Simulation design method applied to simulation surface acoustic wave filter

Abstract

The invention provides a simulation design method applied to a simulation surface acoustic wave filter, and the method comprises the steps: measuring an actual surface acoustic wave filter model, obtaining a Y parameter of the filter, and carrying out the de-embedding processing of an admittance characteristic curve; according to the admittance characteristic curve, COM model parameters are extracted, and static capacitance is extracted; coupling mode model data simulation is carried out according to the extracted COM model parameters, and comparison with actually measured data is carried out; carrying out geometric modeling on the SAW filter to obtain a multi-physical field model and a grid thereof; adding corresponding boundary conditions and excitation conditions to the geometric model, and generating a grid file; importing the grid file into a spectral element algorithm, and saving calculation time by adopting distributed operation; and comparing the generated simulation curve of the COM model with a simulation curve obtained by a spectral element algorithm and actually measured data. And the problem of overlong simulation time of an existing acoustic wave filter is greatly solved while the simulation precision is not influenced.

Description

Technical Field

[0001] The invention belongs to the technical field of radio frequency chips, and in particular relates to a simulation design method for simulating surface acoustic wave filters. Background Art

[0002] In recent years, with the rapid development of communication technology and the popularization of electronic devices, the performance requirements for RF modules have become increasingly higher. In RF modules, surface acoustic wave filters (SAW filters), as an important type of filter, are widely used in wireless communications, radar, television, wireless sensors and other fields. The performance of SAW filters directly affects the performance and stability of the entire RF module.

[0003] However, due to the complexity of materials and limitations of manufacturing processes, the design and optimization process of SAW filters remains challenging. Traditional design methods usually rely on limited theoretical models, empirical parameters, and finite element simulations, but multi-physics simulation software takes up too many computing resources. Therefore, the layout design of new structures relies more on coupled mode algorithms for simulation iterations, and finally obtains the optimal geometric parameters of the SAW filter and then performs spectral element method simulation to optimize the thermodynamic and electrical characteristics.

[0004] In view of this, it is very meaningful to propose a simulation design method for simulating surface acoustic wave filters. Summary of the invention

[0005] In order to solve the problems of excessive calculation amount and long calculation time in the existing SAW model simulation, the present invention provides a simulation design method for simulating surface acoustic wave filters to solve the above-mentioned technical defects.

[0006] In a first aspect, the present invention provides a method for simulation design of a surface acoustic wave filter, the method comprising the following steps:

[0007] Measuring an actual surface acoustic wave filter model, obtaining Y parameters of the filter, including open-circuit and short-circuit Y parameters, and de-embedding the admittance characteristic curve;

[0008] Extracting COM model parameters and static capacitance according to the admittance characteristic curve;

[0009] Simulating coupled mode model data according to the extracted COM model parameters and comparing the data with the measured data;

[0010] Geometric modeling of the SAW filter, including setting the relevant parameters of the filter model and the calculation area to obtain the multi-physics field model and its mesh;

[0011] Adding corresponding boundary conditions and excitation conditions to the geometric model and generating a mesh file;

[0012] Importing the grid file into the spectral element algorithm and adopting distributed computing to save computing time;

[0013] The generated simulation curve of the COM model is compared with the simulation curve obtained by the spectral element algorithm and the measured data.

[0014] Preferably, the obtaining of the Y parameters of the filter, including the Y parameters of the open circuit and the short circuit, comprises: Y transistor =((Y dut -Y open ) -1 -(Y short -Y open ) -1 ) -1 , where Y dut is the overall admittance curve of the device, Y open is the interdigital open circuit admittance curve, Y short Preferably, the extraction of COM model parameters according to the admittance characteristic curve includes: obtaining f r 、f ar , f0 and Z imag , f r and f ar are the resonant frequency and anti-resonant frequency of the admittance curve, respectively. f0 is a low frequency point far from the resonant frequency (linear change area), which is used to extract the static capacitance, Z imag is the imaginary part of the impedance corresponding to this point, and then the COM model parameters are calculated using the following formula:

[0015]

[0016]

[0017] Where η is a characterization of the average propagation characteristics of the surface acoustic wave device in the resonant and antiresonant states, |ε| represents the coupling strength difference parameter in the surface acoustic wave device, v(δ0) represents the propagation ratio or proportion of energy in the device under specific conditions, and δ B and δ V An intermediate variable representing some special frequency characteristics or energy states inside a surface acoustic wave device. к represents the energy coupling between different parts inside the device, including amplitude and phase information. B It is used to calculate and analyze the energy coupling strength of the device under a specific working state and its relationship with other parameters, η BIt represents an enhancement factor of equivalent propagation or coupling capability under specific conditions. δ is a comprehensive parameter containing multiple variables related to the working characteristics of the surface acoustic wave device. ω is the angular frequency, p is the electrode period, and v B is the wave velocity, and γ is the Q value loss coefficient.

[0018] Preferably, the extraction of the static capacitance includes: C t is the static capacitance of the IDT device, which is affected by the number of interdigital logarithms and numerical aperture. The low frequency point f0 and its corresponding imaginary impedance Z imag C can be obtained by the following formula t :C t =n*W*∈,∈=p1*∈0, Where n is the number of interdigital pairs, W is the numerical aperture, and ∈0 is the vacuum dielectric constant.

[0019] Preferably, simulating coupled mode model data according to the extracted COM model parameters and comparing with measured data includes: establishing a coupled mode model for the surface acoustic wave filter and establishing a differential equation for its acoustic wave propagation phenomenon: Express the differential equation as a matrix operation: The P matrix unit is represented as an interdigital resonator, and the expressions of each element of the P matrix are as follows:

[0020]

[0021] P 33 =-K1P 31 -K2P 32 exp(jk c L)+2(α2K1-α1K2)L+jωC t

[0022] Where D = s cos (sL) + jδsin (sL0, c 12 =c 21 · =к is the reflection coefficient, α1=α2 * is the coupling coefficient, k c is the wave number, L is the resonator length, generally L = 2n*λ, n is the number of IDTs; λ = 2p, p is the electrode period; C t is the static capacitance of the IDT device, C t =C0*L*W;P 11 and P 22 represents the reflection coefficient of the interdigital electrode, P 12 =P 21represents the transmission coefficient of the interdigital electrode, P 13 and P 23 represents electromechanical coupling, which is the transfer function between the electrical quantity and the acoustic quantity at the left and right ends, P 13 and P 23 Represents the transfer function from acoustic quantity to electrical quantity, P 33 It is an electrical quantity, namely, input admittance.

[0023] Further preferably, the method further includes: assembling the obtained P matrix of the reflection grating, gap, and IDT unit into a series-parallel connection of the filter according to the transmission function of the ABCD matrix, combining the coupled mode method and the transmission matrix to realize the simulation of the surface acoustic wave filter. For a ladder filter, according to the microwave network transmission, the matrix is ​​as follows: According to the ABCD matrix, the series and parallel resonators are expressed as follows:

[0024]

[0025] The ABCD matrix is ​​converted to S parameters according to the following equation:

[0026]

[0027] Among them, S 11 It represents the ratio of the input intensity to the reflected intensity of port 1 when the microwave signal is incident on port 1, and represents the attenuation of the signal. S 21 It indicates the microwave signal incident on port 1, the ratio of the input intensity of port 2 to the input intensity of port 1, and represents the signal throughput; S 22 Represents the microwave signal incident on the 2-port, the ratio of the 2-port input intensity to the reflected intensity, characterizing the signal attenuation, S 12 Indicates the microwave signal incident on 2 ports. The ratio of the input intensity of port 1 to the input intensity of port 2 characterizes the signal throughput.

[0028] Preferably, the geometric modeling of the SAW filter includes setting relevant parameters of the filter model and a calculation area to obtain a multi-physics field model and its grid, including:

[0029] The multi-physics field model is calculated using the spectral element method to obtain two calculation areas, namely the electric field and the displacement field, and the control matrix equation to be solved is obtained:

[0030]

[0031] Where ρ is the density of the piezoelectric material, ω is the operating angular frequency of the filter, u is the displacement vector of the deformation, τ is the stress tensor of the piezoelectric material, D is the electric displacement vector of the piezoelectric material, ρ e is the external charge density, Ω p is the electric field area, then Ωs+p is the displacement field area;

[0032] Under the constitutive relationship between τ and D, the following coupling will be achieved between the electroacoustic signals:

[0033] D=e:ε+∈·E,inΩ s

[0034] τ=c:ee·E,inΩ s+p

[0035] Among them, c is the fourth-order elastic coefficient matrix, e is the third-order piezoelectric coupling coefficient, ∈ is the second-order dielectric constant, ε is the strain tensor, E is the electric field intensity, Ω p is the electric field area, Ω s+p is the displacement field area;

[0036] where the electric field and strain tensors are related to the potential and displacement fields as follows:

[0037]

[0038]

[0039] Where V is the voltage field of the acoustic filter, thus realizing the mutual conversion relationship between the electroacoustic signal, Ω p is the electric field area, Ω s+p is the displacement field area.

[0040] Preferably, the spectral element algorithm solves the surface acoustic wave filter equation as follows:

[0041]

[0042] Among them, Ω is the entire calculation area, Ω p is the electric field area, Ω s+p is the displacement field region, Γ is the boundary surface, n is the boundary surface normal, and n·D represents the output voltage; the physical region Ω of the acoustic wave filter is divided into K non-overlapping hexahedral grid units, and then the grid units in the physical space are mapped to the reference unit through coordinate transformation, and numerical integration, derivation and interpolation are directly performed in the reference unit.

[0043] Preferably, the spectral element algorithm uses Lagrange function as basis function, selects Gauss-LobattoLegendre (GLL) points as interpolation points, and constitutes the matrix equation for solving the physical model by the spectral element algorithm. In the reference unit, the N-order basis function has N+1 GLL points, which are defined as:

[0044]

[0045] Where i∈[-1,1],i=1,2,…,N+1, is the i-th basis function, L N (ζ i ) is the Nth order Lagrange polynomial, L' N (ζ i ) is the first-order derivative of the N-th order Lagrange polynomial;

[0046] The assembled matrix A and the excitation source matrix b are obtained, and then the linear equations (Ax = b) are solved by direct method or iterative method to obtain the displacement field and voltage field of the entire model. The weak form is tested by Galerkin method, and the reference domain is integrated and discretized:

[0047]

[0048] Among them, M ij represents the mass matrix, ü j represents the acceleration of the node, is the external force, is the internal force due to the piezoelectric effect, K ij represents the stiffness matrix, and is the corrected stiffness matrix u j represents the displacement of the node, V j represents the node voltage, Q i is the load term related to the electric field.

[0049] Preferably, the generated simulation curve of the COM model is compared with the simulation curve obtained by the spectral element algorithm and the measured data, including: using the Q_bode method to calculate the Q value for comparison to verify the correctness of the simulation result, and the Q value formula is as follows:

[0050]

[0051] Among them, S 11 is the S parameter converted from the admittance curve, group_delay(S 11 ) is the group delay expression, that is, the gradient operation of the imaginary part of S11.

[0052] Compared with the prior art, the beneficial results of the present invention are:

[0053] The present invention proposes an acoustic wave filter simulation method based on a coupled mode method and a spectral element method. The method can simulate an actual model and filter composition without material parameters based on the coupled mode method. In addition, the spectral element method adopts the concepts of grid degradation, matrix equivalence and distributed registration, which greatly speeds up the current problem of long acoustic wave filter simulation time without affecting the simulation accuracy, and makes a certain contribution to the thermal and electrical simulation analysis of acoustic wave filters. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] The accompanying drawings are included to provide a further understanding of the embodiments and are incorporated into and constitute a part of this specification. The accompanying drawings illustrate the embodiments and are used together with the description to explain the principles of the present invention. It will be easy to recognize other embodiments and many expected advantages of the embodiments because they become better understood by reference to the following detailed description. The elements of the drawings are not necessarily to scale with each other. The same reference numerals refer to corresponding similar parts.

[0055] Figure 1 It is a flow chart of a method for simulation design of a simulated surface acoustic wave filter according to an embodiment of the present invention;

[0056] Figure 2 is a comparison diagram of measured data and de-embedded data of an embodiment of the present invention;

[0057] Figure 3 is an optical microscopic schematic diagram of a practical model of an embodiment of the present invention;

[0058] Figure 4 A schematic diagram of establishing a coupled mode model for a surface acoustic wave filter problem according to an embodiment of the present invention;

[0059] Figure 5 A comparison diagram of the COM algorithm and de-embedded data in an embodiment of the present invention;

[0060] Figure 6 is a schematic diagram of a topological model of a ladder filter according to an embodiment of the present invention;

[0061] Figure 7 It is a simulation schematic diagram of a surface acoustic wave filter implemented based on a COM algorithm according to an embodiment of the present invention;

[0062] Figure 8 It is a comparison diagram of de-embedding data, COM simulation data and SEM simulation data of an embodiment of the present invention;

[0063] Fig. 9 It is a schematic diagram of a simplified SEM geometric model of an actual model of an embodiment of the present invention;

[0064] Fig.10 A mesh division diagram of a single pair of electrode model of an embodiment of the present invention;

[0065] Fig.11 It is a comparison chart of the de-embedded data and SEM data of the Q_bode value of an embodiment of the present invention. DETAILED DESCRIPTION

[0066] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the relevant invention, rather than to limit the invention. It is also necessary to explain that, for ease of description, only the parts related to the relevant invention are shown in the accompanying drawings.

[0067] Figure 1 It is shown that the embodiment of the present invention discloses a method for simulation design of a surface acoustic wave filter, such as Figure 1 As shown, the method comprises the following steps:

[0068] S1, measuring an actual surface acoustic wave filter model, obtaining Y parameters of the filter, including open circuit and short circuit Y parameters, and de-embedding the admittance characteristic curve;

[0069] S2. Extracting COM model parameters and static capacitance according to the admittance characteristic curve;

[0070] S3, simulating coupled mode model data according to the extracted COM model parameters, and comparing the data with the measured data;

[0071] S4, geometric modeling of the SAW filter, including setting relevant parameters of the filter model and a calculation region to obtain a multi-physics field model and its mesh;

[0072] S5, adding corresponding boundary conditions and excitation conditions to the geometric model, and generating a mesh file;

[0073] S6, importing the grid file into the spectral element algorithm, and adopting distributed computing to save computing time;

[0074] S7, comparing the generated simulation curve of the COM model with the simulation curve obtained by the spectral element algorithm and the measured data.

[0075] Specifically, the secondary simulation design is mainly aimed at the actual model, and the measured data is de-embedded to obtain the admittance characteristic curve of the forked electrode of the surface acoustic wave filter. The measured admittance characteristic curve includes two parts, one is the admittance characteristic curve of the forked electrode itself, and the other is the parasitic parameters generated by the transmission line at high frequency, which are not reflected in the finite element method simulation and coupled mode algorithm simulation.

[0076] To solve the de-embedding problem, it is necessary to first measure the overall admittance curve, the interdigital short-circuit admittance curve and the interdigital open-circuit admittance curve. The interdigital electrode admittance expression is (1), as follows:

[0077] Y transistor =((Y dut -Y open ) -1 -(Yshort -Y open ) -1 ) -1 ⑴

[0078] where Y dut is the overall admittance curve of the device, Y open is the interdigital open circuit admittance curve, Y short The interdigital short-circuit admittance curve is shown in Figure 2. The comparison results before and after de-embedding are as follows: Figure 2 As shown in Figure 3, the parasitic parameters of the measured data have a greater impact on the admittance data.

[0079] In this embodiment, the actual model geometric parameters are shown in Table 1, and its microscopic optical images are shown in Table 1. Figure 3 shown.

[0080] parameter value Electrical Cycle 2.0[μm] Electrode Materials Al&Cu / Ti Piezoelectric Materials <![CDATA[YX42°LiTaO3]]> Electrode thickness 0.9*2.0*2.0[μm] Piezoelectric thickness 2*2.0[μm] Number of electrode pairs 60 Reflection grating pairs 40 Numerical aperture 40*2*2[μm]

[0081] Table 1 Geometric parameters of the filter model for measured data

[0082] For the measured data after de-embedding, we get f r 、f ar , f0 and Z imag .f r and f ar are the resonant frequency and anti-resonant frequency of the admittance curve, respectively. f0 is the low frequency point far away from the resonant frequency (linear change area), which is used to extract the static capacitance, Z imag is the imaginary part of the impedance corresponding to this point, and then the COM model parameters are calculated using the following formula:

[0083]

[0084]

[0085] Where η is a characterization of the average propagation characteristics of the surface acoustic wave device in the resonant and antiresonant states, |ε| represents the coupling strength difference parameter in the surface acoustic wave device, v(δ0) represents the propagation ratio or proportion of energy in the device under specific conditions, and δ B and δ V An intermediate variable representing some special frequency characteristics or energy states inside a surface acoustic wave device. к represents the energy coupling between different parts inside the device, including amplitude and phase information. B It is used to calculate and analyze the energy coupling strength of the device under a specific working state and its relationship with other parameters, η B It represents an enhancement factor of equivalent propagation or coupling capability under specific conditions. δ is a comprehensive parameter containing multiple variables related to the working characteristics of the surface acoustic wave device. ω is the angular frequency, p is the electrode period, and v Bis the wave velocity, and γ is the Q value loss coefficient.

[0086] A coupled mode model is established for the surface acoustic wave filter problem, such as Figure 4 shown.

[0087] A differential equation can be established for the above sound wave propagation phenomenon:

[0088]

[0089] Express the differential equation as a matrix operation:

[0090]

[0091] At this time, the P matrix unit is represented as an interdigital resonator. Modifying the parameters can be fed back to the admittance curve without modeling and simulating again, saving a lot of computing resources. The expressions of each element of the P matrix are as follows:

[0092]

[0093] P 33 =-K1P 31 -K2P 32 exp(jk c L)+2(α2K1-α1K2)L+jωC t . (2.18)

[0094] Where, D = s cos (sL) + jδsin (sL), c 12 =c 21 * = к is the reflection coefficient, α1 = α2 * is the coupling coefficient, k c is the wave number, P 11 and P 22 represents the reflection coefficient of the interdigital electrode, P 12 =P 21 represents the transmission coefficient of the interdigital electrode, P 13 and P 23 represents electromechanical coupling, which is the transfer function between the electrical quantity and the acoustic quantity at the left and right ends, P 13 and P 23 Represents the transfer function from acoustic quantity to electrical quantity, P 33 is the electrical quantity, namely, input admittance; C t is the static capacitance of the IDT device, which is affected by the number of interdigital pairs, the numerical aperture, the f0 and Z mentioned above. imag C can be obtained by the following formula t :

[0095] Ct =n*W*∈ (2.19)

[0096] ∈=p1*∈0 (2.20)

[0097]

[0098] Where n is the number of interdigital logarithms, W is the numerical aperture, and ∈0 is the vacuum dielectric constant. From the above, we can get that, similarly, since the IDT of the reflection grating only has acoustic characteristics, its port expression only needs to be changed to c 12 =c 21 * = 0, and the comparison between the COM algorithm and the de-embedded data is obtained. Figure 5 As shown in the figure, the curves have a high degree of fit, which proves the accuracy of the method. Based on this method, the simulation design process of the surface acoustic wave filter can be greatly accelerated.

[0099] Based on this, we get the P matrix of the reflector, gap, and IDT unit. According to the transfer function of the ABCD matrix, we can assemble the series and parallel connection of the filter. For a ladder filter, we generally use the following Figure 6 The topological model.

[0100] According to microwave network transmission, the matrix is ​​as follows:

[0101]

[0102] According to the ABCD matrix, we express the series and parallel resonators as follows:

[0103]

[0104] The conversion of the ABCD matrix into S parameters is based on the following equation:

[0105]

[0106] Among them, S 11 It represents the ratio of the input intensity to the reflected intensity of port 1 when the microwave signal is incident on port 1, and represents the attenuation of the signal. S 21 It indicates the microwave signal incident on port 1, the ratio of the input intensity of port 2 to the input intensity of port 1, and represents the signal throughput; S 22 Represents the microwave signal incident on the 2-port, the ratio of the 2-port input intensity to the reflected intensity, characterizing the signal attenuation, S 12 Indicates the microwave signal incident on 2 ports. The ratio of the input intensity of port 1 to the input intensity of port 2 characterizes the signal throughput.

[0107] We combine the coupled mode method with the transmission matrix to simulate the surface acoustic wave filter. Figure 7 and Figure 8 shown.

[0108] Furthermore, the embodiment of the present invention also proposes a method for solving the piezoelectric multi-physics field coupling equation based on the spectral element method, including:

[0109] First, the SAW filter is geometrically modeled, including setting the relevant parameters of the filter model and the calculation area to obtain the multi-physics field model and its mesh, such as Fig. 9 and Fig.10 As shown, Fig. 9 It is a schematic diagram of a simplified SEM geometric model of an actual model of an embodiment of the present invention; Fig.10 It is a mesh division diagram of a single pair of electrode model according to an embodiment of the present invention.

[0110] The multi-physics field model is calculated using the spectral element method to obtain two calculation areas, namely the electric field and the displacement field, and the control matrix equation to be solved is obtained:

[0111]

[0112]

[0113] Where ρ is the density of the piezoelectric material, ω is the operating angular frequency of the filter, u is the displacement vector of the deformation, τ is the stress tensor of the piezoelectric material, D is the electric displacement vector of the piezoelectric material, ρ e is the external charge density, Ω p is the electric field area, then Ω s+p is the displacement field area.

[0114] In addition to solving the matrix relationship of the two physical fields separately, the coupling relationship between the two must also be considered. Under the constitutive relationship of τ and D, the following coupling will be achieved between the electroacoustic signals:

[0115] D=e:ε+∈·E,in Ω s (4.3)

[0116] τ=c:ee·E,in Ω s+p (4.4)

[0117] Among them, c is the fourth-order elastic coefficient matrix, e is the third-order piezoelectric coupling coefficient, ∈ is the second-order dielectric constant, ε is the strain tensor, E is the electric field intensity, Ω p is the electric field area, then Ω s+p is the displacement field area.

[0118] In the above two equations, the first equation is the piezoelectric effect that applies a stress field to an object to generate an electrical signal; the second equation is the inverse piezoelectric effect that applies an electric field to a piezoelectric material to generate an acoustic wave signal. Among them, the relationship between the electric field and strain tensor and the electric potential and displacement field is as follows:

[0119]

[0120] Among them, V is the voltage field of the acoustic filter, thus realizing the mutual conversion relationship between the electroacoustic signal, Ω p is the electric field area, then Ω s+p is the displacement field area.

[0121] The spectral element algorithm solves the surface acoustic wave filter equation based on the weighted residual method to calculate the surface acoustic wave filter equation, so the original equation should be rewritten into the following weak form:

[0122]

[0123] Among them, Ω p is the electric field area, then Ω s+p is the displacement field region, Γ is the boundary surface, and n is the normal direction of the boundary surface. The physical region Ω of the acoustic wave filter is divided into K non-overlapping hexahedral grid units. The grid units in the physical space are then mapped to the reference unit through coordinate transformation, and numerical integration, derivation and interpolation can be performed directly in the reference unit.

[0124] The spectral element method of the present invention uses Lagrange function as basis function and selects Gauss-Lobatto Legendre (GLL) points as interpolation points. In the reference unit, the N-order basis function has N+1 GLL points and is defined as:

[0125]

[0126] Where i∈[-1,1],i=1,2,…,N+1, is the i-th basis function, L N (ζ i ) is the Nth order Lagrange polynomial, L' N (ζ i ) is the first-order derivative of the N-th order Lagrange polynomial.

[0127] The above constitutes the matrix equation for solving the physical model by the spectral element algorithm. Based on the needs of practical applications, it is necessary to add initial conditions to the matrix equation, which are often called boundary conditions. For the electric coupling model such as the acoustic wave filter, its boundary conditions can be divided into solid mechanics boundary conditions and electrostatic field boundary conditions. At this time, it is necessary to add corresponding boundary conditions according to the actual model.

[0128] After obtaining the assembly matrix A and the excitation source matrix b, the linear equations (Ax=b) are solved by direct or iterative methods to obtain the displacement field and voltage field of the entire model. The weak form is tested by the Galerkin method, and the reference domain is integrated and discretized:

[0129]

[0130] Among them, M ij represents the mass matrix, ü j represents the acceleration of the node, is the external force, is the internal force due to the piezoelectric effect, K ij represents the stiffness matrix, and is the corrected stiffness matrix u j represents the displacement of the node, V j represents the node voltage, Q i is the load term related to the electric field.

[0131] Distributed computing is performed, and the linear equations obtained above are divided into n tasks according to the frequency points. Distributed storage technology is used to divide the total equation solution into n task nodes and deploy them on the server to achieve the integration of computing resources. The computing memory is used in exchange for the computing time, which greatly shortens the simulation time of the wide-band SAW resonator. At the same order, the computing time will also be greatly reduced. The actual resonator model parameters are shown in Table 2.1, and the simulation data analysis is shown in Table 2.2:

[0132]

[0133] Table 2.1 Geometric parameters of spectral element method simulation model

[0134]

[0135]

[0136] Table 2.2 Comparison of spectral element method acceleration

[0137] From the analysis of the chart data, we can see that the spectral element algorithm has a high degree of simulation accuracy while greatly shortening the simulation time, and the addition of multi-core distributed computing further accelerates the simulation time.

[0138] The Q value is an important feature that characterizes the superior performance of a resonator. Finally, the Q value is calculated by the Q_bode method to verify the correctness of the simulation results. The results are as follows Fig.11 As shown in the figure, the spectral element method accelerates the simulation speed while the obtained curve is quite consistent with the measured curve. The Q value formula is as follows:

[0139]

[0140] Among them, S 11 is the S parameter converted from the admittance curve, group_delay(S 11 ) is the group delay expression, that is, the gradient operation of the imaginary part of S11.

[0141] In summary, the present invention proposes a method for simulating an acoustic wave filter based on a coupled mode method and a spectral element method. Based on the coupled mode method, the method can simulate the actual model and the filter composition without material parameters. In addition, the spectral element method adopts the concepts of grid degradation, matrix equivalence and distributed storage, which greatly speeds up the current problem of long acoustic wave filter simulation time without affecting the simulation accuracy, and makes a certain contribution to the thermoelectric simulation analysis of acoustic wave filters.

[0142] The above description is only a preferred embodiment of the present invention and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the present invention is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above inventive concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present invention (but not limited to) to form a technical solution.

Claims

1. A method for simulation design of a surface acoustic wave filter, characterized in that: The method comprises the following steps: Measuring an actual surface acoustic wave filter model, obtaining Y parameters of the filter, including open-circuit and short-circuit Y parameters, and de-embedding the admittance characteristic curve; Extracting COM model parameters and static capacitance according to the admittance characteristic curve; Simulating coupled mode model data according to the extracted COM model parameters and comparing the data with the measured data; Geometric modeling of the SAW filter, including setting the relevant parameters of the filter model and the calculation area to obtain the multi-physics field model and its mesh; Adding corresponding boundary conditions and excitation conditions to the geometric model and generating a mesh file; Importing the grid file into the spectral element algorithm and adopting distributed computing to save computing time; The generated simulation curve of the COM model is compared with the simulation curve obtained by the spectral element algorithm and the measured data.

2. The method for simulation design of a surface acoustic wave filter according to claim 1, characterized in that: The obtaining of the Y parameters of the filter, including the open circuit and short circuit Y parameters, comprises: AND transistor =((And dut -AND open ) -1 -(AND short -AND open ) -1 ) -1 where Y dut is the overall admittance curve of the device, Y open is the interdigital open circuit admittance curve, Y short is the interdigital short-circuit admittance curve.

3. The method for simulation design of a surface acoustic wave filter according to claim 1, characterized in that: The extracting of COM model parameters according to the admittance characteristic curve includes: obtaining f r 、f ar , f0 and Z imag , f r and f ar They are the resonant frequency and anti-resonant frequency of the admittance curve, f0 is a low frequency point far from the resonant frequency, the linear change area is used to extract the static capacitance, Z imag is the imaginary part of the impedance corresponding to this point, and then the COM model parameters are calculated using the following formula: Where η is a characterization of the average propagation characteristics of the surface acoustic wave device in the resonant and antiresonant states, |ε| represents the coupling strength difference parameter in the surface acoustic wave device, v(δ) represents the propagation ratio or proportion of energy in the device under specific conditions, and δ B and δ V An intermediate variable representing some special frequency characteristics or energy states inside a surface acoustic wave device. к represents the energy coupling between different parts inside the device, including amplitude and phase information. B It is used to calculate and analyze the energy coupling strength of the device under a specific working state and its relationship with other parameters, η B It represents an enhancement factor of equivalent propagation or coupling capability under specific conditions. δ is a comprehensive parameter containing multiple variables related to the working characteristics of the surface acoustic wave device. ω is the angular frequency, p is the electrode period, and v B is the wave velocity, and γ is the Q value loss coefficient.

4. The method for simulation design of a surface acoustic wave filter according to claim 1, characterized in that: The extraction of the static capacitance includes: C t is the static capacitance of the IDT device, which is affected by the number of interdigital logarithms and numerical aperture. The low frequency point f0 and its corresponding imaginary impedance Z imag C can be obtained by the following formula t : C t =n*W*∈ ∈＝p1*∈0 Where n is the number of interdigital pairs, W is the numerical aperture, and ∈0 is the vacuum dielectric constant.

5. The method for simulation design of a surface acoustic wave filter according to claim 1, characterized in that: The coupled mode model data simulation is performed according to the extracted COM model parameters, and compared with the measured data, including: establishing a coupled mode model for the surface acoustic wave filter, and establishing a differential equation for its acoustic wave propagation phenomenon: Express the differential equation as a matrix operation: The P matrix unit is represented as an interdigital resonator, and the expressions of each element of the P matrix are as follows: P 33 =-K1P 31 -K2P 32 exp(jk c L)+2(α2K1-α1K2)L+jωC t Where, D = s cos (sL) + jδsin (sL), c 12 =c 21 * =к is the reflection coefficient, α1=α2 * is the coupling coefficient, k c is the wave number, L is the resonator length, generally L = 2n*λ, n is the number of IDTs; λ = 2p, p is the electrode period; C t is the static capacitance of the IDT device, C t =C0*L*W;P 11 and P 22 represents the reflection coefficient of the interdigital electrode, P 12 =P 21 represents the transmission coefficient of the interdigital electrode, P 13 and P 23 represents electromechanical coupling, which is the transfer function between the electrical quantity and the acoustic quantity at the left and right ends, P 13 and P 23 Represents the transfer function from acoustic quantity to electrical quantity, P 33 It is an electrical quantity, namely, input admittance.

6. The method for simulation design of a surface acoustic wave filter according to claim 5, characterized in that: Also includes: The P matrix of the reflector, gap, and IDT unit is assembled into the series-parallel connection of the filter according to the transmission function of the ABCD matrix. The coupled mode method and the transmission matrix are combined to realize the simulation of the surface acoustic wave filter. For a ladder filter, according to the microwave network transmission, the matrix is ​​as follows: According to the ABCD matrix, the series and parallel resonators are expressed as follows: The ABCD matrix is ​​converted to S parameters according to the following equation: Among them, S 11 It represents the ratio of the input intensity to the reflected intensity of port 1 when the microwave signal is incident on port 1, and represents the attenuation of the signal. S 21 It indicates the microwave signal incident on port 1, the ratio of the input intensity of port 2 to the input intensity of port 1, and represents the signal throughput; S 22 It represents the ratio of the 2-port input intensity to the reflected intensity when the 2-port microwave signal is incident, and it characterizes the signal attenuation; S 12 Indicates the microwave signal incident on 2 ports. The ratio of the input intensity of port 1 to the input intensity of port 2 characterizes the signal throughput.

7. The method for simulation design of a surface acoustic wave filter according to claim 1, characterized in that: The geometric modeling of the SAW filter includes setting relevant parameters of the filter model and a calculation area to obtain a multi-physics field model and its mesh, including: The multi-physics field model is calculated using the spectral element method to obtain two calculation areas, namely the electric field and the displacement field, and the control matrix equation to be solved is obtained: Where ρ is the density of the piezoelectric material, ω is the operating angular frequency of the filter, u is the displacement vector of the deformation, τ is the stress tensor of the piezoelectric material, D is the electric displacement vector of the piezoelectric material, ρ e is the external charge density, Ω p is the electric field area, then Ω s+p is the displacement field area; Under the constitutive relationship between τ and D, the following coupling will be achieved between the electroacoustic signals: D=e:ε+∈·E,inΩ s τ=c:ee·E,inΩ s+p Among them, c is the fourth-order elastic coefficient matrix, e is the third-order piezoelectric coupling coefficient, ∈ is the second-order dielectric constant, ε is the strain tensor, E is the electric field intensity, Ω p is the electric field area, Ω s+p is the displacement field area; where the electric field and strain tensors are related to the potential and displacement fields as follows: Where V is the voltage field of the acoustic filter, thus realizing the mutual conversion relationship between the electroacoustic signal, Ω p is the electric field area, Ω s+p is the displacement field area.

8. The method for simulation design of a surface acoustic wave filter according to claim 1, characterized in that: The spectral element algorithm solves the surface acoustic wave filter equation as follows: Among them, Ω is the entire calculation area, Ω p is the electric field area, Ω s+p is the displacement field region, Γ is the boundary surface, n is the boundary surface normal, and n·D represents the output voltage; the physical region Ω of the acoustic wave filter is divided into K non-overlapping hexahedral grid units, and then the grid units in the physical space are mapped to the reference unit through coordinate transformation, and numerical integration, derivation and interpolation are directly performed in the reference unit.

9. The method for simulation design applied to simulated surface acoustic wave filters according to claim 1, characterized in that: The spectral element algorithm uses Lagrange function as basis function and selects Gauss-LobattoLegendre (GLL) points as interpolation points to form the matrix equation for solving the physical model. In the reference unit, the N-order basis function has N+1 GLL points, which are defined as: Among them, ζ∈[-1,1],i=1,2,…,N+1, is the i-th basis function, L N (ζ i ) is the Nth order Lagrange polynomial, L' N (ζ i ) is the first-order derivative of the N-th order Lagrange polynomial; The assembled matrix A and the excitation source matrix b are obtained, and then the linear equations (Ax = b) are solved by direct method or iterative method to obtain the displacement field and voltage field of the entire model. The weak form is tested by Galerkin method, and the reference domain is integrated and discretized: Among them, M ij represents the mass matrix, represents the acceleration of the node, is the external force, is the internal force due to the piezoelectric effect, K ij represents the stiffness matrix, and is the corrected stiffness matrix u j represents the displacement of the node, V j represents the node voltage, Q i is the load term related to the electric field.

10. The method for simulation design of a surface acoustic wave filter according to claim 1, characterized in that: The simulation curve of the generated COM model is compared with the simulation curve obtained by the spectral element algorithm and the measured data, including: the Q value is calculated by the Q_bode method to verify the correctness of the simulation result. The Q value formula is as follows: Among them, S 11 is the S parameter converted from the admittance curve, group_delay(S 11 ) is the group delay expression, that is, the gradient operation of the imaginary part of S11.

Images