Neuron architecture based on jamming avoidance response of an electric fish

The authors propose a digital simulation system of a jamming avoidance response model (JAR) to obtain a robust system of phase information processing. The JAR can be observed in an actual neuron system of an electric fish. In the model, a new digital neuron is first proposed for realizing a processing mechanism of both power amplitude and phase information. It is also shown that the designed network can realize the behavior of JAR even when data are observed under condition of low SNR.<<ETX>>