S.G. Gurzhin - Ph.D.(Eng.), Associate Professor, Department «Information-Measuring and Biomedical Engineering», Ryazan State Radio Engineering University, Winner of the Lenin Komsomol Prize in Science and Technology (1984.), Laureate of the Ryazan Region on Science and Technology and the Silver Medal Academician V.F. Utkin
A.A. Dunaev - Dr.Sc.(Eng.), Professor, Department «Computer science and computer engineering», Ryazan State University n.a. S.A. Esenina
L.P. Zhalnenkova - teacher of mathematics and informatics, Municipal Budget Educational Institution «Secondary comprehensive school № 7», c. Kasimov

Improving the efficiency of the operational methods and integrated correlation- spectral analysis of bioelectric signals acquires special relevance in connection with the growing need for objective, accurate, rapid control of the parameters of human functional state during various operations during physiotherapy sessions, as well as for fast and accurate setting diagnosis for emergency assistance. Speed and complexity of algorithms for microprocessor of diagnostic systems is mainly determined by duration, number and complexity of computing operations correlation and spectral analysis. Existing analysis algorithms using the classical methods such as the method of multiplication, when assessing the normalized correlation function and subsequent cosine transform it or multiple fast Fourier transform (FFT), have a rather complex structure and contain the implementation of a large number of arithmetic operations and other auxiliary character, such as memorization, rearrangement, etc. In most cases, the major factor limiting the feasibility of complex algorithms and measuring systems operating in real-time computing capabilities are used modules and microprocessors. One of the promising directions in the field of synthesis algorithms and measurement systems with minimized structure for correlation and spectral analysis are methods that use a small number of quantization levels, interpolation functions, probabilistic methods, polar, conditional mean, the conditional mean sign function correlation filter, moving autoregression various modifications of the method of weighted conditional mean. Among modified only method of weighted average conditional contains substantially less arithmetic and auxiliary operations than classical methods have. Correlation analysis on the weighted average conditional uses information about conditional values of processes that receive and assess other statistical characteristics, such as one-dimensional, multi-dimensional, conditional probability density conditional expectation, conditional variance, etc. Therefore, the proposed method of correlation analysis on interval conditional mean algorithmically and structurally compatible with the computational procedures evaluate a large number of other statistical characteristics, which makes it easily integrated into multi-functional algorithms and measuring systems.