The nanoscale MOSFET has a short channel effect, which increases the standby power by increasing leakage current, and has difficulty in controlling threshold voltage owing to the gate field influence becoming weak in a nanoscale channel. Recently, many problems occurring in designing next-generation integrated circuits (ICs) have been emerging, owing to continuous scaling for a metal-oxide-semiconductor field-effect transistor (MOSFET). A hybrid inverter and an integrate and fire (I&F) circuit for a spiking neural network, which consisted of NMOSFETs and an NFBFET, were simulated using the circuit simulator to verify a validation of the proposed NFBFET macro-model. The proposed model implements not only the I DS- V GS characteristics but also the I DS- V DS characteristics. The parameters of the NMOSFET and diode used in this proposed model were fitted from TCAD data of the NFBFET, divided into two parts. To solve this problem, we connected a physics-based diode model with an ideal switch in series to the current generator circuit.

The previous current generator circuit could implement I DS- V GS characteristics but could not accurately implement I DS- V DS characteristics. For the previous model, the current generator circuit consisted of one ideal switch and one resistor. This circuit implements the charging characteristics of NFBFET, which occur in the channel region.

The charge integrator circuit consisted of one N-type metal-oxide-semiconductor field-effect transistor (NMOSFET), one capacitor, and one resistor. One is a charge integrator circuit and the other is a current generator circuit. The macro-model of the NFBFET is configured into two parts. In this study, we propose an improved macro-model of an N-type feedback field-effect transistor (NFBFET) and compare it with a previous macro-model for circuit simulation.

Types of transistor download free