AlGaN/GaN HEMT device small signal model extraction method

Abstract

The invention discloses a method for extracting a small signal model of an AlGaN/GaN HEMT (High Electron Mobility Transistor) device, and relates to the field of radio frequency power devices, S parameters related to different parasitic parameters are respectively measured by applying corresponding direct current bias to three electrodes of the device, and then the small signal model of the AlGaN/GaN HEMT device is extracted according to the measured S parameters on the basis of the principle of a two-port network. The method comprises the following steps of: converting a parasitic parameter matrix into a Y parameter matrix related to parasitic capacitance and a Z parameter matrix related to parasitic resistance and parasitic inductance, and then de-embedding the parasitic parameter to obtain an intrinsic parameter. The relative errors of the S parameters before and after de-embedding are solved one by one, and the errors of all frequencies are averaged, so that the parameter extraction precision is ensured while the parameter extraction speed is ensured.

Description

technical field [0001] The invention belongs to the field of radio frequency power devices, and relates to a method for extracting a small signal model of an AlGaN / GaN HEMT device, which can be used to extract a small signal equivalent circuit model of a gallium nitride high electron mobility transistor device to guide device process optimization and circuit Simulation design. Background technique [0002] Gallium Nitride High Electron Mobility Transistor (GaN HEMT) has the advantages of low intrinsic delay, good RF performance, low gate leakage, and low static power consumption, and is widely used in RF and microwave applications. And the establishment of the device equivalent circuit model has very important research significance and practical value for guiding the process steps, verifying the accuracy of the process, improving the device structure, and studying the influence of process parameters on the high-frequency performance of the device. [0003] In actual work, t...

AI Technical Summary

Problems solved by technology

[0004] At present, the mainstream small-signal parameter extraction process is mostly based on the small-signal model parameter extraction method proposed by GaAs HEMT. The parasitic parameters that have nothing to do with the bias are extracted through the cold-field S-parameters when the drain-source is short-circuited, and then the transistor operating frequency De-embed all the S-parameters and extract the intrinsic parameters, but there are other effects that cannot be ignored in GaN devices

[0005] Therefore, the accuracy of the intrinsic parameters largely depends on the accuracy of the extracted parasitic parameters, and the extraction of the parasitic parameters is related to the device process and the actual structural parameters, and there is a certain error in the structural parameters, so the empirical approximation It will lead to errors in parasitic parameters; in addition, the traditional parameter optimization method does not remove the influence of parasitic capacitance in the optimization process of parasitic inductance and resistance, and the accumulation of errors will narrow the scope of parameter optimization, so that parameter errors cannot be further reduced

Images