RF MOSFET Modeling Of SMIC 0.13μm Process Based On BSIM4

With the rapid progress of the radio frequency (RF) wireless communications market, especially the fast growth of the internet of things, have induced a great market demand for radio frequency integration circuits (RFICs). And as the continuous scaling down of CMOS technology, the cutoff frequency of MOSFET has exceeded 100 GHz, which makes the CMOS technology a good choice for the design of RFICs. The CMOS RFIC design has been the most popular way of circuit design for its high quality, low cost and high level of integration. However, it is still difficult to design RFIC in CMOS technology, compared with the MOSFET models for both digital and analog application at low frequencies, the models for high frequency application are more difficult to develop due to the bias-dependence of the parasitic and geometry scaling. So it is meaningful to investigate the RF application MOSFET model in CMOS technology.This thesis will carry out the investigation on RF MOSFET modeling on 0.13um CMOS technology. BSIM4 is the model of short-channel IG-FET, which was brought forward by UC, Berkeley in 1990s. It is the continuation of the BSIM3v3, but it has great improvement in the DC, CV, RF and STI stress part. As the CMOS technology feature size continues to reduce, especially at the node of 0.13um and 90nm, the BSIM4 will become more useful.First of all, the thesis reviewed current MOSFET modeling technique and some kinds of dominating MOSFET models. Secondly, a brief introduction of the equations of BSIM4 threshold voltage model, the mobility model, leakage current model, gate tunneling current model, substrate current model and the eigenvalue capacitance model and external capacitor is presented. Then, the design of the layout for RF MOSFET in SMIC 0.13um CMOS technology, the extraction of the model parameters and the test plan are introduced, good agreement is achieved between the simulated and measured results, followed by a brief introduction of the STI stress influence on the device, the BSIM4 STI stress model. Based on the equations, the stress parameters extracted and analysis. Finally, the extraction of the BSIM4 RF model is made, including the intrinsic and the extrinsic parameters, the results are verified by the measured data from 50 MHz to 20.05 GHz.