2687-0517 14 1 2021 1-70

RAR RUS 7-21 Shashikhin Vladimir Turulin Aleksandr sanya.turulin.98@list.ru Budnik Svetlana budnik.sveta@mail.ru Image encryption algorithm based on controlled chaotic maps The article reviews the problem of ensuring the security of storage, processing, and transmission of images based on a cryptographic method using chaotic maps. The encryption algorithm is based on a three-dimensional mapping. The encryption algorithm strength when using systems with chaotic dynamics depends on the value of the largest (positive) Lyapunov characteristic exponent. Therefore, the problem of increasing resistance to various kinds of attacks is reduced to determining the control parameters, at which the leading Lyapunov characteristic exponent increases. The authors propose a procedure for changing the chaotic map characteristics (entropy and Lyapunov characteristic exponents) based on introducing feedback into the system. The procedure is developed using the modal control method based on reducing the system to the canonical Frobenius form. The use of the proposed algorithm is considered on the example of the Rössler system. The test results confirmed an increase in the strength of the proposed encryption algorithm against statistical and differential analysis due to an increase in the Lyapunov characteristic exponent. 10.18721/JCSTCS.14101 004.056 Image encryption chaotic maps control of the spectrum of Lyapunov characteristic exponents modal control canonical Frobenius form https://infocom.spbstu.ru/article/2021.68.1/

RAR RUS 22-32 Kotlyarov Evgeniy ekotlyarov86@gmail.com Mikhailov Viktor vmikhaylov@niime.ru Zubov Igor izubov@niime.ru Nuikin Andrey anuykin@niime.ru Iljin Alexander alexiljin0@gmail.com P-5603-2015 14042371400 0000-0003-0608-7824 National Research University of Electronic Technology Putrya Mikhail pmg@miee.ru

Shokin Square, Bld. 1, Zelenograd, Moscow, 124498, Russia

CMOS inductor design features for LTE devices This work is devoted to some aspects of the development of planar elements of the microwave path, which are used in the design of low-noise LTE range amplifiers, namely inductors, for further employment as part of the NB-IoT transceiver. General theoretical calculations on the design of high-frequency inductors are given. By the example of a multilayer CMOS 90 nm inductor with a nominal value of 7 nH, we demonstrated the influence of the structure thickness obtained by replicating layers used to get the required skin layer thickness and to achieve the best quality factor in the LTE frequency range from 0.5 to 3.5 GHz by means of electromagnetic (EM) models. For a better understanding of the inductor operation, the models obtained as a result of the EM simulation for different values of the substrate conductivity are compared. The obtained data partially refute the need of increasing the maximum thickness of the inductors by a set of upper metals combined using multiple TSV arrays for silicon process stacks. Due to the increasing of the capacitive influence of the substrate on the lower metal layers of the inductor, the highest values of Q-factor and self-resonance frequencies are achieved by the structures with a minimum number of metal layers, despite the negative influence of the skin effect for low frequencies. 10.18721/JCSTCS.14102 621.372 LTE NB-IoT inductance coil 3D inductor CMOS EM analysis https://infocom.spbstu.ru/article/2021.68.2/ 22-32.pdf

RAR RUS 33-42 Aslanov Gadarbek uits@dstu.ru Kazibekov Rustam kazibrus11@mail.ru Musaibov Rashid rashid_musaibov@mail.ru Disadvantages of the instrumental error determining method of airfield quasi-Doppler automatic direction finders An increase in the aircraft flights intensity leads to the need to improve the efficiency of the airfield radio equipment functioning. One of the ways to improve the accuracy of navigation equipment is to reduce the instrumental error. The article analyzes the methods of checking the instrumental error of the ADF. Is shown that the so-called “electric rotation” of the antenna system (AS) head used in the radio direction finders ADF-80K, ADF-80, ADF-85, “Platan”, DF-2000 for checking the instrumental error leads to incorrect results. The authors employed the linearity property of the Fourier transform in the simulation to prove the incorrectness of the method for determining the ADF instrumental error with the “electric rotation” of the AS head. The simulation results showed that in the ADF operating in the quasi-Doppler mode, the failure of the vibrators located along the bearing line to the radio source does not lead to the appearance of bearing error, while the failure of the vibrators located orthogonally to the bearing line can result in a bearing error reaching 3.750. The simulation results confirmed that unlike the ADF that use the AS head “electric rotation”, the ADF with mechanical rotation show reliable results of the instrumental error measurement. The paper proposes a new method for checking the instrumental bearing errors using the values of the AS phase non-identities which ensures the reliability of the measurement results. 10.18721/JCSTCS.14103 621.396 instrumental error aerodrome automatic direction finders phase non-identity antenna system electric rotation https://infocom.spbstu.ru/article/2021.68.3/ 33-42.pdf

RAR RUS 43-49 Vargauzin Viktor var@mail.spbstu.ru Nikolaev Denis nikolaev.d.i@yandex.ru High precision passive radar algorithm The work is devoted to the study of the location determination algorithm for an arbitrary number of receiving stations using the differential-rangefinder method. The algorithm uses all possible Time Differences of Arrival (TDOAs) of the signal from the radio emission source to the receiving stations. In this case, the concept of a “reference” receiving station is excluded, relative to which the range differences are estimated in the classical method, and the signal TDOAs from the source between all possible pairs of receiving stations are used. It is shown that for a given number of receiving stations, the transition from the algorithm with one “reference” station to the proposed algorithm can significantly increase the accuracy of determining the location. Moreover, with an increase in the number of receiving stations, the efficiency of such a transition increases. In addition, for both methods, it has been demonstrated that adding a new receiving station improves positioning accuracy, but the gain decreases with the increasing number of stations. The work can find application in various monitoring systems, since it can significantly increase the accuracy of location determination only through algorithmic solutions, without costly replacement of equipment. 10.18721/JCSTCS.14104 621.396.4 determination of object location receiving station time difference of arrival radio emission source positioning accuracy https://infocom.spbstu.ru/article/2021.68.4/ 43-49.pdf

RAR RUS 50-59 Cajas Guijarro Carlos David cajasgcarlos@hotmail.com Budanov Dmitriy dmitriy.budanov@gmail.com On programming an application to mitigate DoS attack using OpenDaylight controller in software-defined networking Denial of Service (DoS) attacks try to deplete system resources by consuming bandwidth. In this paper the application using Software-Defined Networking (SDN) principles for DoS attack mitigation based on traffic monitoring in a network is proposed. The most important details about the programming aspects of the application using OpenDaylight (ODL) are explained. The application generates both proactive and reactive rules that should be installed in the network devices. Therefore, it is possible to have statistics of the flows and track possible anomalies such as an unexpected increase of the throughput in one or more of the flows. This allows to detect a DoS attack and mitigate it, installing the appropriate rules. Simulation results obtained with the application when using virtual switches in a network with a linear topology are presented. 10.18721/JCSTCS.14105 004.72 DoS attack software-defined networking OpenDaylight controller proactive rules reactive rules throughput https://infocom.spbstu.ru/article/2021.68.5/ 50-59.pdf

RAR RUS 60-69 Hanafi M. Yassine hanafi.med.yassine@gmail.com AAH-8784-2019 35303230700 Peter the Great St.Petersburg Polytechnic University Vyacheslav P. shkodyrev@imop.spbstu.ru

Polytechnicheskaya, 29, St.Petersburg, 195251, Russia

Reinforcement learning for industrial manufacturing control system The problem posed is a very general case of optimal control of a dynamic, potentially stochastic, and partially observable system for which a model is not necessarily available. We analyze the disadvantages of classical approaches of the control theory and present a new modified numerical reinforcement learning rule of machine learning algorithm. Control theory is a field that has been studied for a very long time and which deals with the behavior of dynamic systems and how to influence it. Among the best-known examples are LQG (Linear Quadratic Gaussian) or PID (Proportional Integral Derivative) controllers. Most of the existing approaches presuppose (analytical) knowledge of the dynamic system, and one of the constraints is the need to be able to free oneself from a priori models. We focus on modified reinforcement learning approach to adaptive control policy as perspective area of control of complex dynamical system under uncertainty. 10.18721/JCSTCS.14106 004 reinforcement learning multi agent system oil manufacturing Bellman equations dynamic programming https://infocom.spbstu.ru/article/2021.68.6/ 60-69.pdf