An Improved Small-Signal Model for SiGe HBT Under OFF-State, Derived From Distributed Network and Corresponding Model Parameter Extraction

IEICE Electron. Express

et al. 2018

This study presents the extraction and verification of a smallsignal model suitable for characterizing THz InP double heterojunction bipolar transistors (DHBTs). The π-type topology is adopted in the intrinsic model. Capacitances C cx and C ce are used to characterize the capacitive parasitics caused by the routing line connecting the collector terminal, base terminal and emitter terminal, respectively. The inductive parasitics introduced by the routing line are also considered. The initial values of the model parameters are extracted using a direct extraction method. The model and extraction method for the model parameters are verified by adopting an InP DHBT with 1 emitter finger and an emitter size of 0.5 µm × 5 µm. The simulation results correspond with the measured results in the frequency range from 200 MHz to 325 GHz.

Section: Introductionmentioning

confidence: 99%

Extraction and verification of the small-signal model for InP DHBTs in the 0.2–325 GHz frequency range

Zhou

Sun

IEICE Electron. Express

et al. 2018

“…The small-signal model is important since it is the basis of the large signal model, and provides important information to the foundry to guide structure optimization and process improvement. Small-signal models for InP HBT/DHBT devices have been extensively reported, [21][22][23][24][25][26][27][28][29][30][31][32][33][34][35] with verification, typically below 100 GHz, provided.…”

Section: Introductionmentioning

confidence: 99%

Machine learning technique based extremely broadband small‐signal behavioral model forInP DHBTs

Cai

Int J RF Microw Comput Aided Eng

King

et al. 2020

A novel modeling methodology for indium phosphide (InP) double heterojunction bipolar transistors (DHBTs) based on the theory of Bayesian inference, a well‐known method from the field of machine learning, is presented in this article. An extremely broadband small‐signal behavioral model, from 200 MHz to 325 GHz, is built, tested, and validated in this work, with excellent agreement obtained between the extracted model and the experimental data in the form of S‐parameters. A single finger InP DHBT device, with emitter size of 0.5 × 5 μm2 exhibiting an ft of over 550 GHz, is used in the verification example. Taking advantage of regression techniques based on machine learning concepts, the proposed black‐box behavioral model can more accurately predict the behavior of the device compared with the traditional equivalent circuit modeling method. Several sets of measured vs modeled data are shown, indicating the efficacy of the method.

“…Our previous research has discussed the distributed characteristics of SiGe HBTs under OFF-state and a simple lump voltage-controlled current source is used to mode the forward-active operation [13]. In this paper, the distributed concept is extended to the entire transistor network.…”

Section: Introductionmentioning

confidence: 99%

Distributed Small-Signal Equivalent Circuit Model and Parameter Extraction for SiGe HBT

Sun

et al. 2019

IEEE Access

Self Cite

In this paper, we present an improved high frequency small-signal distributed model for SiGe HBTs under forward-active mode based on the transmission line theory. The distributed nature of the transistor structure is taken into account in the proposed model. The single SiGe HBT is considered to be a cascade of many infinitesimal transistors, connected with the intrinsic base resistance. The closed-form solutions of admittance parameters for the distributed model are derived by solving the transmission line equation. With reasonable approximation and simplification, the model parameters are then directly extracted based on the nonlinear rational function fitting. The new improved distributed model and parameter extraction technique are validated with a 1 × 1.2 × 30 µm 2 SiGe HBT from 100 MHz to 20.89 GHz. The simulated S-parameters in the proposed transmission line model are in close agreement with the measured data, and the frequency characteristics of the transistors are well predicted.