2687-0517 13 1 2020 1-79

RAR RUS 8-18 Peter the Great St. Petersburg Polytechnic University Kochovski Petar petako_bt@hotmail.com

St. Petersburg, Russian Federation

56049610600 0000-0003-1116-7765 Peter the Great St. Petersburg Polytechnic University Drobintsev Pavel drobintsev_pd@spbstu.ru

St. Petersburg, Russian Federation

An approach for automated deployment of cloud applications in the Edge-to-Cloud computing continuum satisfying high Quality of Service requirements Modern component-based software engineering environments allow deployment of cloud applications on various computing infrastructures, such as Edge-to-Cloud infrastructures. The heterogeneous nature of such computing resources results in variable Quality of Service (QoS). Therefore, the deployment decision can seriously affect the application’s overall performance. This study presents an approach for automated deployment of cloud applications in the Edge-to-Cloud computing continuum that considers non-functional requirements (NFRs). In addition, the authors explore multiple methods for selection of optimal cloud infrastructure, such as IaaS. The paper presents an experimental evaluation performed using a cloud application for storing data under different workloads. For the purposes of the experimental evaluation, a Kubernetes cluster composed of 44 computing nodes was used. The cluster nodes were geographically distributed computing infrastructures hosted by several service providers. The proposed approach allows a reliable selection of infrastructures, which satisfy high QoS requirements for cloud applications, from heterogeneous Edge-to-Cloud computing environments. 10.18721/JCSTCS.13101 004.75 cloud computing cloud application deployment Quality of Service Infrastructure as a Service Edge-to-Cloud https://infocom.spbstu.ru/article/2020.64.1/

RAR RUS 19-30 Peter the Great St. Petersburg Polytechnic University Daeef Feras ferasit87@gmail.com

St. Petersburg, Russian Federation

Application of the computer vision system for controlling a mobile robot in a dynamic environment The article proposes a new approach to the use of computer vision when controlling a robot in a dynamic environment. While moving along an unchanged path to the target point, a robot can encounter any new object (static or moving). We describe a visual analysis to determine the detection distance of moving objects to prevent collisions with the robot in a timely manner. An obstacle detection algorithm in the robot zone was developed based on data from an RGB-D video camera using computer vision methods. Based on ROS in a Gazebo virtual environment with a Turtlebot robot kit and an open source library (opencv), we adopted software implementation of the developed approaches which confirmed their applicability to the detection of objects in the mobile robot environment. 10.18721/JCSTCS.13102 004.896 mobile robots dynamic environment navigation vision system object detection https://infocom.spbstu.ru/article/2020.64.2/

RAR RUS 31-41 Peter the Great St.Petersburg Polytechnic University Antonov Alexander antonov@eda-lab.ftk.spbstu.ru

Polytechnicheskaya, 29, St.Petersburg, 195251, Russian Federation

Peter the Great St. Petersburg Polytechnic University Besedin Denis 1310nero@mail.ru

St. Petersburg, Russian Federation

Filippov Aleksey filippov@eda-server.ftk.spbstu.ru Efficiency analysis of high-level synthesis tools for hardware implementation of sorting algorithms The article is devoted to the research of efficiency of Xilinx’s high-level synthesis tools, the Vivado HLS package version 2019.2, for synthesis of a hardware implementation of sorting algorithms. The relevance of creating hardware implementation of sorting algorithms is determined by modern approaches to building high-performance heterogeneous computing systems and modern criteria for the efficiency of such systems – the ratio of performance to power consumption and the ratio of real performance to peak performance. The authors carried out a comparative analysis of the implementation of the selected sorting algorithms on a universal processor and on the basis of the VLSI Xilinx submarine research. The article discusses approaches to optimize the description of algorithms and control the Vivado HLS package to achieve optimal performance of the resulting hardware solutions. The article shows that the main performance gain is provided by parallelizing of the source arrays processing, which is achieved both by the settings of the design tool, the Vivado HLS package, the selected description style, as well as the features of the sorting algorithm selected for hardware implementation. 10.18721/JCSTCS.13103 004 hardware acceleration sorting algorithms high-level synthesis reconfigurable hardware accelerator FPGA https://infocom.spbstu.ru/article/2020.64.3/

RAR RUS 42-52 Peter the Great St. Petersburg Polytechnic University Nguyen Dac Cu daccu91.spb@gmail.com

St. Petersburg, Russian Federation

Zavialov Sergey 0000-0001-7726-8492 Volvenko Sergey Transmission efficiency of multi-frequency signals in MBC using amplitude limitation on the transmitting module The ability to receive electromagnetic waves reflected from meteor traces is the basis for the construction of meteor burst communication (MBC) systems. MBC has a low data rate (about a few dozen kilobits per second). Therefore, from the point of view of increasing the information transfer rate, we can use multi-frequency signals, for example, OFDM signals. These signals occupy a smaller frequency band, but have a high peak-to-average power ratio. In this article, we used simulation modeling to transmit information in an MBC system using multi-frequency signals under the condition of amplitude limitation on the transmitting module. 10.18721/JCSTCS.13104 621.39 meteor burst communications peak-to-average power ratio amplitude limitation MBC OFDM PAPR https://infocom.spbstu.ru/article/2020.64.4/

RAR RUS 53-64 St. Petersburg Institute for Informatics and Automation of RAS Rostova Ekaterina rostovae@mail.ru

St. Petersburg, Russian Federation

Rostov Nikolay Peter the Great St. Petersburg Polytechnic University Yan Zhengjie yanzhengjie1019@gmail.com

St. Petersburg, Russian Federation

Neural network compensation of dynamic errors in a position control system of a robot manipulator This paper considers a position control system of a 3-link robot manipulator. The authors reviewed publications on nonlinear compensation of dynamic errors with the use of neural networks in robot manipulator control systems. The paper presents mathematical description of the control system with the compensation of nonlinear dynamics of the robot mechanism. We carried out training of multivariable neural network compensators of dynamic errors occurring because of the influence of inertia, Coriolis and gravity load torques. We developed computer models of the control system with different types of neural network compensators which are included in feedforward and feedback of the system and carried out a computer simulation of control systems with prototype and different kinds of neural network compensators. We also conducted a comparative analysis of dynamic errors in the system with different combinations of neural network compensators and gave recommendations on program realization of neural network compensators for real robot manipulator position control systems. 10.18721/JCSTCS.13105 004.896 robot manipulator position control system neural networks nonlinear multivariable compensators simulation dynamic errors https://infocom.spbstu.ru/article/2020.64.5/

RAR RUS 65-78 Nesterov Sergey nesterov@saiu.ftk.spbstu.ru Peter the Great St. Petersburg Polytechnic University Smolina Elena smolensk9595@mail.ru

St. Petersburg, Russian Federation

The assessment of the results of a massive open online course using Data Mining methods The paper presents the results of a grade reports analysis for five sessions of a massive open online course “Data Management” at openedu.ru. For our research, we used clustering and classification in the R programming environment. Clustering showed the presence of four groups of course participants with nearly similar course results. These clusters were similar for all five sessions of the course we analyzed. We also showed it is possible to predict whether a participant completes the course or drops out, based on the test results during the first half of the course. The course lecturers can use the results to plan measures for keeping the students in the course. Also, such a type of analysis helps to understand the reasons why the students drop out of the course. The lecturers can take them into account to modify the course structure and learning content. This new knowledge about the course participants can be used during the next course sessions. We expect that for other courses with a similar structure, the clustering results will be also similar. The approach to predict whether a student drops out or completes the course used in the paper is applicable for other courses as well. 10.18721/JCSTCS.13106 004.85, 004.62, 378.147 MOOC learning management systems Data Mining clustering classification https://infocom.spbstu.ru/article/2020.64.6/