Kulik Y. Methods and devices for processing digital signals in networks with high noise

The object of research - the process of processing of received digital signal in distributed computer networks with high levels of noise. The goal is to improve the processed data in distributed networks with high levels of. Research methods. Were used computer methods, and circuit simulation, probability theory, mathematical statistics, spectral transformation, wavelet transform and filtering of signals, probabilistic analysis, numerical methods using environments Matlab, Maple, LabView. Scientific novelty. Was developed method for determining the coefficients of expansion, using replacement biortohonal coefficients of wavelet decomposition approximating sum using a number of quasi-positive delta-sequences ; Improved methods of evaluation of the signal for the adaptive threshold method which uses a multi-averaged estimate of the logarithmic spectrum signal; proposed the method of fast median filtering, which operates with a limited amount of data by splitting vector data into multiple data vectors; improved method of parallel fast wavelet transform that uses partitioning vector data into blocks of data to be processed using local wavelet transform in diagonal sequence. The practical significance of the results. Carried researches allowed increase the range in the transmission of information of 30 dB to 10 dB. Obtained in the research results are implemented in the form of algorithms, software and block diagrams. The developed software that allows you to perform data processed using fast median filtering. The flow diagram for the device to transmit information using fast median filtering and parallel fast wavelet transform. The degree of implementation - the results of the thesis are introduced in the form of software in information systems for measuring PE "Anklyuz" PE "TBA-Servis" Vinnitsa Chamber of Commerce "with the use of fast median filtering, which helped to improve the quality of communication CCTV. Scope (area) of using - communication, signal processing, information-measuring system.