Ultra-wideband gradient line power divider and design method thereof

Abstract

The invention relates to an ultra-wideband tapered line power divider and a design method thereof, and the design method comprises the following steps: obtaining the design requirements of the tapered line power divider, and selecting a tapered line to achieve transmission; modeling is carried out on the gradual change line, and the line width is calculated; calculating an odd-even mode impedance sequence; based on a gradient line isolation resistance formula, determining an isolation resistance sequence about a gradient line position, and regionally dividing the isolation resistance sequence into isolation resistance films; establishing a gradual change line and isolation resistor unit length model, analyzing the attenuation coefficient, cascading a plurality of gradual change line and isolation resistor unit length models, and calculating the insertion loss of the total power divider; and judging whether performance index requirements are met or not based on the insertion loss and the isolation degree of the total power divider, if so, establishing a total model of the broadband gradient line power divider and performing simulation, and otherwise, adjusting the isolation resistor sequence. Compared with the prior art, the design method provided by the invention has relatively high flexibility in reflection and isolation performance, and the designed ultra-wideband tapered line power divider has the advantage of high isolation.

Description

Technical Field

[0001] The present invention relates to the field of passive device power dividers, and in particular to an ultra-wideband gradient line power divider and a design method thereof. Background Art

[0002] Broadband Wilkinson power splitters are widely used in wireless and measurement systems. The main performance characteristics of a two-port splitter include transmission, reflection, and isolation. For a two-port matching network with different port impedances, multi-section impedance transformers and tapered line impedance transformers are used to improve broadband transmission and reflection performance. The reflection bandwidth of a multi-section impedance transformer is limited by the number of sections, while a tapered line can theoretically exhibit infinite reflection bandwidth. To improve bandwidth, CN113904088A discloses a design method for an ultra-wideband power splitter based on a vertically mounted substrate. First, using the circuit structure given by the equivalent circuit diagram, a multi-section coupled line is connected in series to simulate the actual tapered line. The odd and even mode impedance values ​​of each node are obtained through simulation and optimization using two-dimensional electromagnetic simulation software. The obtained odd and even mode impedances are then divided into strong coupling parts and weak coupling parts. Finally, the physical dimensions of each node obtained in the previous steps are combined to fit a function curve, which is then modeled, simulated, and optimized using three-dimensional electromagnetic simulation software. The present invention achieves equal power distribution based on a vertically mounted substrate circuit structure. The overall structure achieves broadband impedance transformation based on tapered coupling lines, thereby realizing the function of ultra-wideband power distribution. However, when transmission line impedance converters, such as multi-section impedance converters and tapered line impedance converters, are used in power splitter designs, it is difficult to achieve a match between isolation bandwidth and reflection bandwidth. This problem is particularly prominent in tapered line power splitters, resulting in the theoretically infinite reflection bandwidth of the tapered line not being fully utilized. Furthermore, transmission lines with isolation resistor films are actually lossy, and a certain amount of current will flow through the isolation resistors or the resistor films. However, existing designs only consider isolation, not insertion loss. Summary of the Invention

[0003] The purpose of the present invention is to provide an ultra-wideband tapered line power divider and a design method thereof. First, a tapered line isolation resistance formula based on a multi-section impedance transformer and a tapered line impedance transformer position is proposed. This formula can be used to design discrete lumped resistors and continuous isolation resistor films. It has theoretically infinite bandwidth in a lossless case. Based on this, taking into account the transmission, reflection and isolation performance, an ultra-wideband tapered line high-isolation power divider divided into two is designed, and its isolation resistor uses a resistor film form.

[0004] The purpose of the present invention can be achieved by the following technical solutions:

[0005] A design method for an ultra-wideband gradient line power divider comprises the following steps:

[0006] S1, obtain the design requirements of the gradient line power divider and select the gradient line to achieve transmission;

[0007] S2, modeling the gradient line and calculating the line width;

[0008] S3, calculate the odd and even mode impedance sequence based on the modeling results of the gradient line;

[0009] S4, determining an isolation resistor sequence at the gradient line position based on the odd-and-even mode impedance sequence and the gradient line isolation resistance formula, and dividing the isolation resistor sequence into isolation resistor films;

[0010] S5, establishing a gradient line and isolation resistor unit length model, analyzing its attenuation coefficient, cascading multiple gradient line and isolation resistor unit length models, and calculating the total power divider insertion loss;

[0011] S6, based on the total power divider insertion loss and isolation, determine whether the performance index requirements are met. If so, divide the isolation resistor sequence into segments, take the average value of the resistance value of each segment, establish the overall model of the broadband gradient line power divider and perform simulation. Otherwise, adjust the isolation resistor sequence and return to step S4, and divide the adjusted isolation resistor sequence into isolation resistor films.

[0012] The design requirements include reflection coefficient, line length, isolation and insertion loss.

[0013] The step S2 comprises the following steps:

[0014] S21, set the basic parameters: port impedance, load impedance, wavelength, total length of the gradient line and the maximum reflection coefficient within the passband, and create a length partition array based on the total length of the gradient line to segment the gradient line;

[0015] S22, calculating gradient line impedance: calculating a load reflection coefficient based on the port impedance and the load impedance, calculating a ripple value within the passband based on a maximum reflection coefficient limited within the passband, calculating an intermediate variable based on the load reflection coefficient and the ripple value within the passband, and calculating gradient line impedance based on the port impedance, the load impedance, the load reflection coefficient, and the intermediate variable;

[0016] S23, calculating line width: calculating the line width of each gradient line segment;

[0017] S24, based on the calculated gradient line impedance and line width, call Matlab-HFSS-API to model the gradient line, set simulation parameters and perform simulation, and analyze the reflection and transmission characteristics of the gradient line within the operating frequency range.

[0018] The method for calculating line width is:

[0019] Based on the dielectric constant E rAnd the port impedance Z0 are used to calculate the first coefficient A and the second coefficient B respectively:

[0020]

[0021] Calculate the first line width W based on the first coefficient dA :

[0022]

[0023] The second line width W is calculated based on the second coefficient and the dielectric constant dB :

[0024]

[0025] If the first line width is smaller than a preset threshold, the first line width is used as the gradient line width; otherwise, the second line width is used as the gradient line width.

[0026] The method of calling Matlab-HFSS-API to perform gradient line modeling is as follows: using Matlab to generate a HFSS VBS script, which is used to establish the geometric structure of the gradient line and set the port in HFSS, including the following steps:

[0027] According to the length and impedance value of the gradient line, the coordinates of each gradient line segment are calculated and a 2D multi-segment polyline is generated to represent the actual shape of the gradient line;

[0028] Use the hfssBox function to construct a dielectric area and specify the material type;

[0029] Construct a ground layer and assign it as a perfect electrical conductor for electromagnetic wave propagation through the hfssAssignPE function;

[0030] A waveguide port is provided between the dielectric area and the ground layer to ensure that a transmission signal enters or leaves the port;

[0031] Set the frequency range for calculation and use the hfssInsertSolution function and the hfssInterpolatingSweep function to perform frequency sweep within the frequency range.

[0032] The step S3 comprises the following steps:

[0033] S31, based on the modeling result of the gradient line, read the S parameter data set in the simulation result;

[0034] S32, calculate the Z parameter based on the S parameter: Z = Z0*(I+S) / (IS), where I is the identity matrix, S is the S parameter matrix, and Z is the Z parameter matrix. Divide the Z parameter matrix into four sub-matrices according to the size of the S parameter;

[0035] S33, calculating a first matrix and a second matrix based on a submatrix of the Z parameter matrix;

[0036] S34, performing eigenvalue decomposition on the first matrix to obtain an eigenvector matrix and an eigenvalue matrix;

[0037] S35, odd and even mode impedance calculation:

[0038] Normalize the eigenvector matrix;

[0039] The principal value of the transmission constant is calculated based on the eigenvalue matrix, and phase unfolding is performed to remove redundant up-jump points and down-jump points in the principal value of the transmission constant;

[0040] Calculate the complex propagation constant based on the eigenvector matrix and the principal value of the transmission constant;

[0041] calculating characteristic impedance based on the eigenvector matrix, principal values ​​of transmission constants, and a second matrix;

[0042] Calculates odd-mode and even-mode impedances based on the complex propagation constant and characteristic impedance.

[0043] The gradient line isolation resistance formula is:

[0044]

[0045] Among them, G k represents the conductance value, l represents the total line length, Δz represents the specified unit length of the gradient line, A represents the maximum reflection coefficient of the gradient line in the passband, Γ0 represents the terminal reflection coefficient, T i represents the transmission coefficient between each section, F represents the adjustment factor, k represents the sequence number of each section, and I1(x) function represents the first-order modified Bessel function.

[0046] A=acosh(Γ ld / Γ rip )

[0047]

[0048] Among them, Γ ld represents the load reflection coefficient, Γ rip Indicates the ripple value within the passband, a indicates a custom scale, Y k Represents the admittance value, which is the reciprocal of the gradient line isolation resistance series.

[0049] The conductance value is obtained by step-by-step iterative calculation. During the iterative process, the initial value of the adjustment factor is set to 1. When the number of gradient line nodes is too large and the conductance value becomes negative, the value of the adjustment factor is gradually increased so that the adjusted conductance value becomes positive.

[0050] The calculation method of the attenuation coefficient is:

[0051]

[0052] Among them, γ e and γ o They represent the propagation constants of even mode and odd mode respectively, and real represents the real part of the result.

[0053] An ultra-wideband gradient line power divider is designed by the method described above.

[0054] Compared with the prior art, the present invention has the following beneficial effects:

[0055] 1. This invention proposes a gradient isolation resistor formula suitable for infinite-bandwidth gradient Wilkinson power dividers. This formula is also applicable to multi-section impedance transformer power dividers. By introducing an adjustment factor, it balances reflection and isolation performance, providing greater flexibility in the design of isolation resistors that consider both properties.

[0056] 2. The present invention analyzes the influence of resistor films with different resistance values ​​on the insertion loss of the total circuit, thereby taking the influence of the insertion loss into consideration in the design of the power divider, thereby improving the performance of the power divider.

[0057] 3. The present invention extracts the complex propagation constant of the lossy coupled line with an isolation resistor film, providing a basis for selecting between transmission, reflection and isolation performance.

[0058] 4. The high-isolation ultra-wideband one-to-two gradient line power divider designed in the present invention shows excellent performance in terms of reflection, isolation, and insertion loss in broadband response. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 is a flow chart of the design method of the present invention;

[0060] Figure 2 Model diagram of unit length gradient line and isolation resistor film, where (a) is a top view and (b) is a side view;

[0061] Figure 3 Schematic diagram of the effect of different resistance values ​​of resistor films on the attenuation coefficient in one embodiment;

[0062] Figure 4 A diagram of a simulation overall model in one embodiment;

[0063] Figure 5 This is a test diagram of the S parameters of the simulation total model in one embodiment;

[0064] Figure 6 A physical picture of the ultra-wideband gradient line power divider designed for the present invention;

[0065] Figure 7 This is an actual S-parameter test diagram of a physical ultra-wideband gradient line power divider in an embodiment. DETAILED DESCRIPTION

[0066] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0067] This embodiment provides a design method for an ultra-wideband gradient line power divider. Figure 1 As shown, the following steps are included:

[0068] S1. Obtain the design requirements of the gradient line power divider (including reflection coefficient, line length, isolation, and insertion loss) and select the Klopfenstein gradient line to achieve transmission.

[0069] S2, use Matlab-HFSS-API to model the gradient line and calculate the line width.

[0070] Specifically, S2 includes the following steps:

[0071] S21, set the basic parameters: port impedance, load impedance, wavelength, total length of the gradient line and the maximum reflection coefficient limited within the passband, and create a length partition array based on the total length of the gradient line to segment the gradient line.

[0072] In this embodiment, the port impedance z0 is set to 50Ω and the load impedance z l = 100Ω. Assume the operating frequency is 3 GHz, the wavelength λ = 300 / 3 = 100 mm. The total length of the gradient is set to L = 150 mm. The maximum reflection coefficient within the passband is RdB = -20 dB.

[0073] S22, calculate the gradient line impedance.

[0074] The specific steps include:

[0075] S221, based on the port impedance Z0 and the load impedance Z l Calculate the load reflection coefficient Γ ld :Γ ld =ln(Z l / Z0)*0.5;

[0076] S222, calculating the ripple value Γ within the passband based on the maximum reflection coefficient RdB defined within the passband rip :Γ rip=10 RdB / 20 ;

[0077] S223, calculate the intermediate variable X based on the load reflection coefficient and the ripple value in the passband: X=acosh(Γ ld / Γ rip );

[0078] S224, based on port impedance Z0, load impedance Z l , load reflection coefficient Γ ld And the intermediate variable X calculates the gradient line impedance:

[0079]

[0080] Where Psi represents the modified Bessel function, Besseli is a commonly used Bessel function.

[0081] S23, calculating line width: calculating the line width of each divided gradient line segment.

[0082] The specific steps include:

[0083] S231, based on dielectric constant E r And the port impedance Z0 are used to calculate the first coefficient A and the second coefficient B respectively:

[0084]

[0085] S232, calculating a first line width W based on the first coefficient dA :

[0086]

[0087] S233, calculating a second line width W based on the second coefficient and the dielectric constant dB :

[0088]

[0089] S234, if the first line width W dA If the first line width W is less than the preset threshold 2, dA As the gradient line width, otherwise, the second line width W dB As the line width of the gradient line.

[0090] In this embodiment, the dielectric constant E r =2.2.

[0091] S24, based on the calculated gradient line impedance and line width, call Matlab-HFSS-API to model the gradient line, set simulation parameters and perform simulation, and analyze the reflection and transmission characteristics of the gradient line within the operating frequency range.

[0092] Specifically, use Matlab to generate a HFSS VBS script. This script is used to establish the geometric structure of the gradient line and set the port in HFSS. The script includes the following steps:

[0093] S241, calculating the coordinates of each gradient line segment based on the gradient line length and impedance value, and generating a 2D multi-segment polyline representing the actual shape of the gradient line;

[0094] S242, use the hfssBox function to construct a medium area and specify the material type;

[0095] S243, construct a ground layer and assign it as a perfect electrical conductor for electromagnetic wave propagation through the hfssAssignPE function;

[0096] S244, setting a waveguide port between the dielectric area and the ground layer to ensure that the transmission signal enters or leaves from the port;

[0097] S245 , setting the calculated frequency range to be from 2 GHz to 10 GHz, and performing a frequency sweep within the frequency range using the hfssInsertSolution function and the hfssInterpolatingSweep function.

[0098] S3, based on the modeling results of the gradient line, calculate the odd and even mode impedance sequence.

[0099] S3 includes the following steps:

[0100] S31, based on the modeling result of the gradient line, reads the S parameter data set in the simulation result.

[0101] S32. Calculate the Z parameter based on the S parameter: Z = Z0*(I+S) / (IS), where I is the identity matrix, S is the S parameter matrix, and Z is the Z parameter matrix. Divide the Z parameter matrix into four sub-matrices according to the size N of the S parameter:

[0102] Z 11 =Z([1N / 2],[1N / 2],fre);

[0103] Z 12 =Z([1N / 2],[N / 2+1N],fre);

[0104] Z 21 =Z([N / 2+1N],[1N / 2],fre);

[0105] Z 22 =Z([N / 2+1N],[N / 2+1N],fre)

[0106] Where fre is the current frequency.

[0107] S33, calculate the first matrix P and the second matrix Q based on the submatrix of the Z parameter matrix:

[0108] P=Z 11 / Z 21 ;

[0109] Q=Z 11 / Z 21 *Z 22 -Z 12 .

[0110] S34, performing eigenvalue decomposition on the first matrix P to obtain an eigenvector matrix E and an eigenvalue matrix VA.

[0111] S35, odd and even mode impedance calculation:

[0112] S351, normalizing the eigenvector matrix;

[0113] S352, calculating the principal value of the transmission constant based on the eigenvalue matrix VA: γ = acos(VA), and performing phase unfolding to remove redundant up-jump points and down-jump points in the principal value of the transmission constant;

[0114] S353, calculate the complex propagation constant Γ based on the eigenvector matrix E and the principal value γ of the transmission constant:

[0115] Γ=EγE -1 =Ediag(γ1,γ2,…,γ N )E -1

[0116] S354, calculating the characteristic impedance Z based on the eigenvector matrix E, the principal value γ of the transmission constant and the second matrix Q c :

[0117] Z c =E(sinhPV(γl)) -1 E -1 Q

[0118] S355, calculates odd-mode impedance and even-mode impedance based on the complex propagation constant and characteristic impedance:

[0119]

[0120] Here, ω represents the frequency.

[0121] S4, determining an isolation resistor sequence at the gradient line position based on the odd-and-even mode impedance sequence and the gradient line isolation resistor formula, and dividing the isolation resistor sequence into isolation resistor films.

[0122] The gradient line isolation resistance formula is:

[0123]

[0124] Among them, G k Indicates the conductance value, which is the reciprocal of the resistance value, G k =1 / R k , R k represents the isolation resistance value of the gradient line, l represents the total line length, Δz represents the specified unit length of the gradient line, usually 1mm, A represents the maximum reflection coefficient of the gradient line in the passband, Γ0 represents the terminal reflection coefficient, usually 1 / 3, T i represents the transmission coefficient between each section, F represents the adjustment factor, k represents the sequence number of each section, and I1(x) function represents the first-order modified Bessel function.

[0125] A=acosh(Γ ld / Γ rip )

[0126]

[0127] Among them, Γ ld represents the load reflection coefficient, Γ rip Indicates the ripple value within the passband, a indicates a custom scale, Y k Indicates the admittance value, which is the reciprocal of the gradient line isolation resistance series, Y k =1 / Z k .

[0128] When there are too many gradient nodes, the conductance value may become negative. Initially, F is 1. If a negative value appears, gradually increasing F will recalculate a positive value. When using this formula, the required conductance value is generated gradually and iteratively.

[0129] S5, establish a gradient line and isolation resistor unit length model, analyze its attenuation coefficient, cascade multiple gradient line and isolation resistor unit length models, and calculate the total power divider insertion loss.

[0130] The attenuation coefficient is calculated as:

[0131]

[0132] Among them, γ e and γ o They represent the propagation constants of the even mode and odd mode respectively, and real represents the real part of the result.

[0133] After calculating the insertion loss of each section, the total insertion loss of the cascade is the sum of the insertion loss results of each section.

[0134] S6, based on the total power divider insertion loss and isolation, determine whether the performance index requirements are met. If so, divide the isolation resistor sequence into segments, take the average value of the resistance value of each segment, establish the overall model of the broadband gradient line power divider and perform simulation. Otherwise, adjust the isolation resistor sequence and return to step S4, and divide the adjusted isolation resistor sequence into isolation resistor films.

[0135] In this embodiment, the isolation resistor sequence is adjusted by adjusting the adjustment factor F in the gradient line isolation resistance formula. If the insertion loss is too high and the isolation performance is excessive, the value of F is increased in steps of 0.01. Conversely, if the isolation performance is insufficient, the value of F is decreased in steps of 0.01.

[0136] This paper proposes a novel isolation resistor design formula based on the position of multi-section impedance transformers and tapered impedance transformers. This formula is used to design a two-way ultra-wideband tapered high-isolation power splitter. This formula can be used to design discrete lumped resistors and continuous isolation resistor films, theoretically achieving infinite bandwidth in a lossless design. Furthermore, the impact of different resistance values ​​on the total circuit insertion loss is analyzed based on the resistance films.

[0137] In this embodiment, the above method is simulated by MATLAB and verified by HFSS software. Figure 2 The figure shows a transmission line model with an isolation resistor film. The specific dimensions are: W C =1.24mm,L T =3mmL C =1mm, the model is represented by a stripline, the upper and lower substrates are both made of Rogers5880 material with a thickness of 0.787mm, the copper wire and the ground are both made of copper material with a thickness of 0.035mm, and the subsequent models also use this structure and parameters. Figure 3 The effect of resistor films with different resistance values ​​on the attenuation coefficient of the equivalent "lossy" transmission line is provided, which further affects the insertion loss of the total circuit. Figure 4 and Figure 5 The simulation model and simulation test results are shown. Figure 4 The resistance values ​​from right to left are 65,000 ohms, 40,000 ohms, 15,000 ohms and 8,000 ohms. Figure 5 It can be found that the present invention shows an outstanding performance response. Considering the transmission, reflection and isolation performance, this embodiment also provides an ultra-wideband gradient line power divider divided into two, which is designed and processed by the above method, and its isolation resistor uses a resistive film form, such as Figure 6 The measured results of the ultra-wideband gradient line power divider are shown in Figure 7 As shown, it can be found that it exhibits outstanding performance response, which fully meets the performance requirements of ultra-wideband power dividers on the market.

[0138] The above describes in detail the preferred embodiments of the present invention. It should be understood that those skilled in the art can make numerous modifications and variations based on the concepts of the present invention without inventive effort. Therefore, any technical solutions that can be derived by those skilled in the art through logical analysis, reasoning, or limited experimentation based on the concepts of the present invention and the prior art should be within the scope of protection defined by the claims.

Claims

1. A design method for an ultra-wideband gradient line power divider, characterized in that: The following steps are involved: S1, obtain the design requirements of the gradient line power divider and select the gradient line to achieve transmission; S2, modeling the gradient line and calculating the line width; S3, calculate the odd and even mode impedance sequence based on the modeling results of the gradient line; S4, determining an isolation resistor sequence at the gradient line position based on the odd-and-even mode impedance sequence and the gradient line isolation resistance formula, and dividing the isolation resistor sequence into isolation resistor films; S5, establishing a gradient line and isolation resistor unit length model, analyzing its attenuation coefficient, cascading multiple gradient line and isolation resistor unit length models, and calculating the total power divider insertion loss; S6, based on the total power divider insertion loss and isolation, determine whether the performance index requirements are met. If so, divide the isolation resistor sequence into segments, take the average value of the resistance value of each segment, establish the overall model of the broadband gradient line power divider and perform simulation. Otherwise, adjust the isolation resistor sequence and return to step S4, and divide the adjusted isolation resistor sequence into isolation resistor films.

2. The design method of an ultra-wideband gradient line power divider according to claim 1, characterized in that: The design requirements include reflection coefficient, line length, isolation and insertion loss.

3. The design method of an ultra-wideband gradient line power divider according to claim 1, characterized in that: The step S2 comprises the following steps: S21, set the basic parameters: port impedance, load impedance, wavelength, total length of the gradient line and the maximum reflection coefficient within the passband, and create a length partition array based on the total length of the gradient line to segment the gradient line; S22, calculating gradient line impedance: calculating a load reflection coefficient based on the port impedance and the load impedance, calculating a ripple value within the passband based on a maximum reflection coefficient limited within the passband, calculating an intermediate variable based on the load reflection coefficient and the ripple value within the passband, and calculating gradient line impedance based on the port impedance, the load impedance, the load reflection coefficient, and the intermediate variable; S23, calculating line width: calculating the line width of each gradient line segment; S24, based on the calculated gradient line impedance and line width, call Matlab-HFSS-API to model the gradient line, set simulation parameters and perform simulation, and analyze the reflection and transmission characteristics of the gradient line within the operating frequency range.

4. The design method of an ultra-wideband gradient line power divider according to claim 3, characterized in that: The method for calculating line width is: Based on the dielectric constant E r And the port impedance Z0 are used to calculate the first coefficient A and the second coefficient B respectively: Calculate the first line width W based on the first coefficient dA : The second line width W is calculated based on the second coefficient and the dielectric constant dB : If the first line width is smaller than a preset threshold, the first line width is used as the gradient line width; otherwise, the second line width is used as the gradient line width.

5. The design method of an ultra-wideband gradient line power divider according to claim 3, characterized in that: The method of calling Matlab-HFSS-API to perform gradient line modeling is as follows: using Matlab to generate a HFSS VBS script, which is used to establish the geometric structure of the gradient line and set the port in HFSS, including the following steps: According to the length and impedance value of the gradient line, the coordinates of each gradient line segment are calculated and a 2D multi-segment polyline is generated to represent the actual shape of the gradient line; Use the hfssBox function to construct a dielectric area and specify the material type; Construct a ground layer and assign it as a perfect electrical conductor for electromagnetic wave propagation through the hfssAssignPE function; A waveguide port is provided between the dielectric area and the ground layer to ensure that a transmission signal enters or leaves the port; Set the frequency range for calculation and use the hfssInsertSolution function and the hfssInterpolatingSweep function to perform frequency sweep within the frequency range.

6. The design method of an ultra-wideband gradient line power divider according to claim 1, characterized in that: The step S3 comprises the following steps: S31, based on the modeling result of the gradient line, read the S parameter data set in the simulation result; S32, calculate the Z parameter based on the S parameter: Z = Z0*(I+S) / (IS), where I is the identity matrix, S is the S parameter matrix, and Z is the Z parameter matrix. Divide the Z parameter matrix into four sub-matrices according to the size of the S parameter; S33, calculating a first matrix and a second matrix based on a submatrix of the Z parameter matrix; S34, performing eigenvalue decomposition on the first matrix to obtain an eigenvector matrix and an eigenvalue matrix; S35, odd and even mode impedance calculation: Normalize the eigenvector matrix; The principal value of the transmission constant is calculated based on the eigenvalue matrix, and phase unfolding is performed to remove redundant up-jump points and down-jump points in the principal value of the transmission constant; Calculate the complex propagation constant based on the eigenvector matrix and the principal value of the transmission constant; calculating characteristic impedance based on the eigenvector matrix, principal values ​​of transmission constants, and a second matrix; Calculates odd-mode and even-mode impedances based on the complex propagation constant and characteristic impedance.

7. The design method of an ultra-wideband gradient line power divider according to claim 1, characterized in that: The gradient line isolation resistance formula is: Among them, G k represents the conductance value, l represents the total line length, Δz represents the specified unit length of the gradient line, A represents the maximum reflection coefficient of the gradient line in the passband, Γ0 represents the terminal reflection coefficient, T i represents the transmission coefficient between each section, F represents the adjustment factor, k represents the sequence number of each section, and I1(x) function represents the first-order modified Bessel function. A=acosh(Γ ld / C rip ) Among them, Γ ld represents the load reflection coefficient, Γ rip Indicates the ripple value within the passband, a indicates a custom scale, Y k Represents the admittance value, which is the reciprocal of the gradient line isolation resistance series.

8. The design method of an ultra-wideband gradient line power divider according to claim 7, characterized in that: The conductance value is obtained by step-by-step iterative calculation. During the iterative process, the initial value of the adjustment factor is set to 1. When the number of gradient line nodes is too large and the conductance value becomes negative, the value of the adjustment factor is gradually increased so that the adjusted conductance value becomes positive.

9. The design method of an ultra-wideband gradient line power divider according to claim 1, characterized in that: The calculation method of the attenuation coefficient is: Among them, γ e and γ o They represent the propagation constants of even mode and odd mode respectively, and real represents the real part of the result.

10. An ultra-wideband gradient line power divider, characterized in that: The ultra-wideband gradient line power divider is designed using the method described in any one of claims 1 to 9.

Images