The method of object type recognition in aerospace space on the basis of analysis of radio technical characteristics taking into account the informative character of the signs

Аuthors

Aldohina V. N.¹, Demyanov A. V.¹, Gudaev R. A.^1*, Byk V. S.¹, Vikulova Y. M.^2**

1. Mlitary spaсe Aсademy named after A.F. Mozhaisky, Saint Petersburg, Russia
2. Group of Companies «RTI», 10/1, 8 Marta str., Moscow, 127083, Russia

*e-mail: mrgudaev@mail.ru
**e-mail: yvikulova@oaorti.ru

Abstract

The aim of the work is to increase the probability of correct recognition of the type of object in the airspace based on the analysis of the radio technical characteristics of emissions, taking into account the informativeness of the recognition features to a level not lower than required by the regulatory documents.

To achieve this goal, we propose a method of recognition using the principle of multichannel with discrete accumulation and using weight coefficients calculated on the basis of the entropy approach. Based on the analysis of the data presented, an increase in the number of characters makes it possible to increase the probability of correct recognition, and the application of weighting coefficients gives, albeit a small, but guaranteed increase in the indicator under consideration.

The scientific novelty of the method of recognizing the types of radiating objects on the basis of a set of radio engineering features and weighting coefficients is that it differs from existing ones in that the determination of the type of the radiating object is made on the basis of comparing the additive convolution of particular solutions multiplied by the corresponding weight coefficients reflecting their informativeness with Threshold calculated for the current recognition situation.

The theoretical significance of the developed method for recognizing the types of radiating objects on the basis of a set of radio engineering features and weighting coefficients is that it is a development of the theory of pattern recognition and differs from existing ones in that the definition of the type of the observed object is made on the basis of a comparison of the additive convolution of particular solutions multiplied by the corresponding weighted ones Coefficients reflecting their informativeness, with a threshold calculated for the current recognition situation.

The practical significance of the method for recognizing the types of radiating objects on the basis of a set of radio technical features and weighting factors is to improve the methodological and algorithmic support of the space monitoring system aimed at providing the required probability of correct recognition.

The result of this method implementation will be a decision on the recognition of the type of radiating objects based on the analysis of radio technical information taking into account its informativeness. The considered method of recognition of types of radiating objects allows:

— to make the recognition of the types of radiating objects on the basis of radio technical information;

— expand the characteristic space used to describe types of recognized radiating objects;

— determine the informativeness of the features of the recognition of the types of radiating objects;

— minimize the error in recognizing the types of radiating objects;

— Reduce the computational cost of conducting recognition of the types of radiating objects.

Keywords:

detection, signs of recognition, entropy

References

1. Gorelik A.L., Skripkin V.A. Metody raspoznavaniya (Recognition methods), Moscow, Vysshaya shkola, 1984, 208 p.

2. Chebotarev A.S., Zhukov A.O., Makhnenko Yu.Yu., Turlov Z.N. Monitoring kosmicheskikh apparatov na osnove primeneniya korrelyatsionno-fazovykh pelengatorov (Monitoring of spacecraft on the basis of correlation-phase), Moscow, Fizmatlit, 2011, 120 p.

3. Tou J.T., Gonzalez R.C. Pattern Recognition Principles. London-Amsterdam-Dom Mills, Ontario-Sydney-Tokyo. Addison-Wesley Publishing Company. 1974. 378 p.

4. Maiorov D.A., Perekhozhev V.A., Shemyakov A.O. Trudy MAI, 2013, no. 71, available at: http://www.mai.ru/science/trudy/eng/published.php?ID=47077

5. Kulikov S.V., Gudaev R.A., Baldychev M.T., Gaichuk Yu.N. Naukoemkie tekhnologii, 2015, no. 12, pp. 26-30.

6. Chernavskii D.S., Karp V.P., Rodshtat I.V., Nikitin A.P., Chernavskaya N.M. Raspoznavanie. Autodiagnostika. Myshlenie. Sinergetika i nauka o cheloveke (Recognition. Autodiagnosis. Thinking. Synergetics and the science of man), Moscow, Radiotekhnika, 2004, 272 p.

7. Fedotov N.G. Teoriya priznakov raspoznavaniya obrazov na osnove stokhasticheskoi geometrii i funktsional’nogo analiza (Theory of features of pattern recognition based on stochastic geometry and functional analysis), Moscow, Fizmatlit, 2010, 304 p.

8. Rogov D.A., Katyukha R.V., Gudaev R.A., Logunov S.V., Gudaev R.A. Trudy Voenno-kosmicheskoi akademii imeni A.F. Mozhaiskogo, 2015, no. 647, pp. 137-142.

9. Gudaev R.A. Yubileinaya 70-ya Vserossiiskaya nauchno-tekhnicheskaya konferentsiya, posvyashchennoi Dnyu radio. Sbornik tezisov, St.-Petersburg, 2015, pp. 74-75.

Download