Font Size: a A A

Research On Physical-based Large Signal Model Of Hign Power Microwave GaN HEMT

Posted on:2022-11-15Degree:MasterType:Thesis
Country:ChinaCandidate:X L YuFull Text:PDF
GTID:2518306764473804Subject:Wireless Electronics
Abstract/Summary:PDF Full Text Request
Gallium nitride(GaN)high electron mobility transistor(HEMT)is widely used in5G base station communications and radar due to their high-power density,high frequency,and high temperature resistance.In order to better guide device and circuit design,it is necessary to conduct model studies on GaN HEMTs.At present,scholars at home and abroad have carried out a lot of research on the models of GaN HEMT,mainly including empirical-based models and physical-based models.Among them,the physical-based models have the characteristics of process correlation and rapid update,and can guide the improvement of processes and devices,which have become a research hotspot.At the same time,with the further increase of the demand for high-power devices in wireless communication and radar,the research direction of the physical-based models of high-power devices has become important in the current modeling field.However,high-power devices have problems such as difficulty in testing,difficulty in describing multi-physics coupling effects,and complex physical-based models,which bring many challenges to the physical-based modeling of high-power devices.In this regard,this paper uses a physical-based model based on direct measurement to study the output characteristics of the device under high power conditions.The main research contents include:1.Research on small-signal equivalent circuit model of GaN HEMT with large gate width.Aiming at the problem that the high-power GaN devices used for this modeling cannot be directly tested on-chip,a fixture-based on-chip testing method was adopted and a fixture model was established to de-embed the measuerd data.Based on the 19-element small-signal equivalent circuit model,an accurate small-signal model is established and the model parameters are extracted.2.Research on physical-based large-signal model of GaN HEMT.First,based on the quasi-physical zone division(QPZD)model,the electron areal density(ns)model is improved,and the characterization accuracy of the saturation current(Imax)model is improved to more than 95%in the near pinch-off region.Secondly,according to the high-voltage working characteristics of high-power devices,the physical-based trap model and the self-heating model are embedded into the current model,and a physical-based current model that can accurately characterize characteristics of the high voltage is established.Then,according to the influence of the field plate on the capacitance of high-power devices,based on the surface potential theory,the analytical expression of the physical-based capacitance considering the field plates is deduced.Finally,the technical difficulties of poor convergence of the physical-based model and slow simulation speed were overcome,and a physical-based large-signal model was established.3.Validation of GaN HEMT physical-based large signal model.First,the high-power GaN HEMT physical-based large signal is verified in pulse mode and continuous mode,and the model accuracy is calculated.The results show that under the excitation of RF pulse single-tone,continuous single-tone and continuous double-tone signals,the model's characterization accuracy of the RF large-signal characteristics of the device is over 93%,92%,and 93.64%,respectively,which verifies the accuracy of the physical-based model.Then,the influence of key physical parameters in the model on the RF performance of the model is analyzed and verified,which further verifies the correctness of the physical-based large-signal model.Finally,based on the above work,a physical-based large signal model that is consistent with the physical mechanism,has high characterization accuracy and is suitable for high-power conditions is established.
Keywords/Search Tags:GaN HEMT, high power, Thesis Template, Physical-based, large signal model
PDF Full Text Request
Related items