A.A. Kopylov – Post-graduate Student, 

Moscow Institute of Physics and Technology (State University);

Engineer, PJSC «Radiofizika» (Moscow)

E-mail: kopylov@phystech.edu

E.G. Parinov – Post-graduate Student, 

Moscow Institute of Physics and Technology (State University);

Engineer, PJSC «Radiofizika» (Moscow)

I.V. Zimin – Ph.D. (Eng.), Head of the Laboratory of Radiophysics and Satellite Communication, 

Moscow Institute of Physics and Technology (State University)

Classification of radar objects is one of the most important problem in the development of radar systems. There are many approaches to solving this problem. One of them based on the use of polarizing scattering characteristics of objects obtained using radar systems operating in the full polarization basis. The main advantage of this approach is the ability to obtain additional information that can be used for classification.

To solve the classification problem, we use realizations of polarization scattering matrix obtained for radar objects moving along the trajectory. Realization of PSM is used to calculate various groups of polarization attributes, formatting a set of parameters of the object and comparing it with the standards of the main classes of radar objects. In this case, any classifier known from the classification theory can be selected.

The developed model of the classification process allows us to use it for any polarizing attributes that can be added to the model as separate modules.

The degree of polarization anisotropy and scattering entropy are considered as such polarization parameters in this paper.

Factors that determine the practical use of the polarizing characteristics were considered. For example: movement of radar object on a trajectory, rotation and precession of radar object, the accuracy of the measurement of the PSM, depending on the value of the SNR.

For modeling the classification process the input data for the developed model is realizations of PSM for radar objects of the main classes moving along the trajectory obtained using a special model.

Two cases were considered: the classification of bodies with sizes close to resonant, and the classification of bodies with sizes of the order of 10wavelengths.

The results of the simulation of the classification process for the resonant case and transient case are presented. Results show that the use of scattering entropy makes it possible to obtain a more accurate classification result in the resonant case, and the degree of polarization anisotropy – in the transient case. At the same time, considering the precession of radar objects allows to get a more accurate classification result for each of the parameters.

1. Kanareikin D.B., Pavlov N.F., Potekhin V.A. Polyarizatsiya radiolokatsionnykh signalov. М.: Sovetskoe radio. 1966 (In Russian).
2. Kozlov A.I., Logvin A.I., Sarychev V.A. Polyarizatsiya radiovol. Kn. 2. Radiolokatsionnaya polyametriya. М.: Radiotekhnika. 2007. (In Russian).
3. Claude S.R. Polarisation: Applications in Remote Sensing. Oxford: Oxford University Press. 2009. 352 p.
4. Olyunin N.N., Sazonov V.V. Analiz informativnosti polyarizatsionnykh priznakov v zadache radiolokatsionnogo raspoznavahiya. Sb. dokl. III Vseross. konf. «Radiolokatsiya i radiosvyaz′». М.: IRE im. V.A. Kotel′nikova RAN. 2009 (In Russian). 
5. Fuyou Wang, Rujiang Guo, Yinhe Huang. Radar target recognition based on some invariant properties of the polarization scattering matrix. Proceedings of 2011 IEEE CIE International Conference on Radar. 2011.
6. Parinov E.G., Zimin I.V. Modelirovanie polyarizatsionnykh matrits rasseyaniya ob′ektov nablyudeniya v polnom polyarizatsionnom bazise. Radiotekhnika. 2019. № 4. S. 19-24. DOI: 10.18127/j00338486-201904-02 (In Russian).