CMOS PWM VLSI implementation of neural network
Chen Lü, Bingxue Shi · 2000
A neural network VLSI implementation based on pulse width modulation is analyzed. A simple synapse multiplier is proposed, which has high precision and large linearity range. A voltage-mode sigmoid circuit with adjustable gain is analyzed, which is used for neuron activation functions. A voltage-pulse conversion circuit required for PWM is suggested, which has high conversion precision and linearity. On the basis of the above circuits, a PWM VLSI neural network to solve XOR problem is designed. The simulation result shows its correct function and fast speed, it is suitable for VLSI implementation of neural network.