A one-transistor synapse circuit with an analog LMS adaptive feedback for neural network VLSI

A one-transistor (1T) synapse circuit which uses a single MOS transistor and is more efficient for VLSI implementation of adaptive neural networks than other synapse circuits is presented. This 1T synapse circuit can be used to implement multiply/divide/sum circuits for realizing an adaptive neural network. The feasibility of using this circuit in adaptive neural networks is demonstrated by a 4-b analog-to-digital converter circuit based on the Hopfield modified neural network model with an analog LMS adaptive feedback. DC and transient study shows that the 1T synapse circuits with an analog adaptive feedback circuit can be more efficiently used for VLSI implementation of adaptive neural networks