ADAPTIVE DETECTION ALGORITHMS FOR SIGNAL REFLECTED FROM SMALL-SIZED GROUND TARGETS

Abstract

Background. Solving the problems of detection, selection and recognition of small ground targets (MNTs) (vehicles of various classes, lightly armored targets – armored infantry vehicles and armored personnel carriers, heavily armored targets – tanks) is carried out both by optical, optoelectronic, thermal imaging, and radar reconnaissance means. The most promising means of reconnaissance, which have the properties of all-weather and all-day operation, are radar facilities and integrated systems. The use of differences in the fine structure of the reflected signals makes it possible to form polarization, Doppler, rangefinder, angle-frequency, space-time and other combined radar portraits of targets, which are the basis for solving the problem of detection and recognition of MNCs in the presence of means of reducing visibility and means of electronic protection, as well as the presence powerful interference reflections from local objects. The solution to the problem of selection of stationary and slow-moving MNCs is possible only with the use of polarization and polarization-rangefinder portraits and adaptive signal detection algorithms created on their basis. Materials and methods. The article proposes to analyze the integral distribution functions for the longitudinal-angular and transverse-angular oscillations of the MNC, presented as a set of shiny points relative to the center of mass and moving along an uneven surface. To reveal what differences in the spectral characteristics of these oscillations for cars and armored vehicles are the most critical. Determine the algorithms for adaptive detection of a signal reflected from the considered radar targets using sample values of the spectrum of the elements of the scattering matrix and evaluate their effectiveness. Results and conclusions. Thus, algorithms for adaptive detection of a signal reflected from small targets based on sample values of the spectrum of the elements of the scattering matrix are defined, which reduce the correlation of interference samples at the output of the processor and make it possible to find decision rules in explicit form and obtain equations for estimating the maximum likelihood of unknown signal and noise parameters. vectors.

Keywords

• spectral characteristic
• objects of armored vehicles
• target
• algorithm