Neural network VLSI of pulse operation for reduced power consumption

The neural network application specific VLSI has been preferred in respect of its relatively fast speed, huge network size and effective cost in implementation. Broad ranges of neural network VLSI have been developed for the purpose of implementing the neural network of massive parallel operation. Among those implementations, the principle of pulse-stream operation has made a substantial improvement in speed or size with analog-digital mixed VLSIs. In this work, a way of reducing the power consumption is suggested, which is based on both the synapse circuit using analog MOSFET resistances and the pulse operation principle