Радиофизика и электроника (Mar 2018)
ESTIMATION OF MATCHING OF OPTIMAL DYNAMIC SPECTRAL FILTRATION IN ELECTRO-OPTICAL SYSTEM OF TARGET DETECTION
Abstract
The article is devoted to the study of the properties of the electro-optical systems with matched spectral dynamic filtration. It is usually supposed that the optimal optical signal detector is synthesized under the assumption of a priori information about the statistical characteristics of the target and background signals and the equality of the target and the background correlation matrices. However, in practice, there is always a mismatch between the input and the reference (expected) optical signals. The purpose of this article is to generalize and develop the methods for estimation of matching of the optimal signal processing in electro-optical systems with dynamic spectral filtration, which makes it possible to study the effect of the difference between the input and reference signals on the processing quality. In this paper, the information criterion – the normalized Kullbak–Leibler divergence, which is the ratio of the divergences at the output and input of the dynamic spectral filter – is used as a measure of matching of the optimal signal processing in electro-optical systems with dynamic spectral filtration. The analysis of the properties of the suggested information matching indicator and its comparison with the probability of correct detection is carried out. A mathematical model of the process of optimal dynamic spectral filtration in electro-optical systems is constructed. The examples show the extent to which the differences in the statistical characteristics of input and reference signals affect the information matching indicator. In particular, the article poses and solves three problems that illustrate the process of optimal spectral filtration in the following situations: 1) the statistical properties of the input and the reference signals coincide completely; 2) the mean brightness of the optical radiation of the target and the background at the input of the electro-optical system has changed; 3) the statistical characteristics of the background input signal differ from the reference values.
Keywords
