Rapid electromagnetic optimization method for integrated waveguide filter

Through a rapid optimization method based on normalized coupling matrix and quasi-Newtonian algorithm, the problem of time-consuming optimization of traditional electromagnetic simulation is solved, and the rapid optimization of substrate integrated waveguide filter is realized, which reduces the computational complexity and time and improves stability and convergence speed.

CN118966124AInactive Publication Date: 2024-11-15NANJING RUANHE ELECTRONIC TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410310651.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-03-18
Publication Date
2024-11-15
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Traditional full-wave electromagnetic simulation optimization requires high computing resources and long-term time, making it difficult to achieve fast, simple, reliable and stable electromagnetic design optimization of substrate integrated waveguide filters.

Method used

Using a fast optimization method based on normalized coupling matrix and quasi-Newtonian algorithm, the feature parameter model is approximated to the fine electromagnetic model response through a quasi-Newtonian algorithm with few iterations, and using Jacobian matrix to estimate and correct physical dimension errors to achieve rapid filter optimization.

Benefits of technology

The number of optimized electromagnetic solutions is reduced, the complexity, good stability and fast convergence speed is improved, the local minimum value is avoided, the feature parameter extraction process is simplified, and it is suitable for lossy and lossless networks.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118966124A_ABST
    Figure CN118966124A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of microwave and millimeter wave circuits, and discloses a rapid optimization method for a substrate integrated waveguide filter, and the method comprises the steps: obtaining an optimal characteristic parameter, a target response and an initialized physical size of the substrate integrated waveguide filter based on a normalized coupling matrix of the substrate integrated waveguide filter; in combination with a quasi-Newton algorithm with few iterations, enabling the characteristic parameter model of the substrate integrated waveguide filter to approach the response of the fine electromagnetic model of the substrate integrated waveguide filter, and obtaining the characteristic parameter deviation of the fine electromagnetic model of the current substrate integrated waveguide filter; the deviation is substituted into a Jacobian matrix based on the substrate integrated waveguide filter, the precise physical size of the substrate integrated waveguide filter is obtained through repeated iterative calculation, rapid optimization of the integrated waveguide filter is achieved, complexity is low, stability is good, the convergence speed is high, and the situation that the integrated waveguide filter is caught in a local minimum value can be avoided.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of microwave and millimeter wave circuits, and in particular to a fast electromagnetic optimization method for an integrated waveguide filter. Background Art

[0002] With the continuous advancement of communication technology, RF front-end systems are developing towards miniaturization, light weight, high reliability and low cost. Microwave passive filters are an important part of wireless communication RF front-end systems. Substrate integrated waveguide filters, as a perfect combination of waveguide structure and microstrip line structure, have the advantages of both, namely low loss and easy integration, showing great application value, and are therefore widely used in filter design.

[0003] Traditional full-wave electromagnetic simulation obtains the response of substrate integrated waveguide filters through direct optimization methods, which has wide applicability. However, due to the large number of metal vias in the SIW structure, high-density grids are required in the full-wave simulation solver, so traditional full-wave electromagnetic simulation optimization requires high computing resources and long computing time. The design optimization of a filter may take several weeks and hundreds of electromagnetic solutions.

[0004] Therefore, realizing a fast, simple, reliable and stable electromagnetic design optimization method for substrate integrated waveguide filters is a technical difficulty that needs to be solved and improved. Summary of the invention

[0005] To achieve the above purpose, the present invention adopts the following technical solutions:

[0006] A method for rapid optimization of a substrate integrated waveguide filter comprises the following steps:

[0007] Step 1 S1: based on the normalized coupling matrix of the substrate integrated waveguide filter, obtaining the optimal characteristic parameters, target response and initialization physical size of the substrate integrated waveguide filter;

[0008] Step 2 S2: using a quasi-Newton algorithm with a small number of iterations to make the characteristic parameter model of the substrate integrated waveguide filter approach the response of the current fine electromagnetic model, and obtain the characteristic parameter deviation of the current substrate integrated waveguide filter fine electromagnetic model;

[0009] Step three S3: using a characteristic parameter vector to characterize the physical size of the substrate integrated waveguide filter, substituting the characteristic parameter deviation into the Jacobian matrix based on the substrate integrated waveguide filter, estimating and correcting the physical size error of the substrate integrated waveguide filter, and calculating the current response of the filter.

[0010] Step 4 S4: using the current response of the substrate integrated waveguide filter calculated in S3 as a reference response of a new fine electromagnetic model;

[0011] Step five S5: repeat S2-S4. When the response of the substrate integrated waveguide filter obtained in S3 reaches a preset condition close to the target response, a set of precise physical dimensions of the substrate integrated waveguide filter is obtained, thereby realizing rapid optimization of the substrate integrated waveguide filter.

[0012] The invention discloses a method for rapid optimization of a substrate integrated waveguide filter. The method obtains the optimal characteristic parameters, target response and initialization physical size of the substrate integrated waveguide filter by using a normalized coupling matrix based on the substrate integrated waveguide filter. The method combines a quasi-Newton algorithm with a small number of iterations to make the characteristic parameter model of the substrate integrated waveguide filter approach the response of its fine electromagnetic model, thereby obtaining the characteristic parameter deviation of the fine electromagnetic model of the current substrate integrated waveguide filter. The method substitutes the above deviation into the Jacobian matrix based on the substrate integrated waveguide filter, repeatedly iterates and calculates to obtain the precise physical size of the substrate integrated waveguide filter, thereby realizing rapid optimization of the integrated waveguide filter. The method has low complexity, good stability, fast convergence speed and can avoid falling into a local minimum.

[0013] The present invention has the following beneficial effects:

[0014] (1) Low complexity: The optimization method of the present invention adopts a direct response approximation based on the characteristic parameter model of the substrate integrated waveguide filter and adopts a quasi-Newton algorithm to simplify the characteristic parameter extraction process. This method is applicable to both lossy networks and lossless networks.

[0015] (2) Good stability: The Jacobian matrix update scheme based on the substrate integrated waveguide filter in the present invention can avoid the propagation of parameter errors from the current iteration to the next iteration, thereby obtaining better robustness and enabling the physical size correction of the filter to be controlled within a reasonable range.

[0016] (3) Fast convergence speed: The number of optimization electromagnetic solutions is reduced by one order of magnitude. The optimization method of the substrate integrated waveguide filter proposed in the present invention does not directly search for solutions in the fine electromagnetic model space, but uses an iterative process to extract the characteristic parameters of the current fine electromagnetic model and calculate the dimensional error. It ensures that the error estimate is accurate enough to enable the design variables to converge quickly to the optimal solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flow chart of a method for rapid optimization of a substrate integrated waveguide filter according to an embodiment of the present disclosure is shown;

[0018] Figure 2 An accurate electromagnetic model of a 6th-order Chebyshev-type substrate integrated waveguide filter according to an embodiment of the present disclosure is shown;

[0019] Figure 3 A schematic diagram showing an equivalent circuit of a 6th-order Chebyshev-type substrate integrated waveguide filter according to an embodiment of the present disclosure is shown;

[0020] Figure 4 The target response of a 6th-order Chebyshev-type substrate integrated waveguide filter according to an embodiment of the present disclosure is shown;

[0021] Figure 5 The detailed electromagnetic model response and the approximate response of the characteristic parameter model of the 6th-order Chebyshev-type substrate integrated waveguide filter according to an embodiment of the present disclosure are shown;

[0022] Figure 6 The graph shows the evolution of characteristic parameter deviation of a 6th-order Chebyshev-type substrate integrated waveguide filter according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0023] The present invention is described in detail below in conjunction with embodiments:

[0024] The above embodiments are preferred embodiments of the present invention, and the protection scope of the present invention should not be limited thereto. Any modifications, equivalent substitutions and improvements made on the ideas and solutions of the present invention should be included in the protection scope of the present invention.

[0025] The following will describe in detail an exemplary embodiment, features and aspects of the present disclosure with reference to the accompanying drawings. The same reference numerals in the accompanying drawings represent elements with the same or similar functions. Although various aspects of the embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless otherwise specified.

[0026] The word “exemplary” is used exclusively herein to mean “serving as an example, example, or illustration.” Any embodiment described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments.

[0027] In addition, in order to better illustrate the present disclosure, numerous specific details are given in the following specific embodiments. It should be understood by those skilled in the art that the present disclosure can also be implemented without certain specific details. In some examples, methods, means, components and circuits well known to those skilled in the art are not described in detail in order to highlight the subject matter of the present disclosure.

[0028] Figure 1 A flow chart of a method for rapid optimization of a substrate integrated waveguide filter according to an embodiment of the present disclosure is shown.

[0029] It should be noted that the substrate integrated waveguide filter can be a Chebyshev response substrate integrated waveguide filter, or a Butterworth substrate integrated waveguide filter, etc., which is not limited here. In order to facilitate the understanding of this scheme, the following is an example of a 6th-order Chebyshev response substrate integrated waveguide filter. The filter related indicators are as follows:

[0030] Center frequency: 12GHz

[0031] Bandwidth: 1.3GHz

[0032] In-band return loss: 20dB

[0033] Source / load impedance: Z0 = 50Ω

[0034] like Figure 1 As shown, the method may include:

[0035] Step 1 S1: based on the normalized coupling matrix of the substrate integrated waveguide filter, the optimal characteristic parameters, target response and initialization physical size of the substrate integrated waveguide filter are obtained.

[0036] Optionally, the dielectric substrate can be a ceramic, ferrite and ferromagnetic material board or a PCB board of the Rogers, Taconic, etc. series. At the same time, the dielectric substrate can be a single layer or a multi-layer, fixed together by screws or adhesive layers, and the number of layers, thickness and material of the dielectric substrate are selected according to design indicators such as the working frequency band.

[0037] In one embodiment of the present disclosure, a 6th-order SIW Chebyshev filter uses a Rogers RO4003C dielectric substrate with a thickness of 0.406 mm and a relative dielectric constant of 3.55.

[0038] Figure 2-Figure 4 An accurate electromagnetic model, an equivalent circuit and a target response schematic diagram of a 6th-order SIW Chebyshev filter according to an embodiment of the present disclosure are respectively shown.

[0039] like Figure 2 As shown in the figure, the 6th-order SIW Chebyshev filter has a symmetrical structure, which reduces the number of physical size design variables. The characteristic parameter vector of the substrate integrated waveguide filter can be expressed as:

[0040] x=[f 0i , k ij , Q u , Q ext ] Formula (1),

[0041] f 0i =[f 0,1 , f 0,2 , f 0,3 ]=[f 0,6 , f 0,5 , f 0,4 ] Formula (2),

[0042] k ij =[k 12 , k 23, k 34 ]=[k 56 , k 45 , k 34 ] Formula (3),

[0043] Q ext =Q ext , S = Q ext,L Formula (4),

[0044] Among them, f 0,i represents the resonant frequency of the i-th resonant cavity, k ij The elements of represent all coupling coefficients between adjacent resonant cavities, and the unloading quality factor Q of the SIW resonant cavity u About 350.

[0045] In addition, by Figure 2 It can also be seen that in the cavity width W g Under the premise that the above embodiment filter is unchanged, L i To control the resonant frequency, M i,i+1 and s control the coupling coefficient and external quality factor respectively. Therefore, the physical dimension design variable vector of substrate integrated waveguide filter is:

[0046] y=[L1,L2,L3,M 12 , M 23 , M 34 , s] Formula (5),

[0047] In one embodiment of the present disclosure, other physical dimension parameter values ​​of the substrate integrated waveguide filter are: g =9.69mm, d=0.4mm, p≈1.2mm, q=1.374mm.

[0048] In one embodiment of the present disclosure, according to the basic design rules of substrate integrated waveguide filter, such as equations (6)-(8), the initial values ​​of the physical size design variables of the filter can be obtained.

[0049]

[0050]

[0051]

[0052] In one embodiment of the present disclosure, the characteristic parameters may be associated with a classical normalized coupling matrix. The normalized N+2 coupling matrix M0 of an N-order bandpass substrate integrated waveguide filter may be:

[0053]

[0054] The transformation between the generalized coupling matrix and the characteristic parameters can be expressed as:

[0055]

[0056] k ij =F BW ·m ij

[0057]

[0058] According to equations (9)-(10), the optimal characteristic parameter of the substrate integrated waveguide filter can be obtained as f 0,i =[12,12,12]GHz,k ij =[0.0913, 0.0662, 0.0632], Q ext =9.2.

[0059] Step 2 S2: using a quasi-Newton algorithm with a small number of iterations to make the characteristic parameter model of the substrate integrated waveguide filter approach the response of the current fine electromagnetic model, and obtain the characteristic parameter deviation of the current substrate integrated waveguide filter fine electromagnetic model;

[0060] The extraction process of characteristic parameter deviation can be expressed as:

[0061]

[0062] Where y is the current size of the substrate integrated waveguide filter, R f (y) is the response of the fine electromagnetic model of the filter at the current size, and the characteristic parameter x * is the optimal solution of the filter characteristic parameter model, R C (x * ) is the filter target response.

[0063] In one embodiment of the present disclosure, each iteration is performed by converting the filter characteristic parameter model response R C (x * +Δx) is as close as possible to the current fine electromagnetic model response R f (y) is used to extract the deviation Δx of the characteristic parameters.

[0064] In one embodiment of the present disclosure, an equivalent objective function E(Δx) is constructed to solve the deviation extraction problem in step 2, that is, the objective function E(Δx) is minimized to estimate the feature parameter deviation Δx. The equivalent objective function E(Δx) can be expressed as:

[0065]

[0066] Among them, the partial differential term is introduced to better fit the reflection zero point.

[0067] In one embodiment of the present disclosure, the weight coefficient α is set to 0.25.

[0068] It should be noted that the port reference plane relationship between the fine electromagnetic model and the characteristic parameter model of the substrate integrated waveguide filter is uncertain, and there may be a phase error. This may lead to the failure of the approximation process using the real and imaginary s-parameter method. Therefore, all terms of the objective function E(Δx) are constructed using amplitude functions to eliminate the influence of the phase reference plane.

[0069] In one embodiment of the present disclosure, a quasi-Newton algorithm is used to calculate the minimum solution of the objective function E(Δx) in equation (12), and the maximum number of iterations of the quasi-Newton algorithm is set to 15 each time the deviation is estimated.

[0070] In one embodiment of the present disclosure, the resonant frequency Δf 0,i The error range of the error prediction of the remaining parameters of Δx in the quasi-Newton algorithm is approximately ±0.5x * .

[0071] Step three S3: using a characteristic parameter vector to characterize the physical size of the substrate integrated waveguide filter, substituting the characteristic parameter deviation into a Jacobian matrix based on the substrate integrated waveguide filter, estimating and correcting the physical size error of the substrate integrated waveguide filter, and calculating the current response of the SIW filter.

[0072] In one embodiment of the present disclosure, according to the Jacobian matrix of the substrate integrated waveguide filter model, the characteristic parameter deviation Δx of the SIW filter obtained by S2 is converted into an approximate Jacobian matrix of its physical size error Δy, which can be expressed in the form of a block diagonal matrix:

[0073]

[0074] Where Δy=yy * , y is the current physical size of the SIW filter, y * The optimal solution for physical size.

[0075] Optionally, in formula (11) Represents the negative coupling coefficient between adjacent cavities of the SIW filter.

[0076] Substituting equations (6)-(8) into equation (13), we obtain:

[0077]

[0078] in

[0079] Estimation of Δy based on design deviation of SIW filter physical dimensions n, the current design variable y n Make corrections, the correction formula is:

[0080] y n+1 =y n -μΔy n Formula (15),

[0081] Where μ is the correction step factor, and the introduction of the correction step factor makes the algorithm more robust.

[0082] In one embodiment of the present disclosure, the correction factor μ is set to 0.5.

[0083] Step 4 S4: using the current response of the substrate integrated waveguide filter calculated in S3 as a reference response of a new fine electromagnetic model;

[0084] Set the iteration counter n = n + 1, according to the physical size y of the modified SIW filter n+1 Calculate the new filter response R f (y n+1 ), which is used as the current reference response for the characteristic parameter model approximation in step 2.

[0085] Step five S5: repeat S2-S4. When the response of the substrate integrated waveguide filter obtained in S3 reaches a preset condition close to the target response, a set of precise physical dimensions of the substrate integrated waveguide filter is obtained, thereby realizing rapid optimization of the substrate integrated waveguide filter.

[0086] Optionally, in each iteration, if R obtained in step 4 f (y n+1 ) is approximately equal to the target response R of the substrate integrated waveguide filter C (x * ), the iteration is terminated and the optimization process is completed.

[0087] Figure 5 The refined electromagnetic model responses and the approximate responses of the characteristic parameter model for the first to fourth iterations are shown. Figure 6 The evolution of the characteristic parameter deviation is shown in the figure. It can be observed from the figure that the deviation gradually converges to zero. The evolution of the physical size design variables obtained from the first to the fourth iteration is shown in the following table.

[0088]

[0089] In one embodiment of the present disclosure, the optimal solution of the substrate integrated waveguide filter is obtained after 14 iterations, and a perfect equal ripple filter response is obtained, which is consistent with the target response.

[0090] The above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terms used herein are selected to best explain the principles of the embodiments, practical applications, or technical improvements in the marketplace, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A method for rapid optimization of a substrate integrated waveguide filter, comprising the following steps: Step 1 S1: based on the normalized coupling matrix of the substrate integrated waveguide filter, obtaining the optimal characteristic parameters, target response and initialization physical size of the substrate integrated waveguide filter; Step 2 S2: using a quasi-Newton algorithm with a small number of iterations to make the characteristic parameter model of the substrate integrated waveguide filter approach the response of the current fine electromagnetic model, and obtain the characteristic parameter deviation of the current substrate integrated waveguide filter fine electromagnetic model; Step 3 S3: using a characteristic parameter vector to characterize the physical size of the substrate integrated waveguide filter, substituting the characteristic parameter deviation into a Jacobian matrix based on the substrate integrated waveguide filter, estimating and correcting the physical size error of the substrate integrated waveguide filter, and calculating the current response of the filter; Step 4 S4: using the current response of the substrate integrated waveguide filter calculated in S3 as a reference response of a new fine electromagnetic model; Step five S5: repeat S2-S4. When the response of the substrate integrated waveguide filter obtained in S3 reaches a preset condition close to the target response, a set of precise physical dimensions of the substrate integrated waveguide filter is obtained, thereby realizing rapid optimization of the substrate integrated waveguide filter.