Chie-In Lee

Modeling Inductive Behavior of MOSFET Scattering Parameter

A novel physical small-signal equivalent circuit for accurately modeling an unusual phenomenon of inductive in the breakdown regime of RF metal-oxide semiconductor field-effect transistors is presented for the first time. To remove the low-frequency dispersion of the drain-to-source resistance extracted by a conventional approach, a new extraction method of equivalent circuit element values with the introduction of an inductive network is demonstrated in this paper. Excellent agreement between simulated and experimental data is obtained up to 26.5 GHz in the breakdown region. Therefore, this proposed physical model based on the avalanche breakdown mechanism can accurately be used to predict the RF circuit performance when impact ionization occurs.

S _{22}

In this letter, a noise deembedding method is presented to improve accuracy of on-wafer noise measurements of metal-oxide-semiconductor field-effect transistors. This method further removes parasitic effects of the forward coupling and the dangling leg with transverse magnetic mode considered, which is different from the transverse electromagnetic mode of input/output interconnects. Compared with the conventional scalable noise deembedding methods, this modified method can remove these parasitic noise effects and then make the intrinsic parameters of a device more accurate for noise characterization. The presented method demonstrates better deembedded noise performance especially for 10.4% reduction of deembedded minimum noise figure from 1 to 18 GHz.

in the Breakdown Regime

Abstract The channel resistance from the triode to impact ionization region is determined accurately with the inductive breakdown network considered for different channel length metal–oxide–semiconductor field–effect transistors (MOSFETs). The radio frequency (RF) drain breakdown effect (DBE) and substrate current induced body effect (SCBE) of the MOSFETs in the impact ionization region are distinguished to reveal the channel length dependent channel resistance and breakdown network. The deviation between channel resistance with and without considering the inductive breakdown network increases with reducing channel length due to more significant RF DBE. The presented analysis results can be beneficial to the reliability investigation for RF performance of the MOSFETs in the breakdown region.

An Improved Cascade-Based Noise Deembedding Method for On-Wafer Noise Parameter Measurements

Abstract An improved large-signal breakdown model establishment applying an artificial neural network (ANN) approach is presented for metal-oxide-semiconductor field-effect transistors (MOSFETs) operating in the breakdown regime for the first time. A neural network program with the feed-forward back propagation algorithm and Levenberg Marquardt optimization is utilized to obtain the MOSFET large-signal model for breakdown operation. When compared with the conventional model without the breakdown effects considered, more accurate large-signal characteristics in the breakdown regime can be obtained by incorporating the avalanche network using the ANN approach. The breakdown network is demonstrated to be significant for large-signal characterization at high bias according to the analysis of minimum acceptable error. Besides, the divergence problem due to the neglected avalanche network can be avoided during ANN training in the avalanche regime. The accuracy of the presented model can keep about 2% error. The presented ANN approach can be applied to device large-signal modeling in the impact ionization regime.

MOSFET channel resistance characterization from the triode region to impact ionization region with the inductive breakdown network

In this paper, a simple and fast de-embedding procedure suitable for the X-parameter measurement is presented. The X-parameters of the dummy structures including open and short are measured and then directly subtracted from the device under test (DUT) to remove parasitic effects. Compared with the conventional procedure, this approach can provide a fast and simple algorithm based on the matrix transformation to implement DUT de-embedding for the X-parameter measurement. The presented X-parameter de-embedding procedure can be applied to radio-frequency (RF) transistor characterization in the large-signal operating region.

An improved large-signal model by artificial neural network for the MOSFETs operating in the breakdown region

In this paper, reduction of intermodulation distortion (IMD) due to inductive behavior for radio frequency (RF) MOSFETs is explored using Volterra series based on a nonlinear model incorporating a physical inductive breakdown network for the first time. The calculated total magnitude of high-order nonlinearities is lower than individual results from a nonlinear transconductance and breakdown inductance. In addition, the analysis result shows that the phase difference between the nonlinear transconductance and the breakdown inductance is almost π. Cancellation between resistive nonlinearity and reactive nonlinearity from the inductance due to the avalanche delay is first reported. The input third-order intercept point (IIP 3) is improved at biases where the breakdown inductance nonlinearity dominates. Instead, linearity will decrease when the breakdown resistance dominates.

A simple and fast de-embedding procedure of X-parameter measurement for RF transistor characterization in the large-signal operating region

In this paper, an improved four-port radio-frequency (RF) model for metal-oxide-semiconductor field-effect transistors is extended to investigate the various RF breakdown phenomena where reliability is a concern for the first time. Reduction of isolation from drain to source as well as inductive behavior occurred at the drain, and source terminals in the breakdown region are analyzed and explained by using this modified four-port RF model incorporating with an inductive breakdown network. In addition, reduction of inductive source impedance in the breakdown regime is analyzed based on a series feedback mechanism and degradation of transconductance. Parameters of the modified four-port breakdown model are extracted by fitting measured four-port measurement data in the breakdown regime. Good agreement between measured and simulated 16 four-port scattering parameters (S-parameters) is achieved in the breakdown and saturation regions, validating the improved four-port breakdown model. Therefore, this model can be beneficial to RF amplifier designs in the breakdown regime with reliability considered.

Investigation of RF Avalanche Inductive Effect on Reduction of Intermodulation Distortion of MOSFETs Using Volterra Series Analysis

In this letter, the gate length dependence of X₂₁₂₁^S and X₂₁₂₁^T intercept behaviors for the MOSFETs in the breakdown region is first analyzed using the X-parameter model. Shorter gate length decreases X₂₁₂₁^S coefficients and increases X₂₁₂₁^T coefficients at high drain voltage and high input power due to the significant RF avalanche effect in the high field region. X₂₁₂₁^S and X₂₁₂₁^T coefficients are determined using artificial neural network. Good agreement between the measured and fitted output scattering wave is achieved. The presented analysis for the gate length dependence X₂₁₂₁^S and X₂₁₂₁^T can be beneficial to CMOS power amplifier designs with the scaled-down MOSFETs.

An Improved Four-Port Equivalent Circuit Model of RF MOSFETs for Breakdown Operation

In this letter, the high-frequency avalanche effects of silicon germanium (SiGe) heterojunction bipolar transistors (HBTs) on time-domain locus characteristics based on the X-parameter measurement are presented and analyzed by applying the concept of the dead space theory for the first time. The measured X-parameters of the SiGe HBTs are utilized to obtain the time-domain waveform under large-signal drive in the impact ionization region. Input locus of the SiGe HBTs operating in the avalanche regime presents negative slope when compared with the positive slope in the active region. The rotation direction of the transfer locus at breakdown is also investigated.

Gate Length Dependence of Large-Signal Output Characteristics for the MOSFETs in the Breakdown Region Using X-Parameter Model

Abstract In this paper, linearity characteristic of silicon germanium (SiGe) heterojunction bipolar transistors (HBTs) at different temperatures in the avalanche regime is investigated by the Volterra approach incorporating with a physics-based breakdown network for the first time. Third-order intermodulation distortion (IMD 3 ) decreases with increasing temperature in the impact ionization region due to lower nonlinear contributions from individual nonlinearity according to the Volterra analysis results. Calculated gain, output power, and efficiency of SiGe HBTs are in good agreement with measurement results in the avalanche region. This analysis with respect to temperature can benefit the reliability study of linearity for SiGe HBTs in the avalanche regime.