CN105046066A - AlGaN/GaN HETM small-signal model and parameter extraction method thereof - Google Patents

Abstract

The invention provides an AlGaN/GaN HETM small-signal model and a parameter extraction method thereof. The small-signal model comprises a parasitic part and an intrinsic part; and the intrinsic part comprises intrinsic capacitors including Cgd, Cgs and Cds, a gate-source leakage resistor Rgsf, a gate-drain leakage resistor Rgdf, a channel resistor Ri, a source-drain resistor Rds and a transconductor Gm. According to the invention, on the basis of the original AlGaN/GaN HETM small-signal model, the gate-drain leakage resistor Rgsf applicable to representing electric leakage between the gate and the drain and the gate-source leakage resistor Rgdf applicable to representing electric leakage between the gate and the source are additionally arranged; because unavoidable leakage current exists below the gate and at two sides of the gate when a device is in a normal working state, the characteristics of the device can be reflected more precisely after the two components are added, and therefore, the accuracy rate of a device model is improved.

Description

AlGaN/GaN HETM small-signal model and parameter extracting method thereof

Technical Field

The invention belongs to the technical field of integrated circuits, and particularly relates to an AlGaN/GaNHETM small-signal model and a parameter extracting method thereof.

Background

In the whole process of microwave circuit design, the device model plays a key bridge role of effectively linking the process, the device and the circuit design, so that the characteristics of the device can be accurately reflected into the circuit, the circuit simulation result is completed, the circuit characteristics are predicted, and the overall performance and the yield of the circuit are evaluated. The accurate device small signal model provides necessary data for the analysis of the corresponding large signal characteristics, and is an important means for predicting the S parameter characteristics of the device small signal; the adoption of the accurate device large-signal model simplifies the design steps of radio frequency and microwave millimeter wave power circuits to a certain extent, shortens the circuit development period and saves the cost. In addition, after the tape-out is finished, model parameters are extracted, and parameter values of important physical processes of the device can be fed back by comparing and analyzing data under different tape-out conditions, so that the process quality and the process repeatability are monitored, and the optimization direction of the device is guided. Therefore, an accurate small-signal model plays an important role in the initial process, the device and the later circuit design.

AlGaN/gan semiconductor devices (high electron mobility transistors) have a relatively short development history, and relatively few model research results, and mainly follow the relevant models of field effect transistors such as MESFETs and MOSFETs. However, the carriers in AlGaN/gan hemt are high-concentration 2DEG (two-dimensional electron gas) existing in the channel, and the working principle of AlGaN/gan hemt is very different from that of MESFET and MOSFET, and the existence of model application error is inevitable.

In recent years, some researchers have proposed several models to reflect the electrical characteristics of HEMT devices into circuits, and fig. 1 and 2 are the AlGaN/gan HEMT small-signal models commonly used at present and are shown in fig. 1 to 2. The AlGaN/gan hemt small-signal model provided in fig. 1 is composed of parasitic PAD capacitors Cpg, Cpd, and Cpgd, parasitic inductors Lg, Ld, and Ls, parasitic resistors Rg, Rd, and Rs, intrinsic capacitors Cgd, Cgs, and Cds, current input characterization units Igs, Igd, and Ids, source-drain resistors Rds, channel resistors Ri, scattering (including Rrf, Irf, and Crf) caused by frequency due to a trap effect, and heat conduction units (including thermal resistances Rth, Cth, and characterization units Ith representing input of heat conduction current); the AlGaN/gan hemt small-signal model provided in fig. 2 is composed of parasitic PAD capacitors Cpg, Cpd, and Cpgd, parasitic inductors Lg, Ld, and Ls, parasitic resistors Rg, Rd, and Rs, intrinsic capacitors Cgd, Cgs, and Cds, a source-drain resistor Rds, a channel resistor Ri, and a transconductance Gm.

Although the two models in fig. 1 and 2 can substantially reflect the characteristics of the AlGaN/gan hemt device into the circuit, the model in fig. 1 is relatively complex and time consuming; meanwhile, the reliability of the HEMT device is evaluated, the gate leakage current is an index which cannot be ignored and influences the accuracy of the model, but the two models do not well represent the leakage characteristics of the device. This puts higher demands on the modeling workers.

Disclosure of Invention

In view of the above-mentioned shortcomings of the prior art, the present invention is directed to provide an AlGaN/gan hemtm small-signal model and a parameter extracting method thereof, which take the reliability factor of the gate leakage current of the device into consideration, thereby more accurately reflecting the device characteristics. The invention can also monitor the process quality and the process repeatability under the grid, thereby guiding the direction of optimizing the device. Meanwhile, a necessary foundation is provided for the establishment of a large-signal model of the device.

In order to achieve the above objects and other related objects, the present invention provides a parameter extracting method for AlGaN/gan hettm small signal model, the parameter extracting method includes;

1) measuring S parameters of the peripheral open circuit, converting the S parameters into Y parameters, and obtaining numerical values of parasitic PAD capacitors Cpg, Cpd and Cpgd according to the Y parameters;

2) measuring S parameters of the peripheral open circuit, removing the parasitic PAD capacitors Cpg, Cpd and Cpgd obtained in the step 1), converting the S parameters of the parasitic PAD capacitors Cpg, Cpd and Cpgd to obtain Z parameters, and obtaining values of parasitic inductances Lg, Ls and Ld according to the Z parameters;

3) measuring S parameters of the peripheral open circuit, removing the parasitic PAD capacitors Cpg, Cpd and Cpgd obtained in the step 1), converting the S parameters of the parasitic PAD capacitors Cpg, Cpd and Cpgd to obtain Z parameters, and obtaining values of the parasitic resistors Rg, Rs and Rd according to the Z parameters;

4) measuring S parameters of the AlGaN/GaNHETM device, embedding parasitic PAD capacitors Cpg, Cpd and Cpgd obtained in step 1), parasitic inductors Lg, Ls and Ld obtained in step 2) and parasitic resistors Rg, Rs and Rd obtained in step 3) to obtain intrinsic S parameters of internal parameters, transforming the intrinsic S parameters to obtain Y parameters, and obtaining values of intrinsic capacitors Cgd, Cgs and Cds, channel resistors Ri, transconductance gm, transconductance delay factor tau, output admittance Gds and conductance Ggsf and Ggdf of the intrinsic parameters according to the Y parameters.

As a preferred embodiment of the parameter extracting method for the AlGaN/gan hettm small signal model of the present invention, in step 1), the S parameter of the peripheral open circuit is measured under low frequency and cut-off conditions; measuring an S parameter of the peripheral open circuit under a forward bias condition of low frequency, Vgs > Vth, and Vds ═ 0 in the step 2); measuring S parameters of the peripheral open circuit under the conditions of high frequency and cut-off in the step 3); and 4) measuring the S parameter of the AlGaN/GaNHETM device under the forward bias condition that Vgs is larger than Vpinch-off and Vds is equal to 0 in the step 4).

As a preferred embodiment of the method for referencing the AlGaN/gan hettm small signal model of the present invention, the specific method for obtaining the values of the parasitic PAD capacitances Cpg, Cpd, and Cpgd in step 1) includes:

and transforming the S parameter to obtain a Y parameter, and converting to obtain the following formula:

<math><mrow> <mfrac> <mrow> <mi>Im</mi> <mrow> <mo>(</mo> <msub> <mi>Y</mi> <mn>11</mn> </msub> <mo>)</mo> </mrow> </mrow> <mi>w</mi> </mfrac> <mo>=</mo> <msub> <mi>C</mi> <mrow> <mi>p</mi> <mi>g</mi> </mrow> </msub> <mo>+</mo> <msub> <mi>C</mi> <mrow> <mi>p</mi> <mi>g</mi> <mi>d</mi> </mrow> </msub> <mo>+</mo> <mi>W</mi> <mo>&CenterDot;</mo> <mrow> <mo>(</mo> <msub> <mi>C</mi> <mrow> <mi>g</mi> <mi>d</mi> <mi>o</mi> </mrow> </msub> <mo>+</mo> <msub> <mi>C</mi> <mrow> <mi>g</mi> <mi>s</mi> <mi>o</mi> </mrow> </msub> <mo>)</mo> </mrow> </mrow></math>

<math><mrow> <mfrac> <mrow> <mi>Im</mi> <mrow> <mo>(</mo> <msub> <mi>Y</mi> <mn>22</mn> </msub> <mo>)</mo> </mrow> </mrow> <mi>w</mi> </mfrac> <mo>=</mo> <msub> <mi>C</mi> <mrow> <mi>p</mi> <mi>d</mi> </mrow> </msub> <mo>+</mo> <msub> <mi>C</mi> <mrow> <mi>p</mi> <mi>g</mi> <mi>d</mi> </mrow> </msub> <mo>+</mo> <mi>W</mi> <mo>&CenterDot;</mo> <mrow> <mo>(</mo> <msub> <mi>C</mi> <mrow> <mi>g</mi> <mi>d</mi> <mi>o</mi> </mrow> </msub> <mo>+</mo> <msub> <mi>C</mi> <mrow> <mi>d</mi> <mi>s</mi> <mi>o</mi> </mrow> </msub> <mo>)</mo> </mrow> </mrow></math>

<math><mrow> <mo>-</mo> <mfrac> <mrow> <mi>I</mi> <mi>m</mi> <mrow> <mo>(</mo> <msub> <mi>Y</mi> <mn>12</mn> </msub> <mo>)</mo> </mrow> </mrow> <mi>w</mi> </mfrac> <mo>=</mo> <msub> <mi>C</mi> <mrow> <mi>p</mi> <mi>g</mi> <mi>d</mi> </mrow> </msub> <mo>+</mo> <mi>W</mi> <mo>&CenterDot;</mo> <msub> <mi>C</mi> <mrow> <mi>g</mi> <mi>d</mi> <mi>o</mi> </mrow> </msub> </mrow></math>

cgdo, Cgso and Cdso are scaling factors, and W is the gate width of the device; according to the linear relation of the imaginary part frequency response im (YIj)/omega of the Y parameter along with the device grid width W, numerical values of the parasitic PAD capacitors Cpg, Cpd and Cpgd can be obtained through the intercept of linear fitting.

As a preferable scheme of the parameter extracting method for the AlGaN/gan hettm small signal model of the present invention, the specific method for obtaining the values of the parasitic inductances Lg, Ls, and Ld in step 2) includes: and converting the S parameters of the capacitors Cpg, Cpd and Cpgd without the parasitic PAD to obtain Z parameters, and dividing the imaginary part of the Z parameters by omega to directly obtain the numerical values of the parasitic inductors Lg, Ls and Ld.

As a preferable scheme of the parameter extracting method for the AlGaN/gan hettm small signal model of the present invention, the specific method for obtaining the values of the parasitic resistances Rg, Rs, and Rd in step 3) includes: and converting the S parameters of the capacitors Cpg, Cpd and Cpgd without the parasitic PAD to obtain Z parameters, and directly obtaining the numerical values of the parasitic resistors Rg, Rs and Rd according to the real part of the Z parameters.

As a preferable scheme of the parameter extracting method for the AlGaN/gan hettm small signal model of the present invention, the specific method for obtaining the intrinsic parameters of intrinsic capacitances Cgd, Cgs, and Cds, channel resistance Ri, transconductance gm, transconductance delay factor τ, output admittance Gds, and conductance Ggsf, and Ggdf in step 4) includes: and transforming the intrinsic S parameters to obtain Y parameters to obtain the following formula:

$Y_{11}=\frac{1}{R_i+\frac{1}{G_{gsf}+{\mathrm{jwC}}_{gs}}}+\frac{1}{\frac{1}{G_{gdf}+{\mathrm{jwC}}_{gd}}}$

$Y_{12}=-\frac{1}{\frac{1}{G_{gdf}+{\mathrm{jwC}}_{gd}}}$

<math><mrow> <msub> <msub> <mi>Y</mi> <mn>2</mn> </msub> <mn>1</mn> </msub> <mo>=</mo> <mfrac> <mrow> <msub> <mi>g</mi> <mi>m</mi> </msub> <msup> <mi>e</mi> <mrow> <mo>-</mo> <mi>j</mi> <mi>w</mi> <mi>&tau;</mi> </mrow> </msup> </mrow> <mrow> <msub> <mi>R</mi> <mi>i</mi> </msub> <msub> <mi>G</mi> <mrow> <mi>g</mi> <mi>s</mi> <mi>f</mi> </mrow> </msub> <mo>+</mo> <msub> <mi>jwC</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> <msub> <mi>R</mi> <mi>i</mi> </msub> <mo>+</mo> <mn>1</mn> </mrow> </mfrac> <mo>-</mo> <mfrac> <mn>1</mn> <mfrac> <mn>1</mn> <mrow> <msub> <mi>G</mi> <mrow> <mi>g</mi> <mi>d</mi> <mi>f</mi> </mrow> </msub> <mo>+</mo> <msub> <mi>jwC</mi> <mrow> <mi>g</mi> <mi>d</mi> </mrow> </msub> </mrow> </mfrac> </mfrac> </mrow></math>

${Y_2}_2=G_{ds}+\frac{1}{\frac{1}{G_{gdf}+jw{\text{C}}_{gd}}}+jw{\text{C}}_{ds}$

the simplified processing is carried out on the formula to obtain Ygs, Ygd, Ygm and Yds:

$Y_{gs}=Y_{11}+Y_{12}=\frac{1}{R_i+\frac{1}{G_{gsf}+jw{\text{C}}_{gs}}}=\frac{G_{gsf}+jw{\text{C}}_{gs}}{1+R_i{G}_{gsf}+{\mathrm{jwR}}_i{C}_{gs}}$

Y_gd＝-Y₁₂＝G_gdf+jwC_gd

<math><mrow> <msub> <mi>Y</mi> <mrow> <mi>g</mi> <mi>m</mi> </mrow> </msub> <mo>=</mo> <msub> <mi>Y</mi> <mn>21</mn> </msub> <mo>-</mo> <msub> <mi>Y</mi> <mn>12</mn> </msub> <mo>=</mo> <mfrac> <mrow> <msub> <mi>g</mi> <mi>m</mi> </msub> <msup> <mi>e</mi> <mrow> <mo>-</mo> <mi>j</mi> <mi>w</mi> <mi>&tau;</mi> </mrow> </msup> </mrow> <mrow> <msub> <mi>R</mi> <mi>i</mi> </msub> <msub> <mi>G</mi> <mrow> <mi>g</mi> <mi>s</mi> <mi>f</mi> </mrow> </msub> <mo>+</mo> <msub> <mi>jwC</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> <mo>+</mo> <mn>1</mn> </mrow> </mfrac> </mrow></math>

Y_ds＝Y_i,22+Y_i,12＝G_ds+jwC_ds

defining: <math><mrow> <mi>A</mi> <mo>=</mo> <mfrac> <mrow> <mo>|</mo> <msub> <mi>Y</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> <msup> <mo>|</mo> <mn>2</mn> </msup> </mrow> <mrow> <mi>Im</mi> <mo>&lsqb;</mo> <msub> <mi>Y</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> <mo>&rsqb;</mo> </mrow> </mfrac> <mo>=</mo> <mfrac> <mrow> <msup> <msub> <mi>G</mi> <mrow> <mi>g</mi> <mi>s</mi> <mi>f</mi> </mrow> </msub> <mn>2</mn> </msup> </mrow> <mrow> <msub> <mi>wC</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> </mrow> </mfrac> <mo>+</mo> <msub> <mi>wC</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> </mrow></math>

<math><mrow> <mi>B</mi> <mo>=</mo> <mfrac> <msub> <mi>Y</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> <mrow> <mi>Im</mi> <mo>&lsqb;</mo> <msub> <mi>Y</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> <mo>&rsqb;</mo> </mrow> </mfrac> <mo>=</mo> <mfrac> <mrow> <msub> <mi>G</mi> <mrow> <mi>g</mi> <mi>s</mi> <mi>f</mi> </mrow> </msub> <mrow> <mo>(</mo> <mn>1</mn> <mo>+</mo> <msub> <mi>R</mi> <mi>i</mi> </msub> <msub> <mi>G</mi> <mrow> <mi>g</mi> <mi>s</mi> <mi>f</mi> </mrow> </msub> <mo>)</mo> </mrow> </mrow> <mrow> <msub> <mi>wC</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> </mrow> </mfrac> <mo>+</mo> <msub> <mi>wR</mi> <mi>i</mi> </msub> <msub> <mi>C</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> <mo>-</mo> <mi>j</mi> </mrow></math>

$C={\left|\frac{Y_{gs}}{Y_{gm}}\right|}^2={\left(\frac{G_{gsf}}{g_m}\right)}^2+{\left(\frac{C_{gs}}{g_m}\right)}^2{w}^2$

<math><mrow> <mi>D</mi> <mo>=</mo> <mrow> <mo>(</mo> <msub> <mi>G</mi> <mrow> <mi>g</mi> <mi>s</mi> <mi>f</mi> </mrow> </msub> <mo>+</mo> <msub> <mi>jwC</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> <mo>)</mo> </mrow> <mfrac> <msub> <mi>Y</mi> <mrow> <mi>g</mi> <mi>m</mi> </mrow> </msub> <msub> <mi>Y</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> </mfrac> <mo>=</mo> <msub> <mi>g</mi> <mi>m</mi> </msub> <msup> <mi>e</mi> <mrow> <mo>-</mo> <mi>j</mi> <mi>w</mi> <mi>&tau;</mi> </mrow> </msup> </mrow></math>

for omega²Linearly fitting the-A omega curve to obtain Cgs; combining the obtained Cgs for ω B to ω²Ri can be obtained by linear fitting of the curve; determining Ggsf according to the real part of Ygs; according to formula Y_gd＝-Y₁₂＝G_gdf+jwC_gdObtaining Cgd and Ggdf by the corresponding relation of the medium complex numbers; combining the obtained Cgs, for ω²The gm can be obtained by linear fitting of the-C curve; carrying out linear fitting on the omega-D curve to obtain tau; according to formula Y_ds＝Y_i,22+Y_i,12＝G_ds+jwC_dsGds and Cds can be obtained by the corresponding relation between the real part and the imaginary part of the medium complex number.

The invention also provides an AlGaN/GaNHETM small-signal model, which comprises a parasitic part and an intrinsic part; the intrinsic portion includes a gate-drain leakage resistance Rgdf adapted to characterize gate-to-drain leakage and a gate-source leakage resistance Rgsf adapted to characterize gate-to-source leakage.

As a preferred scheme of the AlGaN/GaNHETM small-signal model, the intrinsic part comprises intrinsic capacitances Cgd, Cgs and Cds, a gate-source leakage resistance Rgsf, a gate-drain leakage resistance Rgdf, a channel resistance Ri, a source-drain resistance Rds and a transconductance Gm; wherein,

the intrinsic capacitor Cgs is connected with the gate source leakage resistor Rgsf in parallel and then connected with the channel resistor Ri in series to form a series structure;

the intrinsic capacitor Cds is connected with the source-drain resistor Rds and the transconductance Gm in parallel and then connected with the intrinsic capacitor Cgd in series to form a series structure, and the gate-drain leakage resistor Rgdf is connected with the intrinsic capacitor Cgd in parallel;

the series structure formed by the intrinsic capacitor Cds, the source-drain resistor Rds, the transconductance Gm and the intrinsic capacitor Cgd is connected in parallel with the series structure formed by the intrinsic capacitor Cgs, the gate-source leakage resistor Rgsf and the channel resistor Ri.

As a preferred scheme of the AlGaN/gan hettm small signal model of the present invention, the parasitic portion includes parasitic PAD capacitors Cpg, Cpd, and Cpgd, parasitic inductors Lg, Ld, and Ls, and parasitic resistors Rg, Rd, and Rs; wherein,

the parasitic inductance Lg and the parasitic resistance Rg are connected in series to form a series structure, one end of the series structure formed by the parasitic inductance Lg and the parasitic resistance Rg is connected to one end of the parasitic PAD capacitance Cpg, and the other end of the series structure is connected between the intrinsic capacitances Cgd and Cgs; the other end of the parasitic PAD capacitance Cpg is grounded;

the parasitic inductor Ld and the parasitic resistor Rd are connected in series to form a series structure, one end of the series structure formed by the parasitic inductor Ld and the parasitic resistor Rd is connected to one end of the parasitic PAD capacitor Cpd, and the other end of the series structure is connected between the intrinsic capacitors Cgd and Cds; the other end of the parasitic PAD capacitor Cpd is grounded;

one end of the parasitic PAD capacitor Cpgd is connected to the connection end of the parasitic PAD capacitor Cpg and the series structure formed by the parasitic inductor Lg and the parasitic resistor Rg, and the other end of the parasitic PAD capacitor Cpgd is connected to the connection end of the parasitic PAD capacitor Cpd and the series structure formed by the parasitic inductor Ld and the parasitic resistor Rd;

the parasitic inductor Ls and the parasitic resistor Rs are connected in series to form a series structure, one end of the series structure formed by the parasitic inductor Ls and the parasitic resistor Rs is connected to one end, far away from the intrinsic capacitor Cgs, of the channel resistor Ri, and the other end of the series structure is grounded.

As described above, the present invention provides an AlGaN/GaNHETM small signal model and a parameter extracting method thereof, which have the following beneficial effects: the invention adds a grid leakage resistance Rgsf suitable for representing the leakage between grid and drain and a grid source leakage resistance Rgdf suitable for representing the leakage between grid and source on the basis of the existing AlGaN/GaNHEMT small-signal model, and because the device has unavoidable leakage currents under the grid and at two sides of the grid under the normal working state, the device can reflect the characteristics of the device more accurately and improve the model accuracy of the device.

Drawings

Fig. 1 to 2 are circuit diagrams illustrating a AlGaN/gan hemt small-signal model in the related art.

Fig. 3 is a schematic structural diagram of a device corresponding to the AlGaN/gan hettm small-signal model of the present invention.

FIG. 4 is an equivalent circuit diagram of the AlGaN/GaNHETM small-signal model of the present invention.

FIG. 5 is a flowchart of a parameter extracting method of the AlGaN/GaNHETM small-signal model of the present invention.

Fig. 6 is a diagram showing a low-frequency equivalent circuit of a small signal under a cut-off condition in the parameter extracting method of the AlGaN/gan hettm small signal model of the present invention.

FIG. 7 is a circuit diagram showing an equivalent circuit under forward bias conditions in the proposed method of the AlGaN/GaNHETM small-signal model of the present invention.

FIG. 8 is a diagram showing a small-signal high-frequency equivalent circuit under a cut-off condition in the proposed method of the AlGaN/GaNHETM small-signal model of the present invention.

Detailed Description

The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention.

Please refer to fig. 3 to 8. It should be noted that the drawings provided in the present embodiment are only for illustrating the basic idea of the present invention, and although the drawings only show the components related to the present invention and are not drawn according to the number, shape and size of the components in actual implementation, the type, quantity and proportion of the components in actual implementation may be changed arbitrarily, and the layout of the components may be more complicated.

Referring to fig. 3 to 4, the present embodiment provides an AlGaN/gan hemt small-signal model, a device structure corresponding to the AlGaN/gan hemt small-signal model is shown in fig. 3, and the following embodiment describes an equivalent circuit and a reference method using the device structure. The extraction of small signal parameters is an essential link in the modeling of the FET semiconductor device, the accurate extraction of peripheral parasitic parameters can influence the accuracy of internal nonlinear parameters, the extraction of small signal models is the basis for the establishment of large signal models, and the large signal models can be established more accurately only by ensuring that all small signal parameter values are accurate and reasonable.

FIG. 4 is an equivalent circuit diagram of a AlGaN/GaNHETM small-signal model, which can be seen from FIG. 4 and includes a parasitic portion and an intrinsic portion; the intrinsic portion includes a gate-drain leakage resistance Rgdf adapted to characterize gate-to-drain leakage and a gate-source leakage resistance Rgs adapted to characterize gate-to-source leakage.

Specifically, the intrinsic part comprises intrinsic capacitances Cgd, Cgs and Cds, a gate-source leakage resistor Rgsf, a gate-drain leakage resistor Rgdf, a channel resistor Ri, a source-drain resistor Rds and a transconductance Gm; the intrinsic capacitor Cgs is connected with the gate-source leakage resistor Rgsf in parallel and then connected with the channel resistor Ri in series to form a series structure; the intrinsic capacitor Cds is connected with the source-drain resistor Rds and the transconductance Gm in parallel and then connected with the intrinsic capacitor Cgd in series to form a series structure, and the gate-drain leakage resistor Rgdf is connected with the intrinsic capacitor Cgd in parallel; the series structure formed by the intrinsic capacitor Cds, the source-drain resistor Rds, the transconductance Gm and the intrinsic capacitor Cgd is connected in parallel with the series structure formed by the intrinsic capacitor Cgs, the gate-source leakage resistor Rgsf and the channel resistor Ri.

Specifically, the parasitic part comprises parasitic PAD capacitors Cpg, Cpd and Cpgd, parasitic inductors Lg, Ld and Ls, and parasitic resistors Rg, Rd and Rs; the parasitic inductor Lg and the parasitic resistor Rg are connected in series to form a series structure, one end of the series structure formed by the parasitic inductor Lg and the parasitic resistor Rg is connected to one end of the parasitic PAD capacitor Cpg, and the other end of the series structure is connected between the intrinsic capacitors Cgd and Cgs; the other end of the parasitic PAD capacitance Cpg is grounded; the parasitic inductor Ld and the parasitic resistor Rd are connected in series to form a series structure, one end of the series structure formed by the parasitic inductor Ld and the parasitic resistor Rd is connected to one end of the parasitic PAD capacitor Cpd, and the other end of the series structure is connected between the intrinsic capacitors Cgd and Cds; the other end of the parasitic PAD capacitor Cpd is grounded; one end of the parasitic PAD capacitor Cpgd is connected to the connection end of the parasitic PAD capacitor Cpg and the series structure formed by the parasitic inductor Lg and the parasitic resistor Rg, and the other end of the parasitic PAD capacitor Cpgd is connected to the connection end of the parasitic PAD capacitor Cpd and the series structure formed by the parasitic inductor Ld and the parasitic resistor Rd; the parasitic inductor Ls and the parasitic resistor Rs are connected in series to form a series structure, one end of the series structure formed by the parasitic inductor Ls and the parasitic resistor Rs is connected to one end, far away from the intrinsic capacitor Cgs, of the channel resistor Ri, and the other end of the series structure is grounded.

Because the device is under the normal working condition, the gate and both sides of the gate have unavoidable leakage current, therefore, the invention adds the conductance Ggsf suitable for representing the leakage between the gate and the drain and the conductance Ggdf suitable for representing the leakage between the gate and the source on the basis of the existing AlGaN/GaNHEMT small signal model, in fig. 4, the conductance Ggsf and Ggdf representing the gate leakage are represented by the gate-source leakage resistance Rgsf and the gate-drain leakage resistance Rgdf respectively, Ggsf is 1/Rgsf, and Ggdf is 1/Rgdf; after two elements of a gate source leakage resistor Rgsf and a gate drain leakage resistor Rgdf are added into the AlGaN/GaNHEMT small-signal model, the characteristics of the device can be reflected more accurately, and the model accuracy of the device is improved.

Referring to fig. 5 to 8, the present invention further provides a referencing method of an AlGaN/gan hemt small signal model, the referencing method at least includes the following steps:

1) measuring S parameters of the peripheral open circuit under the conditions of low frequency and cut-off, converting the S parameters into Y parameters, and obtaining values of the parasitic PAD capacitors Cpg, Cpd and Cpgd according to the Y parameters; the parasitic PAD capacitors Cpg, Cpd and Cpgd mainly have parasitic effects among the metal of the grid end, the source end and the drain end and the substrate and the total effect of coupling capacitors among three electrodes;

2) measuring S parameters of a peripheral open circuit under the forward bias condition of low frequency, gs > Vth (threshold voltage) and Vds ═ 0, removing parasitic PAD capacitors Cpg, Cpd and Cpgd obtained in the step 1), converting the S parameters without the parasitic PAD capacitors Cpg, Cpd and Cpgd to obtain Z parameters, and obtaining values of parasitic inductances Lg, Ls and Ld according to the Z parameters;

3) measuring S parameters of the peripheral open circuit under the conditions of high frequency and cut-off, removing the parasitic PAD capacitors Cpg, Cpd and Cpgd obtained in the step 1), converting the S parameters of the parasitic PAD capacitors Cpg, Cpd and Cpgd to obtain Z parameters, and obtaining values of the parasitic resistors Rg, Rs and Rd according to the Z parameters;

4) under the forward bias condition that Vgs is larger than Vpinch-off (turn-off voltage) and Vds is equal to 0, measuring S parameters of the AlGaN/GaNHETM device, de-embedding parasitic PAD capacitors Cpg, Cpd and Cpgd obtained in step 1), parasitic inductors Lg, Ls and Ld obtained in step 2) and parasitic resistors Rg, Rs and Rd obtained in step 3) to obtain intrinsic S parameters of internal parameters, converting the intrinsic S parameters to obtain Y parameters, and obtaining values of the intrinsic capacitors Cgd, Cgs and Cds, channel resistance Ri, transconductance gm, transconductance delay factor tau, output admittance Gds and conductance Ggsf and Ggdf according to the Y parameters.

Specifically, the physical meanings of the extracted parameters are respectively as follows:

the parasitic PAD capacitors Cpg, Cpd and Cpgd mainly have parasitic effects among the metal of the grid end, the source end and the drain end and the substrate and the total effect of coupling capacitors among three electrodes;

the parasitic inductances Lg, Ld and Ls are mainly parasitic effects formed by metal on the surface of the device at a gate end, a drain end and a source end, and have larger influence on the performance of the device, especially under a high-frequency condition;

the parasitic resistors Rd and Rs represent ohmic contact metal resistors of a drain end and a source end respectively, and also comprise body resistors diffused and injected into an active region, and the parasitic resistors Rg are mainly brought by Schottky gate metal at a gate end; the parasitic resistances Rg, Rd and Rs sometimes change with the bias voltage, but the resistance values are generally considered to be constant in a small signal model;

the intrinsic capacitance Cgs can be regarded as the sum of capacitances formed between the gate and the source and between the gate and the channel, with the space charge region as the medium; similarly, the intrinsic capacitance Cgd is the sum of capacitances formed between the gate and the drain and between the gate and the channel; the intrinsic capacitance Cds is used for representing the coupling capacitance between the source electrode and the drain electrode;

the transconductance Gm is used for measuring the variable of the change of the input grid-source voltage Vgs on the output drain-source current Ids, the physical parameter gives the internal gain of the device, and the transconductance Gm is an important device index for measuring the application of microwaves and millimeter waves;

the transconductance delay factor tau characterizes the time required for the charge of a space point under the grid to be redistributed from one stable state to another stable state when Vgs changes; the channel resistance Ri is the resistance between the channel and the source;

the output admittance Gds, which is used to measure the variation of the output drain-source voltage Vds over the output drain-source current Ids, characterizes the maximum voltage gain obtained from the device and is important to determine the optimal output impedance match of the device;

in the AlGaN/gan hemt device, the condition of conduction current between the gate source and the gate drain can be equivalent to that a schottky diode exists between the gate source and the gate drain, and the resistance of gate current conduction in the schottky diode is characterized by Ggsf and Ggdf (or Rgsf and Rgdf), wherein Ggsf is 1/Rgsf and Ggdf is 1/Rgdf; it is clear that when the applied gate voltage is greater than the turn-on voltage of the diode, the schottky diode is turned on, the values of Rgsf and Rgdf are small, and the values of Ggsf and Ggdf are large.

Specifically, the specific method for obtaining the values of the parasitic PAD capacitances Cpg, Cpd and Cpgd obtained in the step 1) includes:

ids and gm are both approximate to 0 under the cut-off condition, the influence of parasitic inductance and parasitic resistance can be ignored under the low frequency, an equivalent circuit of the device structure is simplified, the simplified small-signal low-frequency equivalent circuit under the cut-off condition is shown in figure 6, S parameters of a peripheral open circuit under the cut-off condition are measured, the S parameters are converted to obtain Y parameters, and the imaginary part of the Y parameters can be written as:

$\frac{\mathrm{Im}\left(Y_{11}\right)}{w}=C_{pg}+C_{gs}+C_{pgd}+C_{gd}$

$\frac{\mathrm{Im}\left(Y_{22}\right)}{w}=C_{pd}+C_{ds}+C_{pgd}+C_{gd}$

$-\frac{\mathrm{Im}\left(Y_{12}\right)}{w}=C_{pgd}+C_{gd}$

the intrinsic capacitance scaling equation can be written as:

C_gd(W)＝C_gdo·W

C_gs(W)＝C_gso·W

C_ds(W)＝C_dso·W

here, Cgd, Cgs, and Cds are intrinsic capacitances in the off state of the device, Cgdo, Cgso, and Cdso are scaling factors, and W is the gate width of the device.

Substituting the intrinsic capacitance scaling formula into the Y parameter imaginary part expression can obtain the following formula:

<math><mrow> <mfrac> <mrow> <mi>Im</mi> <mrow> <mo>(</mo> <msub> <mi>Y</mi> <mn>11</mn> </msub> <mo>)</mo> </mrow> </mrow> <mi>w</mi> </mfrac> <mo>=</mo> <msub> <mi>C</mi> <mrow> <mi>p</mi> <mi>g</mi> </mrow> </msub> <mo>+</mo> <msub> <mi>C</mi> <mrow> <mi>p</mi> <mi>g</mi> <mi>d</mi> </mrow> </msub> <mo>+</mo> <mi>W</mi> <mo>&CenterDot;</mo> <mrow> <mo>(</mo> <msub> <mi>C</mi> <mrow> <mi>g</mi> <mi>d</mi> <mi>o</mi> </mrow> </msub> <mo>+</mo> <msub> <mi>C</mi> <mrow> <mi>g</mi> <mi>s</mi> <mi>o</mi> </mrow> </msub> <mo>)</mo> </mrow> </mrow></math>

<math><mrow> <mfrac> <mrow> <mi>Im</mi> <mrow> <mo>(</mo> <msub> <mi>Y</mi> <mn>22</mn> </msub> <mo>)</mo> </mrow> </mrow> <mi>w</mi> </mfrac> <mo>=</mo> <msub> <mi>C</mi> <mrow> <mi>p</mi> <mi>d</mi> </mrow> </msub> <mo>+</mo> <msub> <mi>C</mi> <mrow> <mi>p</mi> <mi>g</mi> <mi>d</mi> </mrow> </msub> <mo>+</mo> <mi>W</mi> <mo>&CenterDot;</mo> <mrow> <mo>(</mo> <msub> <mi>C</mi> <mrow> <mi>g</mi> <mi>d</mi> <mi>o</mi> </mrow> </msub> <mo>+</mo> <msub> <mi>C</mi> <mrow> <mi>d</mi> <mi>s</mi> <mi>o</mi> </mrow> </msub> <mo>)</mo> </mrow> </mrow></math>

<math><mrow> <mo>-</mo> <mfrac> <mrow> <mi>Im</mi> <mrow> <mo>(</mo> <msub> <mi>Y</mi> <mn>12</mn> </msub> <mo>)</mo> </mrow> </mrow> <mi>w</mi> </mfrac> <mo>=</mo> <msub> <mi>C</mi> <mrow> <mi>p</mi> <mi>g</mi> <mi>d</mi> </mrow> </msub> <mo>+</mo> <mi>W</mi> <mo>&CenterDot;</mo> <msub> <mi>C</mi> <mrow> <mi>g</mi> <mi>d</mi> <mi>o</mi> </mrow> </msub> </mrow></math>

cgdo, Cgso and Cdso are scaling factors, and W is the gate width of the device; according to the linear relation of the imaginary part frequency response im (YIj)/omega of the Y parameter along with the device grid width W, numerical values of the parasitic PAD capacitors Cpg, Cpd and Cpgd can be obtained through the intercept of linear fitting.

Specifically, the specific method for obtaining the values of the parasitic inductances Lg, Ls and Ld in the step 2) comprises the following steps:

under the forward bias condition that Vgs is greater than Vth and Vds is equal to 0, gm is equal to 0, and Cgs is equal to Cgd is equal to 0, the equivalent circuit of the device structure is simplified, as shown in fig. 7, the equivalent circuit under the simplified forward bias condition measures the S parameter of the open circuit at the periphery, removes the parasitic PAD capacitors Cpg, Cpd, and Cpgd obtained in step 1), transforms the S parameter from which the parasitic PAD capacitors Cpg, Cpd, and Cpgd are removed to obtain the Z parameter, and the values of the parasitic inductors Lg, Ls, and Ld can be directly obtained from the imaginary part and ω of the Z parameter:

$L_g=\frac{\mathrm{Im}(Z_{11}-Z_{12})}{w}$

$L_d=\frac{\mathrm{Im}(Z_{22}-Z_{12})}{w}$

$L_s=\frac{\mathrm{Im}\left(Z_{12}\right)}{w}$

specifically, the specific method for obtaining the values of the parasitic resistances Rg, Rs, and Rd in the step 3) is as follows:

under the cut-off condition, simplifying an equivalent circuit of the device structure, measuring an S parameter of a peripheral open circuit by the simplified small-signal high-frequency equivalent circuit under the cut-off condition as shown in FIG. 8, removing parasitic PAD capacitors Cpg, Cpd and Cpgd obtained in the step 1), converting the S parameter without the parasitic PAD capacitors Cpg, Cpd and Cpgd to obtain a Z parameter, and directly obtaining values of parasitic resistors Rg, Rs and Rd according to a real part of the Z parameter:

R_g＝Re(Z₁₁-Z₁₂)

R_d＝Re(Z₂₂-Z₁₂)

R_s＝Re(Z₁₂)＝Re(Z₂₁)

and extracting the parasitic resistance by using a method of extracting the parasitic resistance by using the S parameter under the cut-off condition. The method does not depend on a forward bias grid Schottky junction, eliminates grid degradation caused by large grid current, all parasitic resistances can be directly obtained, and eliminates uncertainty in a Cold-FET method. The Cold-FET method is a method for determining parasitic resistance and inductance by using a radio frequency S parameter under a forward bias condition of a gate Schottky diode.

In this embodiment, the low frequency is less than 5GHz, and the high frequency is 30GHz to 40 GHz.

Specifically, in step 4), the specific method for obtaining the intrinsic parameters of intrinsic capacitances Cgd, Cgs, and Cds, channel resistance Ri, transconductance gm, transconductance delay factor τ, output admittance Gds, and the values of conductances Ggsf and Ggdf in step 4) is as follows:

under the forward bias condition that Vgs is larger than Vpinch-off (turn-off voltage) and Vds is equal to 0, measuring S parameters of the AlGaN/GaNHETM device, de-embedding parasitic PAD capacitors Cpg, Cpd and Cpgd obtained in the step 1), parasitic inductors Lg, Ls and Ld obtained in the step 2) and parasitic resistors Rg, Rs and Rd obtained in the step 3) to obtain intrinsic S parameters of internal parameters, and converting the intrinsic S parameters to obtain Y parameters:

$Y_{11}=\frac{1}{R_i+\frac{1}{G_{gsf}+{\mathrm{jwC}}_{gs}}}+\frac{1}{\frac{1}{G_{gdf}+{\mathrm{jwC}}_{gd}}}$

$Y_{12}=-\frac{1}{\frac{1}{G_{gdf}+{\mathrm{jwC}}_{gd}}}$

<math><mrow> <msub> <mi>Y</mi> <mn>21</mn> </msub> <mo>=</mo> <mfrac> <mrow> <msub> <mi>g</mi> <mi>m</mi> </msub> <msup> <mi>e</mi> <mrow> <mo>-</mo> <mi>j</mi> <mi>w</mi> <mi>&tau;</mi> </mrow> </msup> </mrow> <mrow> <msub> <mi>R</mi> <mi>i</mi> </msub> <msub> <mi>G</mi> <mrow> <mi>g</mi> <mi>s</mi> <mi>f</mi> </mrow> </msub> <mo>+</mo> <msub> <mi>jwC</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> <msub> <mi>R</mi> <mi>i</mi> </msub> <mo>+</mo> <mn>1</mn> </mrow> </mfrac> <mo>-</mo> <mfrac> <mn>1</mn> <mfrac> <mn>1</mn> <mrow> <msub> <mi>G</mi> <mrow> <mi>g</mi> <mi>d</mi> <mi>f</mi> </mrow> </msub> <mo>+</mo> <msub> <mi>jwC</mi> <mrow> <mi>g</mi> <mi>d</mi> </mrow> </msub> </mrow> </mfrac> </mfrac> </mrow></math>

$Y_{22}=G_{ds}+\frac{1}{\frac{1}{G_{gdf}+{\mathrm{jwC}}_{gd}}}+{\mathrm{jwC}}_{ds}$

the simplified processing is carried out on the formula to obtain Ygs, Ygd, Ygm and Yds:

$Y_{gs}=Y_{11}+Y_{12}=\frac{1}{R_i+\frac{1}{G_{gsf}+{\mathrm{jwC}}_{gs}}}=\frac{G_{gsf}+{\mathrm{jwC}}_{gs}}{1+R_i{G}_{gsf}+{\mathrm{jwR}}_i{C}_{gs}}$

Y_gd＝-Y₁₂＝G_gdf+jwC_gd

<math><mrow> <msub> <mi>Y</mi> <mrow> <mi>g</mi> <mi>m</mi> </mrow> </msub> <mo>=</mo> <msub> <mi>Y</mi> <mn>21</mn> </msub> <mo>-</mo> <msub> <mi>Y</mi> <mn>12</mn> </msub> <mo>=</mo> <mfrac> <mrow> <msub> <mi>g</mi> <mi>m</mi> </msub> <msup> <mi>e</mi> <mrow> <mo>-</mo> <mi>j</mi> <mi>w</mi> <mi>&tau;</mi> </mrow> </msup> </mrow> <mrow> <msub> <mi>R</mi> <mi>i</mi> </msub> <msub> <mi>G</mi> <mrow> <mi>g</mi> <mi>s</mi> <mi>f</mi> </mrow> </msub> <mo>+</mo> <msub> <mi>jwC</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> <mo>+</mo> <mn>1</mn> </mrow> </mfrac> </mrow></math>

Y_ds＝Y_i,22+Y_i,12＝G_ds+jwC_ds

defining: <math><mrow> <mi>A</mi> <mo>=</mo> <mfrac> <msup> <mrow> <mo>|</mo> <msub> <mi>Y</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> <mo>|</mo> </mrow> <mn>2</mn> </msup> <mrow> <mi>Im</mi> <mo>&lsqb;</mo> <msub> <mi>Y</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> <mo>&rsqb;</mo> </mrow> </mfrac> <mo>=</mo> <mfrac> <mrow> <msup> <msub> <mi>G</mi> <mrow> <mi>g</mi> <mi>s</mi> <mi>f</mi> </mrow> </msub> <mn>2</mn> </msup> </mrow> <mrow> <msub> <mi>wC</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> </mrow> </mfrac> <mo>+</mo> <msub> <mi>wC</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> </mrow></math>

<math><mrow> <mi>B</mi> <mo>=</mo> <mfrac> <msub> <mi>Y</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> <mrow> <mi>Im</mi> <mo>&lsqb;</mo> <msub> <mi>Y</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> <mo>&rsqb;</mo> </mrow> </mfrac> <mo>=</mo> <mfrac> <mrow> <msub> <mi>G</mi> <mrow> <mi>g</mi> <mi>s</mi> <mi>f</mi> </mrow> </msub> <mrow> <mo>(</mo> <mn>1</mn> <mo>+</mo> <msub> <mi>R</mi> <mi>i</mi> </msub> <msub> <mi>G</mi> <mrow> <mi>g</mi> <mi>s</mi> <mi>f</mi> </mrow> </msub> <mo>)</mo> </mrow> </mrow> <mrow> <msub> <mi>wC</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> </mrow> </mfrac> <mo>+</mo> <msub> <mi>wR</mi> <mi>i</mi> </msub> <msub> <mi>C</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> <mo>-</mo> <mi>j</mi> </mrow></math>

$C={\left|\frac{Y_{gs}}{Y_{gm}}\right|}^2={\left(\frac{G_{gsf}}{g_m}\right)}^2+{\left(\frac{C_{gs}}{g_m}\right)}^2{w}^2$

<math><mrow> <mi>D</mi> <mo>=</mo> <mrow> <mo>(</mo> <msub> <mi>G</mi> <mrow> <mi>g</mi> <mi>s</mi> <mi>f</mi> </mrow> </msub> <mo>+</mo> <msub> <mi>jwC</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> <mo>)</mo> </mrow> <mfrac> <msub> <mi>Y</mi> <mrow> <mi>g</mi> <mi>m</mi> </mrow> </msub> <msub> <mi>Y</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> </mfrac> <mo>=</mo> <msub> <mi>g</mi> <mi>m</mi> </msub> <msup> <mi>e</mi> <mrow> <mo>-</mo> <mi>j</mi> <mi>w</mi> <mi>&tau;</mi> </mrow> </msup> </mrow></math>

for omega²Linearly fitting the-A omega curve to obtain Cgs; combining the obtained Cgs for ω B to ω²Ri can be obtained by linear fitting of the curve; determining Ggsf according to the real part of Ygs; according to formula Y_gd＝-Y₁₂＝G_gdf+jwC_gdThe Cgd and the Ggdf can be obtained by the corresponding relation of the real part and the imaginary part of the medium complex number; combining the obtained Cgs, for ω²The gm can be obtained by linear fitting of the-C curve; carrying out linear fitting on the omega-D curve to obtain tau; according to formula Y_ds＝Y_i,22+Y_i,12＝G_ds+jwC_dsGds and Cds can be obtained by the corresponding relation between the real part and the imaginary part of the medium complex number.

In summary, the present invention provides an AlGaN/gan hemt small signal model and a parameter extraction method thereof, wherein a gate-drain leakage resistance Rgsf suitable for representing gate-drain leakage and a gate-source leakage resistance Rgdf suitable for representing gate-source leakage are added on the basis of the existing AlGaN/gan hemt small signal model, and unavoidable leakage currents exist under a gate and at two sides of the gate in a normal working state of a device, so that the characteristics of the device can be more accurately reflected and the accuracy of the device model can be improved after the two elements are added.

The above embodiments are merely illustrative of the principles and effects of the present invention, and are not intended to limit the invention, for example, the invention may also be used with tri-or multi-epitaxial layers. Any person skilled in the art can modify or change the above-mentioned embodiments without departing from the spirit and scope of the present invention. Accordingly, it is intended that all equivalent modifications or changes which can be made by those skilled in the art without departing from the spirit and technical spirit of the present invention be covered by the claims of the present invention.

Claims (9)

1. A parameter extraction method for an AlGaN/GaNHETM small signal model is characterized by comprising the following steps of;

1) measuring S parameters of the peripheral open circuit, converting the S parameters into Y parameters, and obtaining numerical values of parasitic PAD capacitors Cpg, Cpd and Cpgd according to the Y parameters;

2) measuring S parameters of the peripheral open circuit, removing the parasitic PAD capacitors Cpg, Cpd and Cpgd obtained in the step 1), converting the S parameters of the parasitic PAD capacitors Cpg, Cpd and Cpgd to obtain Z parameters, and obtaining values of parasitic inductances Lg, Ls and Ld according to the Z parameters;

3) measuring S parameters of the peripheral open circuit, removing the parasitic PAD capacitors Cpg, Cpd and Cpgd obtained in the step 1), converting the S parameters of the parasitic PAD capacitors Cpg, Cpd and Cpgd to obtain Z parameters, and obtaining values of the parasitic resistors Rg, Rs and Rd according to the Z parameters;

4) measuring S parameters of the AlGaN/GaNHETM device, embedding parasitic PAD capacitors Cpg, Cpd and Cpgd obtained in step 1), parasitic inductors Lg, Ls and Ld obtained in step 2) and parasitic resistors Rg, Rs and Rd obtained in step 3) to obtain intrinsic S parameters of internal parameters, transforming the intrinsic S parameters to obtain Y parameters, and obtaining values of intrinsic capacitors Cgd, Cgs and Cds, channel resistors Ri, transconductance gm, transconductance delay factor tau, output admittance Gds and conductance Ggsf and Ggdf of the intrinsic parameters according to the Y parameters.

2. The method of claim 1, wherein the method comprises: measuring S parameters of the peripheral open circuit under the conditions of low frequency and cut-off in the step 1); measuring an S parameter of the peripheral open circuit under a forward bias condition of low frequency, Vgs > Vth, and Vds ═ 0 in the step 2); measuring S parameters of the peripheral open circuit under the conditions of high frequency and cut-off in the step 3); and 4) measuring the S parameter of the AlGaN/GaNHETM device under the forward bias condition that Vgs is larger than Vpinch-off and Vds is equal to 0 in the step 4).

3. The method of claim 1, wherein the method comprises: the specific method for obtaining the values of the parasitic PAD capacitances Cpg, Cpd and Cpgd in the step 1) is as follows:

and transforming the S parameter to obtain a Y parameter, and converting to obtain the following formula:

<math> <mrow> <mfrac> <mrow> <mi>Im</mi> <mrow> <mo>(</mo> <msub> <mi>Y</mi> <mn>11</mn> </msub> <mo>)</mo> </mrow> </mrow> <mi>w</mi> </mfrac> <mo>=</mo> <msub> <mi>C</mi> <mrow> <mi>p</mi> <mi>g</mi> </mrow> </msub> <mo>+</mo> <msub> <mi>C</mi> <mrow> <mi>p</mi> <mi>g</mi> <mi>d</mi> </mrow> </msub> <mo>+</mo> <mi>W</mi> <mo>&CenterDot;</mo> <mrow> <mo>(</mo> <msub> <mi>C</mi> <mrow> <mi>g</mi> <mi>d</mi> <mi>o</mi> </mrow> </msub> <mo>+</mo> <msub> <mi>C</mi> <mrow> <mi>g</mi> <mi>s</mi> <mi>o</mi> </mrow> </msub> <mo>)</mo> </mrow> </mrow> </math>

<math> <mrow> <mfrac> <mrow> <mi>Im</mi> <mrow> <mo>(</mo> <msub> <mi>Y</mi> <mn>22</mn> </msub> <mo>)</mo> </mrow> </mrow> <mi>w</mi> </mfrac> <mo>=</mo> <msub> <mi>C</mi> <mrow> <mi>p</mi> <mi>d</mi> </mrow> </msub> <mo>+</mo> <msub> <mi>C</mi> <mrow> <mi>p</mi> <mi>g</mi> <mi>d</mi> </mrow> </msub> <mo>+</mo> <mi>W</mi> <mo>&CenterDot;</mo> <mrow> <mo>(</mo> <msub> <mi>C</mi> <mrow> <mi>g</mi> <mi>d</mi> <mi>o</mi> </mrow> </msub> <mo>+</mo> <msub> <mi>C</mi> <mrow> <mi>d</mi> <mi>s</mi> <mi>o</mi> </mrow> </msub> <mo>)</mo> </mrow> </mrow> </math>

<math> <mrow> <mo>-</mo> <mfrac> <mrow> <mi>Im</mi> <mrow> <mo>(</mo> <msub> <mi>Y</mi> <mn>12</mn> </msub> <mo>)</mo> </mrow> </mrow> <mi>w</mi> </mfrac> <mo>=</mo> <msub> <mi>C</mi> <mrow> <mi>p</mi> <mi>g</mi> <mi>d</mi> </mrow> </msub> <mo>+</mo> <mi>W</mi> <mo>&CenterDot;</mo> <msub> <mi>C</mi> <mrow> <mi>g</mi> <mi>d</mi> <mi>o</mi> </mrow> </msub> </mrow> </math>

cgdo, Cgso and Cdso are scaling factors, and W is the gate width of the device; according to the linear relation of the imaginary part frequency response im (YIj)/omega of the Y parameter along with the device grid width W, numerical values of the parasitic PAD capacitors Cpg, Cpd and Cpgd can be obtained through the intercept of linear fitting.

4. The method of claim 1, wherein the method comprises: the specific method for obtaining the values of the parasitic inductances Lg, Ls and Ld in the step 2) comprises the following steps: and converting the S parameters of the capacitors Cpg, Cpd and Cpgd without the parasitic PAD to obtain Z parameters, and dividing the imaginary part of the Z parameters by omega to directly obtain the numerical values of the parasitic inductors Lg, Ls and Ld.

5. The method of claim 1, wherein the method comprises: the specific method for obtaining the values of the parasitic resistances Rg, Rs and Rd in the step 3) is as follows: and converting the S parameters of the capacitors Cpg, Cpd and Cpgd without the parasitic PAD to obtain Z parameters, and directly obtaining the numerical values of the parasitic resistors Rg, Rs and Rd according to the real part of the Z parameters.

6. The method of claim 1, wherein the method comprises: the specific method for obtaining the intrinsic parameters, i.e., the values of the intrinsic capacitances Cgd, Cgs and Cds, the channel resistance Ri, the transconductance gm, the transconductance delay factor τ, the output admittance Gds and the conductances Ggsf and Ggdf in the step 4) is as follows: and transforming the intrinsic S parameters to obtain Y parameters to obtain the following formula:

$Y_{11}=\frac{1}{R_i+\frac{1}{G_{gsf}+{\mathrm{jwC}}_{gs}}}+\frac{1}{\frac{1}{G_{gdf}+{\mathrm{jwC}}_{gd}}}$

$Y_{12}=-\frac{1}{\frac{1}{G_{gdf}+{\mathrm{jwC}}_{gd}}}$

<math> <mrow> <msub> <mi>Y</mi> <mn>21</mn> </msub> <mo>=</mo> <mfrac> <mrow> <msub> <mi>g</mi> <mi>m</mi> </msub> <msup> <mi>e</mi> <mrow> <mo>-</mo> <mi>j</mi> <mi>w</mi> <mi>&tau;</mi> </mrow> </msup> </mrow> <mrow> <msub> <mi>R</mi> <mi>i</mi> </msub> <msub> <mi>G</mi> <mrow> <mi>g</mi> <mi>s</mi> <mi>f</mi> </mrow> </msub> <mo>+</mo> <msub> <mi>jwC</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> <msub> <mi>R</mi> <mi>i</mi> </msub> <mo>+</mo> <mn>1</mn> </mrow> </mfrac> <mo>-</mo> <mfrac> <mn>1</mn> <mfrac> <mn>1</mn> <mrow> <msub> <mi>G</mi> <mrow> <mi>g</mi> <mi>d</mi> <mi>f</mi> </mrow> </msub> <mo>+</mo> <msub> <mi>jwC</mi> <mrow> <mi>g</mi> <mi>d</mi> </mrow> </msub> </mrow> </mfrac> </mfrac> </mrow> </math>

$Y_{22}=G_{ds}+\frac{1}{\frac{1}{G_{gdf}+{\mathrm{jwC}}_{gd}}}+{\mathrm{jwC}}_{ds}$

the simplified processing is carried out on the formula to obtain Ygs, Ygd, Ygm and Yds:

$Y_{gs}=Y_{11}+Y_{12}=\frac{1}{R_i+\frac{1}{G_{gsf}+{\mathrm{jwC}}_{gs}}}=\frac{G_{gsf}+{\mathrm{jwC}}_{gs}}{1+R_i{G}_{gsf}+{\mathrm{jwR}}_i{C}_{gs}}$

Y_gd＝-Y₁₂＝G_gdf+jwC_gd

<math> <mrow> <msub> <mi>Y</mi> <mrow> <mi>g</mi> <mi>m</mi> </mrow> </msub> <mo>=</mo> <msub> <mi>Y</mi> <mn>21</mn> </msub> <mo>-</mo> <msub> <mi>Y</mi> <mn>12</mn> </msub> <mo>=</mo> <mfrac> <mrow> <msub> <mi>g</mi> <mi>m</mi> </msub> <msup> <mi>e</mi> <mrow> <mo>-</mo> <mi>j</mi> <mi>w</mi> <mi>&tau;</mi> </mrow> </msup> </mrow> <mrow> <msub> <mi>R</mi> <mi>i</mi> </msub> <msub> <mi>G</mi> <mrow> <mi>g</mi> <mi>s</mi> <mi>f</mi> </mrow> </msub> <mo>+</mo> <msub> <mi>jwC</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> <mo>+</mo> <mn>1</mn> </mrow> </mfrac> </mrow> </math>

Y_ds＝Y_i,22+Y_i,12＝G_ds+jwC_ds

defining: <math> <mrow> <mi>A</mi> <mo>=</mo> <mfrac> <msup> <mrow> <mo>|</mo> <msub> <mi>Y</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> <mo>|</mo> </mrow> <mn>2</mn> </msup> <mrow> <mi>Im</mi> <mo>&lsqb;</mo> <msub> <mi>Y</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> <mo>&rsqb;</mo> </mrow> </mfrac> <mo>=</mo> <mfrac> <mrow> <msup> <msub> <mi>G</mi> <mrow> <mi>g</mi> <mi>s</mi> <mi>f</mi> </mrow> </msub> <mn>2</mn> </msup> </mrow> <mrow> <msub> <mi>wC</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> </mrow> </mfrac> <mo>+</mo> <msub> <mi>wC</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> </mrow> </math>

<math> <mrow> <mi>B</mi> <mo>=</mo> <mfrac> <msub> <mi>Y</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> <mrow> <mi>Im</mi> <mo>&lsqb;</mo> <msub> <mi>Y</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> <mo>&rsqb;</mo> </mrow> </mfrac> <mo>=</mo> <mfrac> <mrow> <msub> <mi>G</mi> <mrow> <mi>g</mi> <mi>s</mi> <mi>f</mi> </mrow> </msub> <mrow> <mo>(</mo> <mn>1</mn> <mo>+</mo> <msub> <mi>R</mi> <mi>i</mi> </msub> <msub> <mi>G</mi> <mrow> <mi>g</mi> <mi>s</mi> <mi>f</mi> </mrow> </msub> <mo>)</mo> </mrow> </mrow> <mrow> <msub> <mi>wC</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> </mrow> </mfrac> <mo>+</mo> <msub> <mi>wR</mi> <mi>i</mi> </msub> <msub> <mi>C</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> <mo>-</mo> <mi>j</mi> </mrow> </math>

$C={\left|\frac{Y_{gs}}{Y_{gm}}\right|}^2={\left(\frac{G_{gsf}}{g_m}\right)}^2+{\left(\frac{C_{gs}}{g_m}\right)}^2{w}^2$

<math> <mrow> <mi>D</mi> <mo>=</mo> <mrow> <mo>(</mo> <msub> <mi>G</mi> <mrow> <mi>g</mi> <mi>s</mi> <mi>f</mi> </mrow> </msub> <mo>+</mo> <msub> <mi>jwC</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> <mo>)</mo> </mrow> <mfrac> <msub> <mi>Y</mi> <mrow> <mi>g</mi> <mi>m</mi> </mrow> </msub> <msub> <mi>Y</mi> <mrow> <mi>g</mi> <mi>s</mi> </mrow> </msub> </mfrac> <mo>=</mo> <msub> <mi>g</mi> <mi>m</mi> </msub> <msup> <mi>e</mi> <mrow> <mo>-</mo> <mi>j</mi> <mi>w</mi> <mi>&tau;</mi> </mrow> </msup> </mrow> </math>

for omega²Linearly fitting the-A omega curve to obtain Cgs; combining the obtained Cgs for ω B to ω²Ri can be obtained by linear fitting of the curve; determining Ggsf according to the real part of Ygs; according to formula Y_gd＝-Y₁₂＝G_gdf+jwC_gdObtaining Cgd and Ggdf by the corresponding relation of the medium complex numbers; combining the obtained Cgs, for ω²The gm can be obtained by linear fitting of the-C curve; carrying out linear fitting on the omega-D curve to obtain tau; according to formula Y_ds＝Y_i,22+Y_i,12＝G_ds+jwC_dsGds and Cds can be obtained by the corresponding relation between the real part and the imaginary part of the medium complex number.

7. An AlGaN/GaNHETM small-signal model, which is characterized in that the AlGaN/GaNHETM small-signal model comprises a parasitic part and an intrinsic part; the intrinsic portion includes a gate-drain leakage resistance Rgdf adapted to characterize gate-to-drain leakage and a gate-source leakage resistance Rgsf adapted to characterize gate-to-source leakage.

8. The AlGaN/GaNHETM small-signal model of claim 7, wherein:

the intrinsic part comprises intrinsic capacitors Cgd, Cgs and Cds, a gate source leakage resistor Rgsf, a gate drain leakage resistor Rgdf, a channel resistor Ri, a source drain resistor Rds and a transconductance Gm; wherein,

the intrinsic capacitor Cgs is connected with the gate source leakage resistor Rgsf in parallel and then connected with the channel resistor Ri in series to form a series structure;

the intrinsic capacitor Cds is connected with the source-drain resistor Rds and the transconductance Gm in parallel and then connected with the intrinsic capacitor Cgd in series to form a series structure, and the gate-drain leakage resistor Rgdf is connected with the intrinsic capacitor Cgd in parallel;

the series structure formed by the intrinsic capacitor Cds, the source-drain resistor Rds, the transconductance Gm and the intrinsic capacitor Cgd is connected in parallel with the series structure formed by the intrinsic capacitor Cgs, the gate-source leakage resistor Rgsf and the channel resistor Ri.

9. The AlGaN/gan hettm small signal model of claim 8, wherein:

the parasitic part comprises parasitic PAD capacitors Cpg, Cpd and Cpgd, parasitic inductors Lg, Ld and Ls, and parasitic resistors Rg, Rd and Rs; wherein,

the parasitic inductance Lg and the parasitic resistance Rg are connected in series to form a series structure, one end of the series structure formed by the parasitic inductance Lg and the parasitic resistance Rg is connected to one end of the parasitic PAD capacitance Cpg, and the other end of the series structure is connected between the intrinsic capacitances Cgd and Cgs; the other end of the parasitic PAD capacitance Cpg is grounded;

the parasitic inductor Ld and the parasitic resistor Rd are connected in series to form a series structure, one end of the series structure formed by the parasitic inductor Ld and the parasitic resistor Rd is connected to one end of the parasitic PAD capacitor Cpd, and the other end of the series structure is connected between the intrinsic capacitors Cgd and Cds; the other end of the parasitic PAD capacitor Cpd is grounded;

one end of the parasitic PAD capacitor Cpgd is connected to the connection end of the parasitic PAD capacitor Cpg and the series structure formed by the parasitic inductor Lg and the parasitic resistor Rg, and the other end of the parasitic PAD capacitor Cpgd is connected to the connection end of the parasitic PAD capacitor Cpd and the series structure formed by the parasitic inductor Ld and the parasitic resistor Rd;

the parasitic inductor Ls and the parasitic resistor Rs are connected in series to form a series structure, one end of the series structure formed by the parasitic inductor Ls and the parasitic resistor Rs is connected to one end, far away from the intrinsic capacitor Cgs, of the channel resistor Ri, and the other end of the series structure is grounded.