Методи статистичного оцінювання параметрів сигналу на фоні негаусових корельованих завад

• Дата: 2021
• Автори: Смірнов, Данііл, Вєдєрніков, Дмитро, Палагіна, Олена, Палагін, Володимир
• Формат: Стаття
• Опубліковано: Kamianets-Podilskyi National Ivan Ohiienko University 2021
• Онлайн доступ: http://mcm-tech.kpnu.edu.ua/article/view/251088
• Назва журналу: Mathematical and computer modelling. Series: Technical sciences

The traditional approach to the development of systems for signal parameters estimation in non-Gaussian noise is characterized by significant difficulties associated with the complexity of algorithmic implementation, which makes it impossible to synthesize high-quality software-algorithmic and hardware statistical signal processing. At the same time, the presence of a statistical relationship between the studied sample non-Gaussian random variables leads to a significant complication of the implementation of computational algorithms. Analysis of scientific research in recent years has shown that to solve problems of estimating unknown parameters of signals in non-Gaussian noise there is another promising approach, which is based on the use of numerical characteristics to describe random processes, namely moment and cumulant functions of higher orders. This allows us to describe the statistical properties of the studied non-Gaussian processes with the necessary approximation. The paper proposes new mathematical models of additive interaction of useful signal and correlated non-Gaussian noise based on the use of one-moment and two-moment cumulative functions of higher orders, which made it possible to describe the parameters and characteristics of non-Gaussian distribution of the studied random process and take into account correlations for the synthesis of algorithms for estimating unknown parameters. Based on the obtained moments and cumulant models of random correlated non-Gaussian processes, polynomial stochastic methods for estimation an unknown parameter of a constant signal for dependent sample values are proposed. This allowed the synthesis of computational algorithms for processing non-Gaussian correlated processes. Based on the proposed methods, the synthesis and analysis of polynomial computational algorithms for the parameter estimation of the useful signal with better accuracy characteristics in the form of reducing the variance of the estimate compared to the known results due to additional information about the studied processes in the form of moment and cumulant functions.