Estimation of information parameters of the signal on background of additive and multiplicative noise with arbitrary distribution

The issues of quasi-optimal signal processing in conditions of simultaneous effect of additive and multiplicative non-Gaussian noise are considered. The resulting expression describing the effect on the value of the normalized a posteriori errors of additive and multiplicative non-Gaussian noise. It is shown that with given a priori information about probability density function of the multiplicative noise, the accuracy of the measurement of the information parameter of the signal, in the case of its slow fluctuations can be significantly worse than in the case of rapid fluctuations of multiplicative noise.