Analysis and Detection of Nonlinear Analogue Based on Variable Threshold Value Neuron

In this paper, a variable threshold value artificial neuron structure is put forward. On the basis, piecewise linearization and piecewise variable slope are used to train the detecting methods of non-linear analogue. Combined with characteristic of distributed control systems, a long-distance intelligent marking method is proposed and is applied to carry out the process of training threshold value and weight coefficient. The method is prone to detect analog signals fast and precise. A timing duplicate marking method is presented to ensure the reliability of data transfer. The simulation of the model is carried out, simulation results of nonlinear function are provided.