HARDWARE IMPLEMENTATION OF NEW ANALOG MEMORY FOR NEURAL NETWORKS

We have fabricated a new analog memory with a floating gate as a key c omponent to store synaptic weights for integrated artificial neural ne tworks. The new analog memory comprises a tunnel junction (poly-Si/pol y-Si oxide/poly-Si sandwich structure), a thin-film transistor, two ca pacitors, and a floating gate MOSFET. The diffusion of the charges inj ected through the tunnel junction is controlled by switching operation of the thin-film transistor, and we refer to the new analog memory as switched diffusion analog memory (SDAM). The obtained characteristics of SDAM are a fast switching speed and an improved linearity between the potential of the floating gate and the number of pulse inputs. SDA M can be used in a neural network in which write/erase and read operat ions are performed simultaneously.