350 rub
Journal Neurocomputers №8 for 2016 г.
Article in number:
The analysis of asynchronous neural networks in terms of validity
Authors:
A.N. Kamenskih - Post-graduate Student, Assistant, Faculty of Electrical Engineering, Department of Automation and Telemechanics, Perm National Research Polytechnic University. E-mail: antoshkinoinfo@yandex.ru
Abstract:
The paper contain introduction, two main parts and conclusions. In introduction, the main results from up-to-date scientific works in the field of development of asynchronous neural networks and improvement of reliability of neural networks are described. The technique of validity estimation is need for developers of asynchronous neural networks because the methods of reliability improvement should be chosen very carefully with taking into account all-important parameters. In first part, self-timed approach and self-timed circuits is analyzed in terms of reliability and validity. In second part, validity of operation of asynchronous neural networks is calculated and analyzed. In conclusion, the complex redundancy can be used to improve reliability of asynchronous neural network in some range.
Pages: 32-35
References

 

  1. Yakovlev A. Energy-modulated computing // Design, Automation & Test in Europe Conference & Exhibition (DATE), 2011 // IEEE. 2011. S. 1-6.
  2. Tyurin S., Kharchenko V., Tagarev V. (edits). Redundant Basises for Critical Systems and Infrastructures: General Approach and Variants of Implementation Proceedings of the 1st International Workshop on Critical Infrastructures Safety and Security, Kirovograd, Ukraine 11-13, May, 2011. V. 2. P. 300-307.
  3. Kamenskih A. N., Tyurin S. F. Advanced approach to development of energy-aware and naturally reliable computing systems // Young Researchers in Electrical and Electronic Engineering Conference (EIConRusNW), 2015 IEEE NW Russia // IEEE. 2015. S. 75-77.
  4. Zhou L., Smith S. C., Di J. Radiation Hardened NULL Convention Logic Asynchronous Circuit Design // Journal of Low Power Electronics and Applications. 2015. T. 5. № 4. S. 216-233.
  5. Murray A. F., Smith A. V. W. Asynchronous arithmetic for VLSI neural systems // Electronics Letters. 1987. T. 23. № 12. S. 642-643.
  6. Varshavskijj V.I., Kishinevskijj M.A., Marakhovskijj V.B. Avtomatnoe upravlenie asinkhronnymi processami v EHVM i diskretnykh sistemakh. M.: Nauka. 1986.
  7. Stepchenkov JU. A. i dr. Samosinkhronnyjj vychislitel dlja vysokonadezhnykh primenenijj // V sb. « Problemy razrabotki perspektivnykh mikro- i nanoehlektronnykh sistem». 2010. S. 418-423.
  8. KHarchenko V.S., Lysenko I.V. Nadezhnost, kontrol i diagnostika EHVM. Metodicheskoe posobie. KHAI. 2001. 65 s.