A detecting method of analog signals based on variable threshold value neuron
Bo Sun, Jingfu Lin, Yong Chen · 2008
The paper puts forward a variable threshold value artificial neuron structure. The neural function between the output and input can be accurately trained by increasing the density of threshold value. 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.