EXTRACTION OF A COMPLEX TREND OF SIGNALS IN SYSTEMS OF INTELLIGENT SUPPORT OF THE MOVEMENT OF OBJECTS

Abstract

The article is devoted to the study of problems of intellectualization of control systems of moving objects to ensure their reliable operation in difficult conditions in various physical environments. Intelligent traffic support systems are developed based on the integration of measurement, computing, communication and control technologies and are designed to ensure the collection, accumulation and processing of information, determination of navigation parameters, formation of control influences, visualization of the positioning and movement trajectories of objects, monitoring of thefunctional state of objects and the state of the operating environment, etc. When restoring the lost trajectory of a moving object in the conditions of its complex movement, when approximating complex algorithms in adaptive control systems, with sudden changes in the trajectory or conditions of movement, there is a need to extract a complex trend from noiselike signals. The existing filtering algorithms do not provide acceptable efficiency, especially in conditions of limited a priori information about the nature of changes in the useful signal component.The article proposes and substantiates the use of wavelet filtering to suppress interference and extract the initial signal of a complex shape. A study of the effectiveness of extracting a complex useful signal against the background of Gaussian noise and harmonic disturbances was conducted. To process the simulated signals, wave functions of various orders from the families of Dobechies, symlets, and coiflets were used. The effectiveness of wavelet filtering was evaluated by the mean square deviation value of the selected trend and the model of the useful signal. The comparative analysis of the obtained results showed the feasibility of using the wave function of the family of symlets, which ensured the minimum value of the root mean square deviation of the selected trend from the model of the useful signal.