to estimate how much the size of biometric data sample can be reduced are provided.