A multiple-input OTA circuit for neural networks

An operational transconductance amplifier (OTA) circuit suitable for modeling neurons in VLSI implementations of artificial neural networks (NN) is described. It generates an output voltage which is a sigmoidal-like function of the linear sum of a number of weighted inputs. The weight of each input is individually controlled by a bias voltage which can be varied continuously and dynamically. Large numbers of these cells can be fashioned in regular arrays. It appears to be efficient because each weighted connection is implemented with only two MOSFET transistors.<>