Operator approach to nonlinear compensator synthesys for communication systems

The recent decade trend is in the increase in simulating tasks complication and requirements to its accuracy with the lack of inner structure and system parameters information simultaneously. That's why the dynamic system modeling approach based on the input/output relation seems to be perspective. The approximation of a nonlinear operator mapping the set of input signals into the set of output signals, i.e. the mathematical model construction of a dynamic system on the basis of the input/output relation, is a considerable problem. There are various shapes and structures of nonlinear models, such as Volterra series, split signal polynomials, regression models and different types of neural networks. The operator approach application is extended to nonlinear compensator synthesis for suppression of nonlinear distortions in communicational channels. The nonlinear compensators are designed on the basis of Volterra series and some neural networks for Wiener model of communicational system. A comparative analysis of synthesized nonlinear compensators is represented.