linear non-stationary filtration
In this article the author demonstrates the need for development of novel subclass of the poly-Gaussian models based on the analysis of current models for description of the signals transduction environment in wideband systems. The above mentioned «Multi-Markov poly-Gaussian models» can be characterised by the following basic properties:
The basic components mode is discrete-continuous multidimensional Markov’s process;
A large set of such processes forms a mix that is based on the parameters of the Markov’s process as well as on the variants of imposing signals;
Numbers of the realised combinations of a mix form a Markov’s chain.
In addition the articles presents basic explanations for the description of the signals transduction environment within the limits of the offered model, which is subsequently used for the synthesis of the algorithm for reception of wideband signals.
Estimation of working capacity and efficiency of the synthesised algorithm is made for real signals of a communication network of standard CDMA One, by means of a programs complex developed in the Lazarus programming environment.
On the basis of the analysis of the results obtained the following conclusions have been drawn:
1. Non linear algorithm synthesised on the basis of Multi-Markov poly-Gaussian models of decision-making enables adequate description of the existing in channel CDMA One signals transduction environment at the lowered fading’s level.
2. The algorithm enables decrease in probability of an error, a dispersion of an estimated signal’s parameters and residual noise in comparison with classical algorithm.
3. Estimations obtained with the help of the algorithm enable algorithm’ performance in the conditions of aprioristic uncertainty