350 rub
Journal Electromagnetic Waves and Electronic Systems №9 for 2019 г.
Article in number:
Method for extracting periodic signals from noise by a neural filter
Type of article: scientific article
DOI: 10.18127/j15604128-201909-06
UDC: 621.391.833
Keywords: Radar systems analogous schemes neuroprocessors signal extraction from noise
Authors:

N.L. Dembitskiy – Ph.D. (Eng.), Associate Professor,

Moscow Aviation Institute (National Research University) E-mail: dembitsky@mai.ru

Abstract:

An increase in the requirements for airborne radars necessitates continuous improvement of methods for processing radar information. The relevance of the development of new approaches and breakthrough technologies is due to a contradiction between the promising characteristics of the radar and the extreme capabilities of the well-known technical solutions for processing radar information.

The neuroprocessor model considered in the article is built on analog circuits. The development goal is to achieve high performance hardware recognition of signals by trained neuro-like devices. The developed analog neurofilter is designed to solve the problems of recognizing periodic signals under conditions of high-intensity interference in the megahertz range. The experiments showed the possibility of isolating signals from interference with a speed of several tens of microseconds, which was achieved through parallel processing of information and the use of analog logic in control circuits.

An analog neurofilter can be a competitive alternative to digital signal recognition systems in compact equipment.

Pages: 51-57
References
  1. Boev S.F., Stupin D.D., Savchenko V.P. i dr. Moshchnye nadgorizontnye RLS dal'nego obnaruzheniya. Razrabotka, ispytaniya, funktsionirovanie. Pod red. S.F. Boeva. M.: Radiotekhnika. 2013. (In Russian).
  2. Tatuzov A.L. Neyronnye seti v zadachakh radiolokatsii. Ser. «Neyrokomp'yutery i ikh primenenie». M.: Radiotekhnika. 2009.  (In Russian).
  3. Osovskiy S. Neyronnye seti dlya obrabotki informatsii. M.: Finansy i statistika. 2002. (In Russian).
  4. Kolesnitskiy O.K., Bokotsey I.V., Yaremchuk S.S. Apparatnaya realizatsiya elementov impul'snykh neyronnykh setey s ispol'zovaniem bispin-priborov. Ch. 1. XII Vseros. nauch.-tekhnich. konf. «Neyroinformatika». M.: MIFI. 2010. S. 122–127. (In Russian).
  5. Anisimov A.S., Kalachev A.V. Realizatsiya iskusstvennykh neyronnykh setey na mnogoyadernom protsessore. Elektronnyy zhurnal, eISSN 1684-1719 «Zhurnal Radioelektroniki». 2010. № 9. (In Russian).
  6. Dembitskiy N.L. Analogovye protsessory. M.: MAI. 2018. (In Russian).
  7. Solonina A.I. i dr. Osnovy tsifrovoy obrabotki signalov. Kurs lektsiy. SPb.: BKhV-Peterburg. 2003. (In Russian).
Date of receipt: 11 октября 2019 г.