Researches on GRNN neural network in RF nonlinear systems modeling

Simple introduces the structure of generalized regression neural network, describes the process that using generalized regression neural network (GRNN) modeling to RF nonlinear system, and through the simulation results demonstrate this new nonlinear modeling method of RF devices. The simulation results of the nonlinear RF devices by General regression neural network (GRNN) control the relative forecasting error within in 0.05 The best predict relative error by BP neural network is 0.17, so, compared with BP neural network, GRNN neural network effectively improved the accuracy and precision of the model. In the simulation process by general regression neural network (GRNN) method, to the nonlinear RF power amplifier device, the prediction curve before data normalized more close to the set output curve.