{"title":"Performance Comparison of FPGA-based Convolutional Neural Networks by Internal Representations","authors":"Marsel I. Iamaev, S. P. Shipitsin","doi":"10.1109/EIConRus49466.2020.9039366","DOIUrl":"https://doi.org/10.1109/EIConRus49466.2020.9039366","url":null,"abstract":"Reconfigurable Field-Programmable Gate Arrays (FPGAs) have prospects for applying in mobile and wearable electronics. FPGA-based neural networks have strong advantage in energy consumption comparing to another solutions. For further improving of their energy efficiency it is appropriate to study the individual network parameters effect on the entire system performance. By the reason, different internal representations variants of convolutional neural network (CNN) were compared and investigated. The study involves an accuracy parameters analysis with restricted memory for weights and increasing the network depth. Binary parameters were chosen for FPGA implementation as more efficient. Binarized CNN was compared with equal CNN by memory comsuption of weights. In addition, the mathematical problem statement of realizing binarized neural network is considered.","PeriodicalId":333365,"journal":{"name":"2020 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering (EIConRus)","volume":"101 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114700817","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Organization of Managing the State of Medium Parameters Uncertainty Based on the Concepts of Entropy Potentials","authors":"V. L. Lazarev, D. Volkova","doi":"10.1109/EIConRus49466.2020.9039214","DOIUrl":"https://doi.org/10.1109/EIConRus49466.2020.9039214","url":null,"abstract":"The present study considers the approaches to development of algorithms and technologies which provide the opportunities to manage the state of medium parameters uncertainty. The study is based on the concepts of entropy potential and integrated entropy potential. Proposed solutions are characterized by both simplicity and convenience for practical use. The study is illustrated with examples of proposed solutions implementation in specific situations. The usage of recommended developments will help to elevate the organizational efficiency of managing the state of uncertainty, which occurs in various nature systems.","PeriodicalId":333365,"journal":{"name":"2020 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering (EIConRus)","volume":"99 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124093381","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Transaction Protection in Corporate Networks Based on Distributed Ledger Technology","authors":"N. Kuchin, Kirill O. Polyakov, N. G. Butakova","doi":"10.1109/EIConRus49466.2020.9039210","DOIUrl":"https://doi.org/10.1109/EIConRus49466.2020.9039210","url":null,"abstract":"In this article, we describe the use of distributed ledger technology based on blockchain technology without mining to protect various kinds of transactions in corporate networks. The relevance of this topic comes from the rapid developments in the digitalization of the economy and other spheres of human activity. The digital economy has taken a strong position worldwide. Data are processed and transmitted electronically. At the same time, it becomes vulnerable to attacks of criminals and competitors during transfer and storage processes.The article proposes an approach to the protection of corporate transactions based on a distributed ledger with elements of blockchain technology in comparison with traditional methods.","PeriodicalId":333365,"journal":{"name":"2020 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering (EIConRus)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126445590","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
D. Ryabkin, Nadezhda A. Taricyna, A. A. Nekhaeva, E. Morozova
{"title":"Real-Time Measurement of Biological Tissue Thermal Diffusivity During Laser Welding","authors":"D. Ryabkin, Nadezhda A. Taricyna, A. A. Nekhaeva, E. Morozova","doi":"10.1109/EIConRus49466.2020.9039356","DOIUrl":"https://doi.org/10.1109/EIConRus49466.2020.9039356","url":null,"abstract":"The article proposes a new method for measuring the thermal diffusivity coefficient in real time during laser welding of biological tissues. Thermal diffusivity coefficients for skin and heart tissue of pigs and chicken experimentally identified.","PeriodicalId":333365,"journal":{"name":"2020 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering (EIConRus)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126558782","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Distributed Linux Build System for Elbrus Hardware Platform","authors":"E. Chernikova, Mikhail A. Shalaev","doi":"10.1109/EIConRus49466.2020.9039306","DOIUrl":"https://doi.org/10.1109/EIConRus49466.2020.9039306","url":null,"abstract":"Modern Linux distributions contain thousands of packages, thus, the rebuilding process of all of them is a timeconsuming task. For example, in some cases, it can take a few days of continuous work. The article discusses issue of minimizing the build time of the operating system distribution. In particular, it concentrates on the development of the distributed build system for Elbrus OS. This paper presents an approach based on package level parallelism and hybrid planning. The approach combines a directed acyclic graph model of package dependencies management and task queue system. The experimental results have confirmed that the use of the proposed distributed build system reduces build time significantly.","PeriodicalId":333365,"journal":{"name":"2020 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering (EIConRus)","volume":"2015 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128042961","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
D. Imaev, M. L. Nemudruk, M. S. Fedorov, I. I. Shpakovskaya
{"title":"Diagnostics of the Cavitation Modes in the Heavy Petroleum Products Pumping Systems","authors":"D. Imaev, M. L. Nemudruk, M. S. Fedorov, I. I. Shpakovskaya","doi":"10.1109/EIConRus49466.2020.9039411","DOIUrl":"https://doi.org/10.1109/EIConRus49466.2020.9039411","url":null,"abstract":"The article formalizes the problem of registering the cavitation’s modes of screw pumps. Possible approaches to solving the problem are given. The \"averaged operator\" algorithm is proposed according to the operator in the form of production rules of the type \"IF-THEN\". In this article propose using of a neural network and cluster analysis to determine cavitation modes.","PeriodicalId":333365,"journal":{"name":"2020 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering (EIConRus)","volume":"124 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128163910","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Multidimensional Matrix Algebra Versus Tensor Algebra or μ > 0","authors":"E. Goncharov, P. Iljin, V. Munerman","doi":"10.1109/EIConRus49466.2020.9039478","DOIUrl":"https://doi.org/10.1109/EIConRus49466.2020.9039478","url":null,"abstract":"The paper compares of the multidimensional matrix algebra and the tensor algebra. It is shown that tensor algebra operations are realized in the multidimensional matrix algebra more efficiently. Examples are given to illustrate this fact. It is concluded that the multidimensional matrix algebra is a natural generalization of the tensor algebra. It is shown how the operation of the multiplication of multidimensional matrices is parallelized. It is said which data processing tasks are effectively formalized and implemented in the multidimensional matrix algebra. The paper was written in order to make the algebra of multidimensional matrices accessible to many software developers.","PeriodicalId":333365,"journal":{"name":"2020 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering (EIConRus)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128167497","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Fearures of the Mathematical Modeling of Frequency Converters with Active Rectifiers for Power Quality Analysis in Internal Power Supply Systems","authors":"A. Nikolaev, I. Gilemov, L. Antropova","doi":"10.1109/EIConRus49466.2020.9038988","DOIUrl":"https://doi.org/10.1109/EIConRus49466.2020.9038988","url":null,"abstract":"The work presents a comparative assessment of two models of a frequency converter with an active rectifier from the position of their influence on the supply network. In the first model, the frequency converter was fully simulated, in the second - an inverter and the synchronous motor were replaced by a current source. For research, two simulation models of a frequency converter with a three-level active rectifier, made according to a 12-pulse circuit, were developed and implemented in the Matlab-Simulink program. The harmonic composition of the consumed current and mains voltage at the connection point of the frequency converter is investigated. The conclusion is drawn about the possibility of simplified modeling of a frequency converter with an active rectifier analyzing the power quality in internal power supply systems of industrial enterprises.","PeriodicalId":333365,"journal":{"name":"2020 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering (EIConRus)","volume":"150 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125957555","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Diagnostic Modeling of a Networks of Digital State Machines","authors":"I. Bystrova, B. Podkopaev","doi":"10.1109/EIConRus49466.2020.9039535","DOIUrl":"https://doi.org/10.1109/EIConRus49466.2020.9039535","url":null,"abstract":"the problem of synthesizing a diagnostic model for a network combining a number of digital state machines (automations) is considered. It is assumed that the diagnostic models of all network components are known preliminarily and they are given by systems of logical equations. The possibility of converting the combination of initial diagnostic models providing fault detection of network components is shown, provided that the detected fault is localized in a single network component. A search procedure of analytical equations determining a diagnostic model for the whole network is proposed. It is shown that the fault detecting ability of that diagnostic network model is preserved while computational complexity and hardware redundancy compared to the initial set of diagnostic models of network components are reduced.","PeriodicalId":333365,"journal":{"name":"2020 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering (EIConRus)","volume":"159 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127931002","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Data Mining Algorithms for Solving Classification Problems","authors":"Zayar Aung, I. S. Mikhaylov, Ye Thu Aung","doi":"10.1109/EIConRus49466.2020.9039386","DOIUrl":"https://doi.org/10.1109/EIConRus49466.2020.9039386","url":null,"abstract":"The purpose of the work is to study the data mining existing approaches to solve the forecasting situations in oil wells problem based on their work parameters accumulated values. The various algorithms functioning were analyzed. From the data presented, it can be seen that the greatest number of correct and accurate classifications of situations is obtained using the machine learning method support vector machine. This algorithm can be used as a basic algorithm for modification in the case of multiple classification in further studies. This program will help the engineer to intervene in the production process in time and prevent high production costs.","PeriodicalId":333365,"journal":{"name":"2020 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering (EIConRus)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121762925","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}