Vol 15, No 4 (2017)

The progress in the development of femtosecond fiber lasers and their applications makes the problem of modeling the high power ultrashort pulses propagation in optical fiber cables extremely important. Modern fiber lasers can generate pulses with a duration of less than ten femtoseconds with a peak power of up to tens GW. At such values of peak power nonlinear effects cannot be ignored. For fused quartz fibers and femtosecond pulse duration it can be assumed that only the Kerr nonlinearity is present. For this assumption the approximate analytical solution for the fundamental mode of an optical fiber with step-index profile was already found. This solution was obtained with Gaussian approximation method and it already takes the Kerr nonlinearity into account. However, it neglects the second degree of approximation, which is unacceptable at peak power values exceeding tens of megawatts. In this paper, we consider a solution that takes into account refractive index dependence on high-order terms of light signal peak power which occurs due to the Kerr nonlinearity. For some optical fibers studied the results of the analysis are presented, showing the dependency of the fundamental mode spot radius and propagation constant on the light signal peak power in the presence of the Kerr nonlinearity. It is shown that, with an error of less than 0.1%, the dependence of the fundamental mode propagation constant on the light signal power can be approximated by a second-order polynomial. For the considered optic fiber samples the values of the approximation coefficients were obtained.

This paper shows the use of median filters with weighted central elements for removing the impulse noise from the image. The influence of the probability of obtaining undistorted pixels and distortion of the image pixel distribution function on the processing result is shown. It is demonstrated that median filters with a 3x3 window allow to obtain a higher quality of image denoising at low probability of impulse noise, and median filters with window sizes of 5x5 and 7x7 allow to obtain a higher quality of image denoising at high probability of impulse noise. The obtained results can be used in medical image processing and adaptive filtering systems for photo and video data processing.

Energy saving in the information and communication technology field is an important problem identified while designing Future Networks. To reduce energy consumption, it is important to analyze it at each stage of the lifecycle. In case of networks the in-use mode is primarily considered because energy consumption in this mode is usually the major issue for always-on network equipment. The energy consumption of this mode can be controlled by network architecture, capabilities and operations. The article describes the method of measuring power consumption of passive optical network (PON) equipment with time division multiplexing (TDM). Analysis method is based on the ITU Recommendations Y.3021 and Y.3022. Evaluation of potential energy savings is performed for the Gigabit Passive Optical Network equipment of Eltex Russian company using sleep and cyclic sleep modes. The results of the analysis show that the power consumption of the passive optical network can be reduced to 30% when using sleep mode and up to 50% in cyclic sleep mode with an average network load.

Currently, there are no analytical methods for precisely determining the characteristics of queuing systems G/M/1 or G/M/m, and this lowers the degree of adequacy of stochastic network models to real computer and telecommunications networks and the quality of design decisions. This article shows the comparative results presented in various publications devoted to the calculation of the average waiting time in queuing systems G/M/1, in particular queuing system with hyper-exponential arrival rate. The results under consideration were obtained by completely different approaches, but the final calculated results are completely identical. These approaches have certain advantages to each other. Using of the analogues simplifies calculation of the characteristics for the system considered, and the author’s approach expands the possibilities of calculating characteristics in terms of determining higher-order moments for the waiting time, which allows extracting information needed to calculate a jitter, for example. Considering the definition of jitter in telecommunications as the spread of the waiting time from its average value, knowledge of higher order moments for the waiting time makes it possible to obtain the jitter value via calculating the variance, i.e. second moment of random waiting time.

In the article a method of estimating business process for a telecommunication company is proposed. The method is based on process mining technology, which allows to identify ineffective operations using logged data. The article reviews the basic steps of estimating a technological process: creating the model of the process and the reference model, testing the model, formalizing the requirements and performing quality estimation. Identification is performed by the decision tree, which is built automatically using machine learning methods. It is shown that the new method allows to decrease the amount of diagnostic operations needed for analyzing a technological process.

Threshold based methods are the most frequently used methods of image segmentation due to their intuitive nature and ease of implementation. In this article, we consider methods of automatic threshold selection, as well as methods for threshold adjustment in accordance with local image properties. The studies were conducted using Matlab software. Global threshold processing cannot produce the desired result if the image background is highly heterogeneous in brightness. In such cases, it is necessary to apply pre-processing to compensate for variations in background brightness, after which a global threshold transformation can be performed. One way to choose a threshold is to visually examine the image histogram. If the histogram has two distinct modes, then it is easy to choose the threshold separating them. Another approach to threshold selection is based on the trial and error method, when different thresholds are selected and checked until the result of the threshold processing satisfies the requirements.

Today the capabilities of intelligent transport systems are largely determined by the characteristics of the hybrid wireless vehicular ad hoc networks (VANET). These networks use orthogonal frequency-division multiplexing (OFDM) technique for direct communications between vehicles as well as communications between vehicles and stationary infrastructure devices. Connections can be direct or they can be relayed by means of cooperative diversity when it is needed to improve noise immunity. One of the specifics of the vehicular ad hoc networks is the presence of high-speed vehicles, for which the Doppler spread results in the lost orthogonality of the OFDM signal and interchannel interference which decreases the noise immunity. The article considers the impact of the interchannel interference on the noise immunity of OFDM system using analog relay based cooperative diversity with two vehicles. Fading is modelled individually for each of the OFDM signal subcarriers using baseband equivalents and Rayleigh statistics. Doppler spread model corresponds to isotropic scattering. Keywords: cooperative diversity, AF relay, Rayleigh fading, spectral Doppler spreading, SEP

Forensic examination often involves studying various kinds of images obtained from various sources. One of the tasks assigned to an expert is to enhance image quality to obtain visually distinguishable forensically significant details, such as registration plate number or appearance details of a vehicle or a person. In the article an expert assessment of the obstacle influence on recognizing of a registration plate number is presented and the limitations for initial image quality are outlined. Furthermore, a modified image recovery algorithm is introduced for reconstructing fragments of small images, which allows to obtain additional forensically significant data.