**SIMULATION OF SIGNALS TIME-FREQUENCY SCATTERING IN A MULTIPATH CHANNEL**

L.N. Kazakov, P.G. Demidov Yaroslavl State University, head of the department of radio engineering systems, Dr.Sci.(Tech.), professor; kazakov@uniyar.ac.ru

D.M. Soloviev, P.G. Demidov Yaroslavl State University, department of radio engineering systems, engineer, Ph.D.; soldm89@gmail.com

E.A. Selianskaia, P.G. Demidov Yaroslavl State University, department of radio engineering systems, assistant; eselyanskaya@mail.ru

I.S. Afanasiev, P.G. Demidov Yaroslavl State University, postgraduate; gipnonya@gmail.com

The method for constructing and implementing a discrete model of frequency-time scattering of a radio signal in a mobile channel is proposed. Described method is optimal by the criterion of minimizing the computing resource. The approach for constructing the discrete channel impulse response is based on analysis of the systems sensitivity to the number of rays. For simulating Doppler scattering a low-pass filter is used. Computing costs increase with increasing filter order. The approach to the selection of the filter order based on sensitivity of the probability of a bit error to the parameters of the filter is proposed. These principles have been used as the basis for a signal simulator implementation. This simulator was used for investigation of a broadband radio communication system. The experiment was carried out for verification described approach. The obtained results indicate a high degree of adequacy of the simulator of frequency-time signal scattering to real processes in the channel.

**Keywords:** multipath, optimal number of beams, simulation, Jakes *spectrum, optimal filter order, computing resources, energy efficiency.*

**QUALITY RESEARCH OF THE IONOSPHERIC CHANNEL DISPERSION CHARACTERISTICS SLOPE ESTIMATION ALGORITHMS**

E.M. Lobov, MTUCI, SRD, senior researcher, Ph.D.; lobov@srd.mtuci.ru

E.O. Smerdova, MTUCI, SRD, first category engineer, lizabeth2@mail.ru

The article compared the two methods of evaluating the slope of the dispersion characteristics of the ionospheric channel, probing its broadband signals — on the maximum likelihood criterion and the method of indirect estimation, through the evaluation of signal delays in the different sub-bands. There are formulas for calculating the dispersion of the estimates in the paper. Curves are plotted standard deviations of the estimates when using broadband signals of various types and the corresponding values obtained through simulation.

**Keywords:** frequency dispersion, wideband signal, distortion,*maximum likelihood, ionosphere.*

**METHOD OF MEASUREMENT AND CALIBRATION OF SIGNAL DELAYS IN MODULATORS OF DVB-T2 STANDARD TRANSMITTERS**

V.L. Karyakin, Povolzhskiy State University of Telecommunications & Informatics, department of radio communication, radio broadcasting and television, Professor, Dr.Sci.(Tech.); vl@karyakin.ru

D.V. Karyakin, Russian representation of Juniper Networks, senior systems engineer, Ph.D.; dm@karyakin.ru

L.A. Morozova, Povolzhskiy State University of Telecommunications & Informatics, department of economics and production, associate professor, Ph.D.; morozova@rcf.ru

The method of calibration of time delays in modulators of transmitters with a controlled delay line and a device for its implementation at the program level providing a high accuracy of measurements when testing transmitters of the SFN network are presented.

**Keywords:** SFN network, DVB-T2 standard, synchronization, calibration of signal delays.

**OPTIMAL CHANNEL RESTRICTION WITH PRECODING LEVEL PULSATION AT THE EDGES OF THE SYMBOL INTERVAL**

V.P. Volchkov, MTUCI, department «General theory of communication», professor, Dr.Sci.(Tech.); volchkovvalery@mail.ru

V.G. Sannikov, MTUCI, department «General theory of communication», professor, Ph.D.; tes_mtuci@mail.ru

The problem of synthesis of optimal channel precoders for digital telecommunication systems with symbol-by-symbol data multiplexing over channels with frequency-selective fading is solved. The selected criterion of optimality can effectively compensate for fading without using the equalizer at the receiver, forming at its entrance a smooth finite channel pulses with low pulsation at the edges of the symbol interval. The results of computational experiment confirming the specified properties of precoder are presented.

**Keywords:** digital telecommunication system, frequency-selective channel, finite signal bases, channel precoding, intersymbol interference, variational method of synthesis, optimal pre-emphasis, feedback channel, Lagrangian function, symbol-by-symbol data multiplexing.

**PURIFICATION OF THE DISTORTED VOICE CALL FROM NOISE ON A BASIS PSYCHOACOUSTICAL MODEL OF ITS ACOUSTICAL PERCEPTION**

V.G. Sannikov, MTUCI, department «General theory of communication», professor, Ph.D.; tes_mtuci@mail.ru

A.A. Korolkov, MTUCI, department «General theory of communication», postgraduate student; alexkorolkov@mail.ru

The modified algorithm of purification of the distorted speech from noise is offered. Novelty of algorithm consists in use of psychoacoustical model of an acoustical speech perception with corrective action of its informative a component in the frequency range. The results of computing experiment showing substantial increase of intelligibility of noisy speech at small signal/noise ratio are presented.

**Keywords:** the distorted speech, psychoacoustical model of acoustical system, algorithm of spectral subtraction, purification of speech from noise, intelligibility of speech.

**SIGNAL DISTORTION ESTIMATION AND COMPENSATION INTO THE DOPPLER’S SPREAD SPECTRUM AND RAYLEIGH’S FADINGS CHANNELS**

N.E. Poborchaya, MTUCI, department «General theory of communication», associate professor, Ph.D.; n.poborchaya@mail.ru

E.R. Khasianova, MTUCI, junior research fellow; ehasyanova@gmail.com

Time dynamic channel estimation problem is considered in the paper due to its following compensation. The task is solved with the help of the least square method under the noise distribution expected uncertainty conditions. The communication channel is approximated by the p-power polynomial inside the time-sliding window. Execution of the computational experiment shows the performance capability of the procedure in the presence of slow Rayleigh fadings.

**Keywords:** estimation, least square method, expected uncertainty, rayleigh fadings, doppler’s spread spectrum.

**ALGORITHM 2N-FFT FOR ESTIMATION OF CYCLIC SPECTRAL DENSITY**

T.Ya. Shevgunov, Moscow Aviation Institute (National Research University), National Research University Higher School of Economics, associate professor, Ph.D.; shevgunov@gmail.com

Eu.N. Efimov, Moscow Aviation Institute (National Research University), engineer, Federal State Institution «Scientific Research Institute for System Analysis of the Russian Academy of Sciences», senior project engineer; omegatype@gmail.com

D.M. Zhukov, Moscow Aviation Institute (National Research University), graduate student, ZAO “RADIY TN”, junior researcher; dima.m.zhukov@gmail.com

The paper introduces 2N-FFT algorithm for cyclic power spectral density estimation performed for digital signals represented by their finite length sample sequences. The algorithm is based on time smoothed cyclic periodogram method and uses frequency interpolation property of FFT in order to achieve required cyclic frequency resolution. The introduced algorithm is applied to two random cyclostationary signals: amplitude modulated (AM) radio signal and pulse amplitude modulated (PAM) sequence. The paper also contains brief analysis of the main advantages and disadvantages of the introduced algorithm and overviews the required computational cost of its current implementation.

**Keywords:** cyclostationarity, cyclic spectral density, CSD, cyclicspectrum, signals and systems, spectral analysis, spectral correlation analysis, FFT.

**METHOD OF DENSITY ESTIMATION DISTRIBUTIONS WITH «HEAVY» TAILS**

V.G. Kartashevskiy, PSUTI, head of the department of multiservice networks and information security, Dr.Sci.(Tech.), professor; kartashevskiy-vg@psuti.ru

M.V. Meshcheryakov, PSUTI, department of multiservice networks and information security, postgraduate student; maxa315@bk.ru

E.A. Kuznetsova, PSUTI, department of multiservice networks and information security, postgraduate student; yablochkinaea@mail.ru

The estimation of approximation quality of the probability densities of the time series is made by the kernel regression method. The potential of the method is demonstrated for the estimation of the distributions with «heavy» tails. Method of obtaining a simple mathematical expression for unknown probability density of limited size sample is presented. The dependence of the root-meansquare error of the histogram approximation on the window width are graphically presented. To determine the optimal window width a two-step approach was used. Advantages and disadvantages of the kernel regression method are shown. The optimal window width is obtained with a minimum approximation error.

**Keywords**: density estimation, histogram approximation, distributions with «heavy» tails.

**METHOD OF DETERMINING OF NOISE IMMUNITY OF MULTIDIMENSIONAL SIGNALS**

M.A. Bykhovskiy, MTUSI, professor, Dr.Sci.(Tech.); bykhmark@gmail.com

Presents a method of evaluating noise immunity of reception of signals belonging to a surface-spherical multidimensional ensemble of signals (PSAS).This ensemble, as it was shown by Shannon, is optimal to transmit messages in communication channels with white Gaussian noise. This method is based on the analysis of the algorithm of their processing is in the optimum demodulator the received signals. The formulas, allowing to estimate distribution of distances between signals belonging to PSUS and the range of weights of codewords from an error-correcting code having a certain length and the code rate, is presented. These results are used to assess error probability in demodulation of signals belonging to PSAS. The estimates obtained are compared with those given in other works, on the basis of which it is concluded that the proposed method improves the accuracy of estimation of error probability in the demodulation of the received signals.

**Keywords:** noise immunity of message reception, method of probability calculation, cascade source coding, multidimensional signal ensemble, complexity realization of demodulators and decoders.

**THE STUDY OF SEMANTIC GATEWAY PERFORMANCE**

V.A. Kulik, The Bonch-Bruevich State University of Telecommunications (SPbSUT), department of communications networks, assistant; vslavk.kulik@gmail.com

A. Muthanna, SPbSUT, department of communications networks, associate professor, Ph.D.; ammarexpress@gmail.com

V.D. Pham, SPbSUT, student; daipham93@gmail.com

A. Hakimov, SPbSUT, student; abdukadir94@gmail.com

R.V. Kirichek, SPbSUT, department of communications networks, associate professor, Ph.D.; kirichek@sut.ru

R.Ya. Pirmagomedov, SPbSUT, department of communications networks, associate professor, Ph.D.; lts.pto@yandex.ru

This article studies IoT gateway performance for information transmission between networks using MQTT, CoAP, HTTP/2 and XMPP protocols. The networking architecture and software for semantic gateways in the Internet of Things have been developed. The experimental results were obtained, as well as the main characteristics of gateway performance (delays due to protocol conversion, CPU and memory load) when converting the above protocols. These results allow evaluating possible use of various protocols in real-time control systems based on the Internet of Things technologies.

**Keywords:** Internet of Things, semantic gateway, MQTT, CoAP, HTTP, XMPP, microservices.