Preview

Informatics

Advanced search

Filtering in the presence of information losses based on the extended least squares method

https://doi.org/10.37661/1816-0301-2022-19-1-50-58

Abstract

O b j e c t i v e s . In radar systems for moving objects tracking, there are often gaps in the measurement of coordinates.

The problem is mostly fully solved in continuous time in the theory of systems with a random structure within the  framework of  statistical  Bayesian theory of  filtration  in  the  presence  of  complete  a  priori  statistical information. This approach leads to complex algorithms that are difficult to implement in practice. The purpose of investigation was to develop a filtering algorithm in conditions of information interruptions based on the use of extended least squares method.

M e t h o d s . Methods of estimation theory are used, in particular, the extended least squares method, which makes it possible to find relatively simple algorithms with a  minimum amount of a  priori knowledge about the characteristics of the impacts.

R e s u l t s . The algorithm for filtering radar signals has been developed, based on measurements of the moments of  breaks  and  extrapolation  of  the  measured  coordinates  at  intervals  of  information  lack.  The  resulting algorithm is nonlinear and therefore tracking disruptions may occur in the filter. The results of the algorithm are demonstrated using a model example. The estimation of the filtering accuracy and tracking failure conditions is carried out.

Co n c l u s i o n . A filtering algorithm has been developed that allows determining the moments of the onset of breaks and extrapolating the estimates of useful information. The comparative simplicity of the algorithm makes it suitable for practical use.

For citations:


Artemiev V.M., Naumov A.O. Filtering in the presence of information losses based on the extended least squares method. Informatics. 2022;19(1):50-58. (In Russ.) https://doi.org/10.37661/1816-0301-2022-19-1-50-58

Views: 433


Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.


ISSN 1816-0301 (Print)
ISSN 2617-6963 (Online)