Parameter calibration method and device of double-layer double-frequency satellite navigation antenna and vehicle

Abstract

The embodiment of the application relates to the technical field of antennas, and discloses a parameter calibration method and device of a double-layer double-frequency satellite navigation antenna and a vehicle. The parameter calibration method of the double-layer double-frequency satellite navigation antenna comprises the following steps: determining a target parameter to be calibrated of the double-layer double-frequency satellite navigation antenna; parameter value disturbance of the target parameter will cause the first target frequency of the first layer antenna and / or the second target frequency of the second layer antenna to deviate; a simulation model of the double-layer double-frequency satellite navigation antenna is used to simulate a disturbance matrix of the target parameter; the disturbance matrix is used to represent the corresponding relationship between the parameter value disturbance amount of the target parameter and the target frequency deviation amount; based on the disturbance matrix and a target function, an optimal calibration parameter value of the target parameter capable of minimizing the performance degradation parameter of the first layer antenna at an ideal working frequency and the performance degradation parameter of the second layer antenna at an ideal working frequency is determined. The technical scheme of the application can effectively simplify the calibration process and improve the calibration speed.

Description

Technical Field

[0001] This invention relates to the field of antenna technology, and specifically to a parameter calibration method, apparatus, and vehicle for a dual-layer dual-frequency satellite navigation antenna. Background Technology

[0002] With the rapid development of intelligent connected vehicle technology, various sensing devices such as LiDAR, forward-facing cameras, surround-view cameras, reversing cameras, and panoramic cameras have been gradually added to vehicles. At the same time, various electromagnetic interference sources, such as roof speakers, are also installed. This dense arrangement of such devices creates a dilemma for GNSS navigation antennas, which have relatively weak anti-interference capabilities, as they face a lack of suitable placement within the vehicle layout. The antenna's placement and the surrounding electromagnetic environment directly affect its core parameters, such as receiving sensitivity and positioning accuracy, thus significantly impacting the accuracy and effectiveness of antenna parameter calibration.

[0003] The related technologies disclose methods for eliminating the effects of multipath errors by using absorbing materials in a microwave anechoic chamber, simulating GNSS transmitted signals using a vector network analyzer, controlling the incident direction of the received signal by the antenna under test using a high-precision rotating turntable, acquiring the phase pattern covering the antenna hemisphere through a measurement system, and then calculating the phase center deviation (PCO) and phase center variation (PCV) using a least squares algorithm to achieve high-precision calibration of antenna parameters. Another related technology discloses a method for calibrating the heading angle of dual GNSS antennas for intelligent driving vehicles. This method requires acquiring the pose data of the vehicle under test while traveling along a preset straight road segment. By calculating a first heading angle based on the position information and a second heading angle based on the heading angle output by the navigation equipment, the installation error of the dual GNSS antennas is determined, thereby completing the heading angle parameter calibration. This method also requires strict judgment of the validity of the vehicle's driving data and has certain requirements for the test site.

[0004] It can be seen that most existing technologies adopt calibration methods that combine real vehicle testing with dedicated equipment or specific test scenarios. Whether it is microwave anechoic chamber calibration or real vehicle road calibration, real vehicles are required, and the calibration process is cumbersome and complex: microwave anechoic chamber calibration relies on expensive equipment such as professional anechoic chamber facilities, vector network analyzers, and high-precision rotary tables, resulting in high equipment investment costs. The calibration process involves multiple steps such as coordinate system one, phase direction pattern acquisition, and multi-round data iteration calculation, which is difficult to operate and time-consuming. Real vehicle road calibration requires the pre-set of dedicated test road sections to collect a large amount of pose data during vehicle driving. The conditions for judging the validity of the data are stringent, and it is greatly affected by external environmental interference, resulting in low calibration efficiency. Summary of the Invention

[0005] In view of the shortcomings of the prior art, the purpose of this application is to provide a parameter calibration method, apparatus and vehicle for a dual-layer dual-frequency satellite navigation antenna, which aims to simplify the parameter calibration process of the dual-layer dual-frequency satellite navigation antenna.

[0006] In a first aspect, embodiments of this application provide a parameter calibration method for a dual-layer dual-frequency satellite navigation antenna, comprising: determining target parameters of the dual-layer dual-frequency satellite navigation antenna to be calibrated; wherein, perturbation of the parameter values ​​of the target parameters will cause a shift in a first target frequency and / or a second target frequency; the first target frequency is the actual operating frequency of the first layer antenna in the dual-layer dual-frequency satellite navigation antenna; the second target frequency is the actual operating frequency of the second layer antenna in the dual-layer dual-frequency satellite navigation antenna; and simulating the perturbation matrix of the target parameters using a simulation model of the dual-layer dual-frequency satellite navigation antenna; wherein the perturbation matrix is ​​used to represent the parameter values ​​of the target parameters. The correspondence between numerical perturbation and target frequency offset; the target frequency offset is the frequency offset of the first target frequency and / or the second target frequency; based on the perturbation matrix and the objective function, the optimal calibration parameter values ​​of the target parameters are determined; wherein, the objective function is used to minimize the performance degradation parameters at the ideal operating frequency of the first-layer antenna and the ideal operating frequency of the second-layer antenna; the performance degradation parameters at the ideal operating frequency of the first-layer antenna and the ideal operating frequency of the second-layer antenna change with the change of the first target frequency and / or the second target frequency.

[0007] Based on the aforementioned technical methods, by precisely selecting target parameters whose perturbations will cause a shift in the target frequency, and then performing quantitative analysis based on the correspondence between the perturbation amount of the target parameters and the target frequency shift (perturbation matrix), the impact of target parameter changes on the actual operating frequencies (first and second target frequencies) of the two-layer antenna can be accurately captured, effectively avoiding frequency shift problems caused by parameter perturbations. By combining the objective function with minimizing the performance degradation parameters at the ideal operating frequencies of the two-layer antenna, the independent performance of the two antennas can be simultaneously considered, ensuring that their performance degradation at the ideal operating frequencies is minimized, thereby improving the operational stability, frequency accuracy, and positioning reliability of the dual-layer dual-frequency satellite navigation antenna. Furthermore, by relying on simulation models to obtain the perturbation matrix, complex physical debugging is eliminated, significantly simplifying the target parameter calibration process, improving the efficiency and accuracy of determining the optimal calibration parameter values, and reducing calibration costs.

[0008] In an exemplary embodiment, a simulation model of a dual-layer dual-frequency satellite navigation antenna is used to simulate the perturbation matrix of the target parameters, including: simulating a first perturbation matrix and a second perturbation matrix using the simulation model; wherein, the first perturbation matrix is ​​used to represent the correspondence between the parameter value perturbation of the target parameters and the target frequency offset under the premise of isolating the inter-layer frequency interference between the first layer antenna and the second layer antenna; the second perturbation matrix is ​​used to represent the correspondence between the parameter value perturbation of the target parameters and the inter-layer frequency interference; and the perturbation matrix is ​​determined based on the first perturbation matrix and the second perturbation matrix.

[0009] Based on the above technical means, by simulating the first perturbation matrix and the second perturbation matrix respectively, the influence of parameter perturbation under the two scenarios of "isolation of inter-layer frequency interference" and "inter-layer interference itself" can be clearly distinguished. The different effects of target parameter changes on antenna frequency offset and inter-layer interference can be accurately captured, and the optimization deviation caused by the mutual influence of inter-layer interference and parameter perturbation can be avoided.

[0010] In the exemplary embodiment, the simulation model is constructed as follows: Based on the scanning information of the dual-layer dual-frequency satellite navigation antenna, an initial simulation model of the dual-layer dual-frequency satellite navigation antenna is constructed; the first-layer antenna and the second-layer antenna in the initial simulation model are decoupled to obtain an intermediate simulation model; decoupling is used to isolate inter-layer frequency interference; and the simulation model is obtained using the intermediate simulation model and the residual structure; wherein, the residual structure is used to represent the frequency offset difference of the target frequency offset when the parameter values ​​of the target parameters are the same in the initial simulation model and the intermediate simulation model.

[0011] Based on the aforementioned technical means, by using the scanning information of the dual-layer dual-frequency satellite navigation antenna, an initial simulation model of the dual-layer dual-frequency satellite navigation antenna can be accurately constructed. Furthermore, the first and second layer antennas in the initial simulation model are decoupled, effectively isolating the inter-layer frequency interference between the two layers of antennas and realizing independent analysis of the electromagnetic characteristics of the upper and lower layers of antennas. In addition, a residual structure is introduced to characterize the target frequency offset difference between the initial simulation model and the intermediate simulation model under the same parameter disturbance, which can effectively compensate for the model deviation caused by the decoupling process. This allows the final simulation model to retain the advantages of independent analysis of inter-layer decoupling while realistically restoring the actual inter-layer coupling effects, significantly improving the modeling accuracy and physical realism of the simulation model.

[0012] In an exemplary embodiment, the target parameters include a first dielectric constant corresponding to the first layer antenna and a second dielectric constant corresponding to the second layer antenna. Simulating the first and second perturbation matrices using a simulation model includes: for each pre-selected parameter value perturbation quantity among at least one pre-selected parameter value perturbation quantity of the target parameters, inputting the pre-selected parameter value perturbation quantity into an initial simulation model and an intermediate simulation model respectively to obtain the target frequency offset output by the intermediate simulation model response and the frequency offset difference output by the residual structure response; generating the first perturbation matrix using the parameter value perturbation quantity and the target frequency offset output by the intermediate simulation model response; and generating the second perturbation matrix using the parameter value perturbation quantity and the frequency offset difference output by the residual structure response.

[0013] Based on the aforementioned technical means, for each pre-selected parameter value disturbance in at least one pre-selected parameter value disturbance of the target parameter, each pre-selected parameter value disturbance is substituted into the initial simulation model and the intermediate simulation model for synchronous simulation and solution. This allows for the separate acquisition of the target frequency offset under decoupled conditions and the frequency offset difference corresponding to the residual structure. Furthermore, a first disturbance matrix is ​​constructed based on the parameter disturbance and the output results of the intermediate simulation model, and a second disturbance matrix is ​​constructed based on the parameter disturbance and the residual frequency offset difference. This can accurately separate the independent effects of parameter disturbances of the first dielectric constant and the second dielectric constant on the inherent frequency offset of the antenna and inter-layer coupling interference, achieving decoupled quantitative characterization of parameter disturbance effects, avoiding parameter coupling deviations caused by inter-layer frequency interference, and improving the accuracy and physical realism of disturbance matrix modeling.

[0014] In the exemplary embodiment, the target parameters include: the patch size corresponding to the first layer antenna, the patch size corresponding to the second layer antenna, and the interlayer spacing between the first layer antenna and the second layer antenna; the element values ​​in the second perturbation matrix are all 0 by default; the simulation model is used to simulate the first perturbation matrix, including: for each of the perturbation quantities of at least one preselected parameter value for the target parameters, inputting the preselected parameter value perturbation quantity into an intermediate simulation model to obtain the target frequency offset output by the intermediate simulation model in response to the preselected parameter value perturbation quantity; and generating the first perturbation matrix using the parameter value perturbation quantity and the target frequency offset output by the intermediate simulation model in response to the preselected parameter value perturbation quantity.

[0015] Based on the aforementioned technical means, the patch size corresponding to the first layer antenna, the patch size corresponding to the second layer antenna, and the interlayer spacing between the first and second layer antennas are used as the core target parameters to be calibrated. The elements of the second perturbation matrix are preset to zero by default, which can ignore the interlayer coupling residual interference caused by such structural parameters. By inputting the perturbation amount of each pre-selected parameter value into the intermediate simulation model one by one, the corresponding target frequency offset is obtained. Then, the first perturbation matrix is ​​generated by combining the mapping relationship between the parameter perturbation amount and the frequency offset. This can accurately quantify the influence of individual perturbations of structural parameters such as patch size and interlayer spacing on the target frequency offset of the antenna, eliminate the interference of irrelevant coupling residuals, simplify the complexity of matrix solution, and improve the efficiency and reliability of perturbation matrix construction.

[0016] In an exemplary embodiment, the objective function is further used to minimize the isolation penalty; wherein the isolation penalty is the product of the penalty coefficient and the total frequency offset of the two layers; the penalty coefficient is used to suppress the deterioration of the isolation effect of inter-layer frequency interference; the total frequency offset of the two layers is the sum of the first frequency difference and the second frequency difference; the first frequency difference is the frequency difference between the first target frequency and the ideal operating frequency of the first layer antenna; the second frequency difference is the frequency difference between the second target frequency and the ideal operating frequency of the second layer antenna.

[0017] Based on the aforementioned technical means, the objective function, while minimizing the performance degradation parameters of the two-layer antennas, further introduces an isolation penalty term consisting of the product of the penalty coefficient and the total frequency offset of the two layers. This penalty coefficient effectively constrains and suppresses the degradation of isolation performance caused by inter-layer frequency interference. The sum of the first frequency difference and the second frequency difference characterizes the total frequency offset of the two layers, comprehensively considering the deviation of the actual operating frequency of each of the upper and lower antenna layers from the ideal operating frequency. This allows the objective function to consider both the frequency offset and performance degradation of a single-layer antenna during the optimization process, while simultaneously constraining inter-layer coupling interference and the overall frequency offset level. This achieves synergistic optimization of single-layer performance, dual-layer frequency offset, and inter-layer isolation effect, avoiding the problems of increased inter-layer interference and decreased isolation caused by optimizing a single indicator. It significantly improves the frequency matching accuracy, inter-layer isolation stability, and overall operational reliability of the dual-layer dual-frequency satellite navigation antenna.

[0018] In an exemplary embodiment, the performance degradation parameters include at least one of the following: return loss, peak gain, and maximum axial ratio.

[0019] Based on the above technical means, at least one of return loss, peak gain, and maximum axial ratio is used as a performance degradation parameter, which can comprehensively characterize the electrical performance of a dual-layer dual-frequency satellite navigation antenna from multiple core dimensions such as impedance matching, radiation gain, and polarization characteristics. By jointly minimizing the above performance degradation parameters through an objective function, the antenna impedance matching effect, signal radiation capability, and polarization reception performance can be taken into account at the same time, avoiding the problem of optimizing a single performance index while ignoring the degradation of other key indicators.

[0020] In an exemplary embodiment, the performance degradation parameters include return loss, peak gain, and maximum axial ratio; the process of constructing the objective function includes: using the weighted sum of return loss, peak gain, and maximum axial ratio at the ideal operating frequency of the first-layer antenna as a first optimization objective; using the weighted sum of return loss, peak gain, and maximum axial ratio at the ideal operating frequency of the second-layer antenna as a second optimization objective; using the minimization of isolation penalty as a third optimization objective; and constructing the objective function using the first, second, and third optimization objectives.

[0021] Based on the aforementioned technical methods, return loss, peak gain, and maximum axial ratio are uniformly used as performance degradation parameters. A first and second optimization objective, calculated by weighting multiple indicators, are constructed for the first and second layer antennas respectively. This allows for comprehensive control of the electrical performance of a single-layer antenna from the perspectives of impedance matching, radiation capability, and polarization characteristics. Simultaneously, a third optimization objective, centered on minimizing isolation penalties, is added to comprehensively constrain inter-layer frequency interference and overall frequency offset. By fusing these three optimization objectives to jointly construct an overall objective function, synergistic optimization of the multi-dimensional electrical performance of the single-layer antenna, the frequency offset accuracy of the dual-layer antenna, and the inter-layer isolation effect can be achieved. This avoids the degradation of other performance aspects caused by one-sided optimization of a single indicator, taking into account both independent performance and inter-layer coupling characteristics, and significantly improving the matching characteristics, radiation gain, polarization performance, and inter-layer isolation stability of the dual-layer dual-frequency satellite navigation antenna.

[0022] In an exemplary embodiment, an objective function is constructed using a first optimization objective, a second optimization objective, and a third optimization objective, including: determining a first dynamic weight corresponding to the first optimization objective and a second dynamic weight corresponding to the second optimization objective; the sum of the first dynamic weight and the second dynamic weight is 1; constructing an objective function based on the first optimization objective, the second optimization objective, the third optimization objective, the first dynamic weight, and the second dynamic weight; wherein, when the first frequency difference is greater than the maximum allowable frequency offset threshold of the first layer antenna, the first dynamic weight attenuates exponentially; when the second frequency difference is greater than the maximum allowable frequency offset threshold of the second layer antenna, the second dynamic weight attenuates exponentially.

[0023] Based on the aforementioned technical means, a reasonable allocation of optimization weights for upper and lower antenna layers is achieved by setting a first dynamic weight and a second dynamic weight that sum to 1. Simultaneously, a maximum allowable frequency offset threshold constraint mechanism is introduced. When the frequency difference of a single-layer antenna exceeds the maximum allowable frequency offset threshold, the corresponding dynamic weight decays exponentially. This allows for adaptive adjustment of the optimization focus based on the actual frequency offset error of the two antenna layers, automatically tilting the optimization intensity towards the antenna layer with the larger frequency offset deviation. By combining the first, second, and third optimization objectives with the dynamic weights to jointly construct the objective function, it can balance the electrical performance of a single-layer antenna, the frequency offset accuracy of the two layers, and the inter-layer isolation suppression requirements. This overcomes the shortcomings of fixed weights, which cannot adapt to changes in operating conditions, improves the adaptive capability and convergence efficiency of multi-objective optimization, effectively reduces the resonant frequency offset of the two antenna layers, balances single-layer performance and inter-layer isolation effects, and enables the antenna to maintain excellent electrical characteristics and operational stability under complex operating conditions, improving the accuracy of parameter calibration and the overall optimization robustness.

[0024] In an exemplary embodiment, the objective function complies with constraints; wherein the constraints are used to constrain the range of return loss, peak gain and maximum axial ratio at the ideal operating frequency of the first-layer antenna, and the range of return loss, peak gain and maximum axial ratio at the ideal operating frequency of the second-layer antenna.

[0025] Based on the above technical means, by setting corresponding constraints for the objective function, reasonable ranges of return loss, peak gain and maximum axial ratio of the upper and lower antenna layers at the ideal operating frequency are respectively limited. This enables boundary constraints on various electrical performance indicators during the multi-objective optimization process, avoiding the problem of excessive optimization of a single objective causing other performance indicators to exceed the engineering allowable range and resulting in performance degradation and loss of control.

[0026] In an exemplary embodiment, determining the target parameters to be calibrated for the dual-layer dual-frequency satellite navigation antenna includes: determining key parameters of the dual-layer dual-frequency satellite navigation antenna; wherein, key parameters are parameters whose value perturbations will cause a shift in the first target frequency and / or the second target frequency; selecting parameters from the key parameters whose perturbation sensitivity is greater than a preset perturbation sensitivity as target parameters; wherein, the perturbation sensitivity of the key parameters is determined based on the fluctuation of the performance degradation parameters at the ideal operating frequency of the first-layer antenna and the performance degradation parameters at the ideal operating frequency of the second-layer antenna under the parameter value perturbation of the key parameters; the fluctuation includes the fluctuation amplitude and the degree of fluctuation intensity.

[0027] Based on the aforementioned technical methods, key parameters that can cause antenna target frequency shifts due to parameter disturbances are first identified. Then, the sensitivity to disturbances is judged based on the fluctuation amplitude and intensity of the parameters that degrade the performance of the two antenna layers under the disturbance. Parameters with sensitivity higher than a preset threshold are selected as target parameters to be calibrated. This method can accurately eliminate redundant parameters with weak or negligible impacts, focusing on the core parameters that have the most significant impact on frequency shifts and electrical performance for calibration and optimization, reducing the number of parameters to be optimized and lowering the computational dimensionality and solution complexity.

[0028] In an exemplary embodiment, decoupling the first-layer antenna and the second-layer antenna in the initial simulation model to obtain an intermediate simulation model includes: switching the metal boundary of the second-layer antenna or the first-layer antenna to a perfect electrical conductor boundary or an RLC boundary to decouple the first-layer antenna and the second-layer antenna in the initial simulation model to obtain an intermediate simulation model; wherein, when the target parameter being simulated is at the first-layer antenna, the metal structure setting of the second-layer antenna is switched to a perfect electrical conductor boundary; when the target parameter being simulated is at the second-layer antenna, the metal structure setting of the first-layer antenna is switched to a lumped parameter resistance-inductance-capacitance impedance boundary; the first-layer antenna is the upper antenna of the second-layer antenna.

[0029] Based on the above technical means, by switching the metal structure of the non-simulated side antenna to a perfect electrical conductor boundary or an RLC impedance boundary, the upper and lower antenna layers can be decoupled. This can accurately isolate the electromagnetic coupling and inter-layer frequency interference between the two antenna layers and construct an independent and controllable intermediate simulation model. By configuring boundary conditions differently for the target parameter level, the electromagnetic parasitic effects brought by the other antenna layer can be shielded to the greatest extent, and the perturbation characteristics of the single-level target parameter can be extracted purely.

[0030] In an exemplary embodiment, the method further includes: determining the optimal calibration parameter values ​​of the target parameters, correcting the initial simulation model to obtain an optimized simulation model of the dual-layer dual-frequency satellite navigation antenna; and performing optimized vehicle design based on the optimized simulation model.

[0031] Based on the above technical means, after obtaining the optimal calibration parameter values ​​of the target parameters, the initial simulation model can be corrected using the optimal parameters. This can eliminate the initial modeling deviation and obtain an optimized simulation model of the dual-layer dual-frequency satellite navigation antenna that better reflects the actual electrical characteristics. By relying on the high-precision optimized simulation model to carry out vehicle optimization design, the electromagnetic compatibility characteristics of the antenna in the vehicle installation environment can be predicted in advance. This allows for the rational planning of the antenna installation position, vehicle body structure layout, and RF wiring scheme, avoiding performance degradation problems caused by vehicle body obstruction, electromagnetic interference, and installation gaps.

[0032] Secondly, this application provides a parameter calibration device for a dual-layer dual-frequency satellite navigation antenna, comprising: a first determining module, a simulation module, and a second determining module; the first determining module is used to determine the target parameters to be calibrated in the dual-layer dual-frequency satellite navigation antenna; wherein, disturbances in the parameter values ​​of the target parameters will cause a shift in the first target frequency and / or the second target frequency; the first target frequency is the actual operating frequency of the first layer antenna in the dual-layer dual-frequency satellite navigation antenna; the second target frequency is the actual operating frequency of the second layer antenna in the dual-layer dual-frequency satellite navigation antenna; the simulation module is used to simulate the disturbance matrix of the target parameters using a simulation model of the dual-layer dual-frequency satellite navigation antenna; wherein, the disturbance... The matrix is ​​used to represent the correspondence between the parameter value perturbation of the target parameter and the target frequency offset; the target frequency offset is the frequency offset of the first target frequency and / or the second target frequency; the second determination module is used to determine the optimal calibration parameter value of the target parameter based on the perturbation matrix and the objective function; wherein, the objective function is used to minimize the performance degradation parameter at the ideal operating frequency of the first layer antenna and the performance degradation parameter at the ideal operating frequency of the second layer antenna; the parameter values ​​of the performance degradation parameter at the ideal operating frequency of the first layer antenna and the performance degradation parameter at the ideal operating frequency of the second layer antenna change with the change of the first target frequency and / or the second target frequency.

[0033] In an exemplary embodiment, the simulation module is specifically used to simulate the first perturbation matrix and the second perturbation matrix using a simulation model; wherein, the first perturbation matrix is ​​used to represent the correspondence between the parameter value perturbation of the target parameter and the target frequency offset under the premise of isolating the inter-layer frequency interference between the first layer antenna and the second layer antenna; the second perturbation matrix is ​​used to represent the correspondence between the parameter value perturbation of the target parameter and the inter-layer frequency interference; and the perturbation matrix is ​​determined based on the first perturbation matrix and the second perturbation matrix.

[0034] In the exemplary embodiment, the simulation model is constructed as follows: Based on the scanning information of the dual-layer dual-frequency satellite navigation antenna, an initial simulation model of the dual-layer dual-frequency satellite navigation antenna is constructed; the first-layer antenna and the second-layer antenna in the initial simulation model are decoupled to obtain an intermediate simulation model; decoupling is used to isolate inter-layer frequency interference; and the simulation model is obtained using the intermediate simulation model and the residual structure; wherein, the residual structure is used to represent the frequency offset difference of the target frequency offset when the parameter values ​​of the target parameters are the same in the initial simulation model and the intermediate simulation model.

[0035] In an exemplary embodiment, the target parameters include a first dielectric constant corresponding to the first layer antenna and a second dielectric constant corresponding to the second layer antenna; the simulation module is specifically used to input each of the preselected parameter value perturbations in at least one preselected parameter value perturbation of the target parameters into an initial simulation model and an intermediate simulation model respectively, so as to obtain the target frequency offset output by the intermediate simulation model response and the frequency offset difference output by the residual structure response; generate a first perturbation matrix using the parameter value perturbation and the target frequency offset output by the intermediate simulation model response; and generate a second perturbation matrix using the parameter value perturbation and the frequency offset difference output by the residual structure response.

[0036] In the exemplary embodiment, the target parameters include: the patch size corresponding to the first layer antenna, the patch size corresponding to the second layer antenna, and the interlayer spacing between the first layer antenna and the second layer antenna; the element values ​​in the second perturbation matrix are all 0 by default; the simulation module is specifically used to input each preselected parameter value perturbation amount in at least one preselected parameter value perturbation amount of the target parameters into the intermediate simulation model to obtain the target frequency offset output by the intermediate simulation model in response to the preselected parameter value perturbation amount; and to generate the first perturbation matrix using the parameter value perturbation amount and the target frequency offset output by the intermediate simulation model in response to the preselected parameter value perturbation amount.

[0037] In an exemplary embodiment, the objective function is further used to minimize the isolation penalty; wherein the isolation penalty is the product of the penalty coefficient and the total frequency offset of the two layers; the penalty coefficient is used to suppress the deterioration of the isolation effect of inter-layer frequency interference; the total frequency offset of the two layers is the sum of the first frequency difference and the second frequency difference; the first frequency difference is the frequency difference between the first target frequency and the ideal operating frequency of the first layer antenna; the second frequency difference is the frequency difference between the second target frequency and the ideal operating frequency of the second layer antenna.

[0038] In an exemplary embodiment, the performance degradation parameters include at least one of the following: return loss, peak gain, and maximum axial ratio.

[0039] In an exemplary embodiment, the performance degradation parameters include return loss, peak gain, and maximum axial ratio; the process of constructing the objective function includes: using the weighted sum of return loss, peak gain, and maximum axial ratio at the ideal operating frequency of the first-layer antenna as a first optimization objective; using the weighted sum of return loss, peak gain, and maximum axial ratio at the ideal operating frequency of the second-layer antenna as a second optimization objective; using the minimization of isolation penalty as a third optimization objective; and constructing the objective function using the first, second, and third optimization objectives.

[0040] In an exemplary embodiment, an objective function is constructed using a first optimization objective, a second optimization objective, and a third optimization objective, including: determining a first dynamic weight corresponding to the first optimization objective and a second dynamic weight corresponding to the second optimization objective; the sum of the first dynamic weight and the second dynamic weight is 1; constructing an objective function based on the first optimization objective, the second optimization objective, the third optimization objective, the first dynamic weight, and the second dynamic weight; wherein, when the first frequency difference is greater than the maximum allowable frequency offset threshold of the first layer antenna, the first dynamic weight attenuates exponentially; when the second frequency difference is greater than the maximum allowable frequency offset threshold of the second layer antenna, the second dynamic weight attenuates exponentially.

[0041] In an exemplary embodiment, the objective function complies with constraints; wherein the constraints are used to constrain the range of return loss, peak gain and maximum axial ratio at the ideal operating frequency of the first-layer antenna, and the range of return loss, peak gain and maximum axial ratio at the ideal operating frequency of the second-layer antenna.

[0042] In an exemplary embodiment, the first determining module is specifically used to determine key parameters of the dual-layer dual-frequency satellite navigation antenna; wherein, the key parameters are parameters whose value perturbation will cause the first target frequency and / or the second target frequency to shift; parameters with perturbation sensitivity greater than a preset perturbation sensitivity are selected from the key parameters as target parameters; wherein, the perturbation sensitivity of the key parameters is determined based on the fluctuation of the performance degradation parameters of the first layer antenna at the ideal operating frequency and the performance degradation parameters of the second layer antenna at the ideal operating frequency under the parameter value perturbation of the key parameters; the fluctuation includes the fluctuation amplitude and the degree of fluctuation intensity.

[0043] In an exemplary embodiment, decoupling the first-layer antenna and the second-layer antenna in the initial simulation model to obtain an intermediate simulation model includes: switching the metal boundary of the second-layer antenna or the first-layer antenna to a perfect electrical conductor boundary or an RLC boundary to decouple the first-layer antenna and the second-layer antenna in the initial simulation model to obtain an intermediate simulation model; wherein, when the target parameter being simulated is at the first-layer antenna, the metal structure setting of the second-layer antenna is switched to a perfect electrical conductor boundary; when the target parameter being simulated is at the second-layer antenna, the metal structure setting of the first-layer antenna is switched to a lumped parameter resistance-inductance-capacitance impedance boundary; the first-layer antenna is the upper antenna of the second-layer antenna.

[0044] In an exemplary embodiment, the second determining module is further configured to, after determining the optimal calibration parameter values ​​of the target parameters, correct the initial simulation model to obtain an optimized simulation model of the dual-layer dual-frequency satellite navigation antenna; and, based on the optimized simulation model, perform optimized vehicle design.

[0045] Thirdly, this application provides a vehicle comprising: a vehicle body and a dual-layer dual-frequency satellite navigation antenna for parameter calibration using the parameter calibration device of the dual-layer dual-frequency satellite navigation antenna described in the second aspect above.

[0046] Fourthly, this application provides an electronic device comprising: a processor and a memory; the memory storing instructions executable by the processor. When the processor is configured to execute the instructions, the electronic device implements the parameter calibration method for the dual-layer dual-frequency satellite navigation antenna described in the first aspect.

[0047] Fifthly, this application provides a computer-readable storage medium storing computer program instructions that, when executed by a processor, implement the parameter calibration method for the dual-layer dual-frequency satellite navigation antenna described in the first aspect.

[0048] In a sixth aspect, this application provides a computer program product comprising computer program instructions that, when executed by a processor, implement the parameter calibration method for the dual-layer dual-frequency satellite navigation antenna described in the first aspect.

[0049] It should be noted that the technical effects of any of the implementation methods in aspects two through six can be found in the technical effects of the corresponding implementation methods in aspect one, and will not be repeated here.

[0050] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0051] To more clearly illustrate the technical solutions in the embodiments of this application or the background art, the accompanying drawings used in the embodiments of this application will be described below.

[0052] Figure 1 This is a schematic diagram of the structure of a vehicle disclosed in an embodiment of this application; Figure 2 This is a functional architecture diagram of a parameter calibration device for a dual-layer dual-frequency satellite navigation antenna disclosed in an embodiment of this application; Figure 3 This is a flowchart illustrating a parameter calibration method for a dual-layer dual-frequency satellite navigation antenna disclosed in an embodiment of this application. Figure 4 This is a flowchart illustrating another parameter calibration method for a dual-layer dual-frequency satellite navigation antenna disclosed in an embodiment of this application; Figure 5 This is a flowchart illustrating another parameter calibration method for a dual-layer dual-frequency satellite navigation antenna disclosed in an embodiment of this application; Figure 6 This is a schematic diagram of the structure of a dual-layer dual-frequency satellite navigation antenna disclosed in an embodiment of this application; Figure 7 This is a schematic diagram of the structure of another dual-layer dual-frequency satellite navigation antenna disclosed in the embodiments of this application; Figure 8 This is a schematic diagram of the structure of another dual-layer dual-frequency satellite navigation antenna disclosed in the embodiments of this application; Figure 9 This is a schematic diagram of the decoupling structure of a dual-layer dual-frequency satellite navigation antenna disclosed in an embodiment of this application; Figure 10 This is a schematic diagram of another decoupling structure for a dual-layer dual-frequency satellite navigation antenna disclosed in this application. Figure 11 This is a flowchart illustrating another parameter calibration method for a dual-layer dual-frequency satellite navigation antenna disclosed in an embodiment of this application; Figure 12 This is a flowchart illustrating another parameter calibration method for a dual-layer dual-frequency satellite navigation antenna disclosed in an embodiment of this application; Figure 13 This is a schematic diagram illustrating the stages of a pre-compensation operation disclosed in an embodiment of this application; Figure 14 This is a schematic diagram illustrating the stages of another pre-compensation operation disclosed in an embodiment of this application; Figure 15 This is a schematic diagram illustrating the stages of another pre-compensation operation disclosed in an embodiment of this application; Figure 16 This is a schematic diagram comparing matrix pre-compensation and no compensation as disclosed in an embodiment of this application; Figure 17 This is a linear schematic diagram of the perturbation amount of a preselected parameter value disclosed in an embodiment of this application; Figure 18 This is a flowchart illustrating another parameter calibration method for a dual-layer dual-frequency satellite navigation antenna disclosed in an embodiment of this application; Figure 19 This is a flowchart illustrating another parameter calibration method for a dual-layer dual-frequency satellite navigation antenna disclosed in an embodiment of this application; Figure 20 This is a flowchart illustrating another parameter calibration method for a dual-layer dual-frequency satellite navigation antenna disclosed in an embodiment of this application; Figure 21 This is a schematic diagram of the structure of an electronic device disclosed in an embodiment of this application. Detailed Implementation

[0053] The terms "first," "second," etc., are used for descriptive purposes only and have no sequential or technical meaning, nor should they be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Directional terms used in this application, such as "upper," "lower," "front," "rear," "left," "right," "inner," and "outer," are merely for reference to the orientation shown in the accompanying drawings. The use of directional terms is for better and clearer explanation and understanding of this application, and does not indicate the orientation of the referred device or component in an actual application scenario.

[0054] In the description of the embodiments of this application, unless otherwise expressly specified and limited, the terms "installation" and "connection" should be interpreted broadly. For example, "connection" can be a detachable connection or a non-detachable connection; it can be a direct connection or an indirect connection through an intermediate medium.

[0055] In the embodiments of this application, "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.

[0056] The embodiments of this application are described below with reference to the accompanying drawings.

[0057] Please see Figure 1 , Figure 1 This is a schematic diagram of the structure of a vehicle disclosed in an embodiment of this application. The vehicle 100 can be, but is not limited to, a pure electric vehicle (PEV / BEV), a hybrid electric vehicle (HEV), a range-extended electric vehicle (REEV), a plug-in hybrid electric vehicle (PHEV), or a new energy vehicle.

[0058] In this embodiment, the vehicle 100 includes a body 101 and a dual-layer dual-frequency satellite navigation antenna 102. The dual-layer dual-frequency satellite navigation antenna 102 can be fixed to a preset mounting point on the inner or outer side of the roof of the body 101, and connected to the body 101 by means of snap-fit, strong adhesive, or bolt fastening. This position has a wide field of view and minimal obstruction, which can effectively ensure satellite signal reception. Alternatively, the dual-layer dual-frequency satellite navigation antenna 102 can be embedded in the inner edge of the windshield or rear windshield of the body 101, integrated with the glass, avoiding interference from surrounding devices, and saving external space of the body 101. Or, the dual-layer dual-frequency satellite navigation antenna 102 can be mounted on a preset mounting seat on the trunk cover of the body 101, and fixed to the body by a waterproof sealing structure, adapting to the overall vehicle electrical layout, and balancing signal reception and the integrity of the vehicle's appearance. It should be noted that the above setup methods must ensure that the receiving surface of the dual-layer dual-frequency satellite navigation antenna 102 is not significantly obstructed and is far away from electromagnetic interference sources such as lidar and ceiling speakers to ensure the stable performance of the dual-layer dual-frequency satellite navigation antenna 102.

[0059] For example, the dual-layer dual-frequency satellite navigation antenna 102 (GNSS antenna) is a satellite navigation antenna that adopts a dual-layer radiating patch structure and supports dual-frequency (such as L1 / L5 band) reception of multiple satellite navigation systems such as BeiDou (BDS) and GPS. It integrates signal filtering, amplification and anti-interference modules to receive navigation radio frequency signals transmitted by satellites, realize dual-frequency signal diversity reception, interference suppression and signal conditioning. In vehicles, it mainly plays a core role in high-precision positioning, heading angle calculation, driving navigation timing and intelligent driving positioning assistance, providing reliable position and time references for vehicle autonomous driving, path planning and precise navigation functions.

[0060] It should be understood that, in situations where a physical antenna is available but its parameter values ​​cannot be obtained, existing technologies typically perform parameter calibration by conducting point-by-point tests using specialized equipment such as on-vehicle road tests, microwave anechoic chamber tests, vector network analyzers, and high-precision turntables. This method not only relies on expensive specialized equipment and specific test scenarios but is also cumbersome, complex, time-consuming, and costly, making it difficult to adapt to the needs of large-scale vehicle development and production. This application provides a parameter calibration device for a dual-layer, dual-frequency satellite navigation antenna, which can calibrate the parameters of the dual-layer, dual-frequency satellite navigation antenna through simulation modeling and optimization.

[0061] Please see Figure 2 , Figure 2 This is a functional architecture diagram of a parameter calibration device for a dual-layer dual-frequency satellite navigation antenna disclosed in an embodiment of this application. The parameter calibration device 200 for the dual-layer dual-frequency satellite navigation antenna includes: a first determining module 201, a simulation module 202, and a second determining module 203. The first determining module 201 is used to determine the target parameters to be calibrated for the dual-layer dual-frequency satellite navigation antenna; wherein, perturbation of the target parameter values ​​will cause a shift in the first target frequency and / or the second target frequency; the first target frequency is the actual operating frequency of the first layer antenna in the dual-layer dual-frequency satellite navigation antenna; the second target frequency is the actual operating frequency of the second layer antenna in the dual-layer dual-frequency satellite navigation antenna; the simulation module 202 is used to simulate the perturbation matrix of the target parameters using a simulation model of the dual-layer dual-frequency satellite navigation antenna; wherein, the perturbation... The perturbation matrix is ​​used to represent the correspondence between the parameter value perturbation of the target parameter and the target frequency offset; the target frequency offset is the frequency offset of the first target frequency and / or the second target frequency; the second determination module 203 is used to determine the optimal calibration parameter value of the target parameter based on the perturbation matrix and the objective function; wherein, the objective function is used to minimize the performance degradation parameter at the ideal operating frequency of the first layer antenna and the performance degradation parameter at the ideal operating frequency of the second layer antenna; the parameter values ​​of the performance degradation parameter at the ideal operating frequency of the first layer antenna and the performance degradation parameter at the ideal operating frequency of the second layer antenna change with the change of the first target frequency and / or the second target frequency.

[0062] In an exemplary embodiment, the simulation module 202 is specifically used to simulate the first perturbation matrix and the second perturbation matrix using a simulation model; wherein, the first perturbation matrix is ​​used to represent the correspondence between the parameter value perturbation of the target parameter and the target frequency offset under the premise of isolating the inter-layer frequency interference between the first layer antenna and the second layer antenna; the second perturbation matrix is ​​used to represent the correspondence between the parameter value perturbation of the target parameter and the inter-layer frequency interference; and the perturbation matrix is ​​determined based on the first perturbation matrix and the second perturbation matrix.

[0063] In the exemplary embodiment, the simulation model is constructed as follows: Based on the scanning information of the dual-layer dual-frequency satellite navigation antenna, an initial simulation model of the dual-layer dual-frequency satellite navigation antenna is constructed; the first-layer antenna and the second-layer antenna in the initial simulation model are decoupled to obtain an intermediate simulation model; decoupling is used to isolate inter-layer frequency interference; and the simulation model is obtained using the intermediate simulation model and the residual structure; wherein, the residual structure is used to represent the frequency offset difference of the target frequency offset when the parameter values ​​of the target parameters are the same in the initial simulation model and the intermediate simulation model.

[0064] In the exemplary embodiment, the target parameters include the first dielectric constant corresponding to the first layer antenna and the second dielectric constant corresponding to the second layer antenna; the simulation module 202 is specifically used to input each of the preselected parameter value perturbations in at least one preselected parameter value perturbation of the target parameters into the initial simulation model and the intermediate simulation model respectively, so as to obtain the target frequency offset output by the intermediate simulation model response and the frequency offset difference output by the residual structure response; generate a first perturbation matrix using the parameter value perturbation and the target frequency offset output by the intermediate simulation model response; and generate a second perturbation matrix using the parameter value perturbation and the frequency offset difference output by the residual structure response.

[0065] In the exemplary embodiment, the target parameters include: the patch size corresponding to the first layer antenna, the patch size corresponding to the second layer antenna, and the interlayer spacing between the first layer antenna and the second layer antenna; the element values ​​in the second perturbation matrix are all 0 by default; the simulation module 202 is specifically used to input each of the preselected parameter value perturbation quantities in at least one preselected parameter value perturbation quantity of the target parameters into the intermediate simulation model to obtain the target frequency offset output by the intermediate simulation model in response to the preselected parameter value perturbation quantity; and to generate the first perturbation matrix using the parameter value perturbation quantity and the target frequency offset output by the intermediate simulation model in response to the preselected parameter value perturbation quantity.

[0066] In an exemplary embodiment, the objective function is further used to minimize the isolation penalty; wherein the isolation penalty is the product of the penalty coefficient and the total frequency offset of the two layers; the penalty coefficient is used to suppress the deterioration of the isolation effect of inter-layer frequency interference; the total frequency offset of the two layers is the sum of the first frequency difference and the second frequency difference; the first frequency difference is the frequency difference between the first target frequency and the ideal operating frequency of the first layer antenna; the second frequency difference is the frequency difference between the second target frequency and the ideal operating frequency of the second layer antenna.

[0067] In an exemplary embodiment, the performance degradation parameters include at least one of the following: return loss, peak gain, and maximum axial ratio.

[0068] In an exemplary embodiment, the performance degradation parameters include return loss, peak gain, and maximum axial ratio; the process of constructing the objective function includes: using the weighted sum of return loss, peak gain, and maximum axial ratio at the ideal operating frequency of the first-layer antenna as a first optimization objective; using the weighted sum of return loss, peak gain, and maximum axial ratio at the ideal operating frequency of the second-layer antenna as a second optimization objective; using the minimization of isolation penalty as a third optimization objective; and constructing the objective function using the first, second, and third optimization objectives.

[0069] In an exemplary embodiment, an objective function is constructed using a first optimization objective, a second optimization objective, and a third optimization objective, including: determining a first dynamic weight corresponding to the first optimization objective and a second dynamic weight corresponding to the second optimization objective; the sum of the first dynamic weight and the second dynamic weight is 1; constructing an objective function based on the first optimization objective, the second optimization objective, the third optimization objective, the first dynamic weight, and the second dynamic weight; wherein, when the first frequency difference is greater than the maximum allowable frequency offset threshold of the first layer antenna, the first dynamic weight attenuates exponentially; when the second frequency difference is greater than the maximum allowable frequency offset threshold of the second layer antenna, the second dynamic weight attenuates exponentially.

[0070] In an exemplary embodiment, the objective function complies with constraints; wherein the constraints are used to constrain the range of return loss, peak gain and maximum axial ratio at the ideal operating frequency of the first-layer antenna, and the range of return loss, peak gain and maximum axial ratio at the ideal operating frequency of the second-layer antenna.

[0071] In an exemplary embodiment, the first determining module 201 is specifically used to determine key parameters of the dual-layer dual-frequency satellite navigation antenna; wherein, the key parameters are parameters whose parameter values ​​perturbation will cause the first target frequency and / or the second target frequency to shift; parameters with perturbation sensitivity greater than a preset perturbation sensitivity are selected from the key parameters as target parameters; wherein, the perturbation sensitivity of the key parameters is determined based on the fluctuation of the performance degradation parameters of the first layer antenna at the ideal operating frequency and the performance degradation parameters of the second layer antenna at the ideal operating frequency under the parameter value perturbation of the key parameters; the fluctuation includes the fluctuation amplitude and the degree of fluctuation intensity.

[0072] In an exemplary embodiment, decoupling the first-layer antenna and the second-layer antenna in the initial simulation model to obtain an intermediate simulation model includes: switching the metal boundary of the second-layer antenna or the first-layer antenna to a perfect electrical conductor boundary or an RLC boundary to decouple the first-layer antenna and the second-layer antenna in the initial simulation model to obtain an intermediate simulation model; wherein, when the target parameter being simulated is at the first-layer antenna, the metal structure setting of the second-layer antenna is switched to a perfect electrical conductor boundary; when the target parameter being simulated is at the second-layer antenna, the metal structure setting of the first-layer antenna is switched to a lumped parameter resistance-inductance-capacitance impedance boundary; the first-layer antenna is the upper antenna of the second-layer antenna.

[0073] In an exemplary embodiment, the second determining module 203 is further configured to, after determining the optimal calibration parameter value of the target parameter, correct the initial simulation model to obtain an optimized simulation model of the dual-layer dual-frequency satellite navigation antenna; and, based on the optimized simulation model, perform vehicle optimization design.

[0074] Please see Figure 3 , Figure 3 This is a flowchart illustrating a parameter calibration method for a dual-layer dual-frequency satellite navigation antenna disclosed in an embodiment of this application.

[0075] In some embodiments, an initial model of a dual-layer dual-frequency satellite navigation antenna can be constructed first, and the parameter values ​​of the target parameters to be calibrated can be continuously adjusted until the optimal calibration parameter values ​​are determined.

[0076] As a feasible implementation method, the parameter calibration method for a dual-layer dual-frequency satellite navigation antenna may include the following steps: S301. Determine the target parameters to be calibrated for the dual-layer dual-frequency satellite navigation antenna.

[0077] As a feasible implementation method, the target parameters include at least one of the following: the first dielectric constant corresponding to the first layer antenna, the patch size corresponding to the first layer antenna, the second dielectric constant corresponding to the second layer antenna, the patch size corresponding to the second layer antenna, and the interlayer spacing between the first layer antenna and the second layer antenna. The patch size includes: patch length and patch width.

[0078] In an exemplary embodiment, perturbation of the target parameter value will cause a shift in the first target frequency and / or the second target frequency.

[0079] The first target frequency is the actual operating frequency of the first layer antenna in the dual-layer dual-frequency satellite navigation antenna. It can receive satellite signals in the first navigation frequency band and realize signal transmission and reception, resonance matching and radiation coverage in the corresponding frequency band.

[0080] The second target frequency is the actual operating frequency of the second layer antenna in the dual-layer dual-frequency satellite navigation antenna. It can receive satellite signals in the second navigation frequency band and complete dual-frequency diversity reception, anti-interference filtering, and high-precision positioning signal analysis.

[0081] For example, the size of the antenna patch directly determines the resonant electrical length. Increasing the size corresponds to a shift of the actual operating frequency to a lower frequency, while decreasing the size corresponds to a shift of the actual operating frequency to a higher frequency. The higher the dielectric constant of the dielectric layer, the shorter the equivalent wavelength and the greater the electrical length, which reduces the overall resonant frequency of the antenna. Changes in the spacing between the upper and lower antenna layers will cause changes in the interlayer electromagnetic coupling strength, thereby interfering with impedance matching and resonance conditions, resulting in a shift in the target frequency, bandwidth and VSWR characteristics of each of the two frequencies.

[0082] As a feasible approach, multiple candidate structural and dielectric parameters of the dual-layer dual-frequency satellite navigation antenna to be calibrated are first obtained. The sensitivity and influence weight of each parameter on the antenna resonant frequency, impedance characteristics, and radiation performance are determined. Key parameters with high sensitivity and decisive effect on dual-frequency operating characteristics are then selected, thereby determining the target parameters of the dual-layer dual-frequency satellite navigation antenna to be calibrated.

[0083] It should be understood that different structural and dielectric parameters have varying degrees of influence on the two operating frequency bands of a dual-layer dual-band antenna. Some parameters are only sensitive to a single frequency band, while others will simultaneously couple and affect the dual-band resonance characteristics. By quantifying the contribution and sensitivity of each parameter to the target frequency, redundant and irrelevant parameters can be eliminated, and only the target parameters that play a dominant role in the dual-band operating performance can be retained for calibration. This ensures calibration accuracy, reduces the number of parameters to be optimized, and improves simulation optimization efficiency.

[0084] S302. Using a simulation model of a dual-layer dual-frequency satellite navigation antenna, simulate the disturbance matrix of the target parameters.

[0085] As a feasible implementation method, the perturbation matrix is ​​used to represent the correspondence between the parameter value perturbation of the target parameter and the target frequency offset.

[0086] The target frequency offset is the frequency offset of the first target frequency and / or the second target frequency.

[0087] In the exemplary embodiment, based on the constructed simulation model of the dual-layer dual-frequency satellite navigation antenna, preset small numerical perturbations are applied to each target parameter to be calibrated. Batch simulations are performed to obtain the first target frequency offset and the second target frequency offset corresponding to each set of parameter perturbations. With the target parameters as independent variables and the frequency offsets as dependent variables, matrix quantization and arrangement are performed to form a multi-parameter-multi-frequency coupling perturbation matrix. This perturbation matrix can intuitively characterize the sensitivity, coupling influence weight, and linear change law of each target parameter, such as the dielectric constant of the first layer antenna, the size of the first layer patch, the dielectric constant of the second layer antenna, the size of the second layer patch, and the spacing between the upper and lower layers, to the first target frequency and the second target frequency, respectively.

[0088] As a feasible implementation method, the simulation model of the dual-layer dual-frequency satellite navigation antenna is used as an electromagnetic simulation model to characterize the antenna's real physical structure, dielectric properties, and feeding characteristics. The simulation model can be constructed by non-destructive scanning and measurement of the physical dual-layer dual-frequency satellite navigation antenna using a 3D scanner.

[0089] In the exemplary embodiment, the physical dual-layer dual-frequency satellite navigation antenna provided by the supplier is disassembled non-destructively. A 3D scanner is used to obtain the initial geometric dimensions of the antenna patches (measurement accuracy up to ±0.02mm); simultaneously, a dial indicator is used to measure the interlayer spacing of the upper and lower antennas (measurement accuracy up to ±0.05mm). During the construction of the simulation model of the dual-layer dual-frequency satellite navigation antenna, the dielectric constants of the upper and lower ceramic substrates are set as variable parameters to be calibrated, and their initial value range is set to 7.0–20.0 based on the properties of alumina ceramic materials. Finally, the feed network is modeled using an equivalent circuit model, and the initial dielectric constant of the feed section is set, completing the parameter assignment and model construction of the initial simulation model of the dual-layer dual-frequency satellite navigation antenna.

[0090] Optionally, since the feed network is mainly used to achieve impedance matching and signal input transmission for the dual-layer dual-frequency satellite navigation antenna, and the core purpose of this application is to calibrate and optimize the performance-related parameters of the dual-layer dual-frequency satellite navigation antenna itself, rather than optimizing the parameters of the feed network itself, the feed network structure can be ignored or simplified while meeting the simulation accuracy requirements. Only the main antenna radiating structure can be modeled and simulated, thereby effectively reducing the complexity of the simulation model, decreasing the computational load, and balancing calibration accuracy and simulation efficiency.

[0091] It should be understood that obtaining initial geometric values ​​through non-destructive physical measurements and setting initial values ​​for dielectric parameters in conjunction with typical material ranges can ensure that the simulation model is highly consistent with the physical characteristics of the real antenna. Based on this, the perturbation matrix obtained by simulation has parameter sensitivity and coupling law that are closer to the actual working conditions of the vehicle, which can provide a reliable model basis for subsequent antenna parameter iteration optimization and accurate calibration.

[0092] S303. Based on the perturbation matrix and the objective function, determine the optimal calibration parameter values ​​for the objective parameters.

[0093] As a feasible implementation method, the objective function is used to minimize the performance degradation parameters of the first-layer antenna at the ideal operating frequency and the second-layer antenna at the ideal operating frequency.

[0094] The performance degradation parameters of the first-layer antenna at its ideal operating frequency and the performance degradation parameters of the second-layer antenna at their ideal operating frequency change with the change of the first target frequency and / or the second target frequency.

[0095] Performance degradation parameters include at least one of the following: return loss, peak gain, and maximum axial ratio. Return loss characterizes the degree of signal reflection at the antenna feed port; a lower return loss value indicates less signal reflection and better matching between the antenna and the feed system. Peak gain characterizes the antenna's signal amplification capability in the optimal radiation direction; a higher peak gain value indicates a stronger ability to receive and radiate satellite navigation signals. Maximum axial ratio characterizes the antenna's polarization performance; a maximum axial ratio value closer to 1 indicates better circular polarization performance and more accurate reception of satellite navigation circularly polarized signals.

[0096] It should be understood that the performance degradation parameter at the ideal operating frequency is the core evaluation indicator for measuring the calibration effect of a dual-layer dual-frequency satellite navigation antenna. The magnitude of the performance degradation parameter directly reflects the accuracy of the target parameter calibration. Moreover, the corresponding dual-layer dual-frequency satellite navigation antenna can only meet the actual usage requirements of high-precision vehicle positioning, navigation, and timing when all performance degradation parameters are within a preset reasonable range. If the performance degradation parameter exceeds the preset range, it will lead to a decrease in antenna receiving sensitivity and a reduction in positioning accuracy, making it unsuitable for the application scenarios of intelligent connected vehicles.

[0097] In the exemplary embodiment, a set of performance degradation parameters (at least one of return loss degradation, peak gain degradation, and maximum axial ratio degradation) at the ideal operating frequency of the first-layer antenna and a set of performance degradation parameters (at least one of return loss degradation, peak gain degradation, and maximum axial ratio degradation) at the ideal operating frequency of the second-layer antenna are defined respectively. A performance degradation quantification expression for the first-layer antenna is constructed based on the performance degradation parameter set at the ideal operating frequency of the first-layer antenna; a performance degradation quantification expression for the second-layer antenna is constructed based on the performance degradation parameter set at the ideal operating frequency of the second-layer antenna. Combining the performance degradation quantification expressions for the first and second-layer antennas yields the objective function. The performance degradation quantification expressions for the first and second-layer antennas are constructed when the first and second-layer antennas are decoupled.

[0098] It should be understood that the performance degradation quantification expression is constructed using a weighted summation method. That is, each performance degradation parameter is assigned a corresponding weight coefficient, and the performance degradation quantification value of a single-layer antenna is obtained by multiplying each performance degradation parameter by its weight coefficient and summing the results. The weight coefficients are set according to the degree of influence of each performance degradation parameter on the antenna positioning performance.

[0099] As another feasible implementation method, in order to further suppress frequency interference between the two antenna layers, ensure the independence and stability of dual-frequency operation, and avoid the deterioration of the interlayer electromagnetic isolation effect due to excessive dual-frequency offset, the objective function is also used to minimize the isolation penalty.

[0100] The isolation penalty is the product of the penalty coefficient and the total frequency offset of the two layers.

[0101] The penalty coefficient is used to suppress the deterioration of the isolation effect of inter-layer frequency interference. It is determined based on the preset inter-layer isolation requirements of the dual-layer dual-frequency satellite navigation antenna, the allowable range of target frequency offset, and actual electromagnetic interference test data. The higher the isolation requirements and the narrower the allowable range of frequency offset, the larger the penalty coefficient value, thereby strengthening the constraint on frequency offset and avoiding the aggravation of inter-layer interference.

[0102] The total frequency offset of the dual-layer is the sum of the first frequency difference and the second frequency difference, and is used to quantify the overall offset of the dual frequencies.

[0103] The first frequency difference is the frequency difference between the first target frequency and the ideal operating frequency of the first layer antenna.

[0104] The second frequency difference is the frequency difference between the second target frequency and the ideal operating frequency of the second layer antenna.

[0105] In an exemplary embodiment, an isolation penalty term is added to the basic objective function (i.e., the objective function consisting only of the performance degradation quantification expression of the first and second layer antennas) without introducing an isolation penalty term to obtain an optimized objective function. The isolation penalty term is obtained when the first and second layer antennas are coupled.

[0106] As a feasible implementation method, the perturbation matrix serves as the core guiding basis for parameter adjustment. It can clarify the direction of parameter value adjustment for each target parameter, ensure that the iterative optimization process is efficient and accurate, avoid blind optimization, shorten the optimization cycle, and improve calibration efficiency.

[0107] In the exemplary embodiment, the perturbation matrix quantifies the correspondence between the minute perturbations of each target parameter (first dielectric constant, patch size corresponding to the first layer antenna, second dielectric constant, patch size corresponding to the second layer antenna, and interlayer spacing) and the offsets of the first and second target frequencies. The direction of parameter adjustment can be directly determined through the perturbation matrix: if the first target frequency is found to be higher than its ideal operating frequency during the iteration process, the rule in the perturbation matrix that "increasing the patch size corresponding to the first layer antenna can cause the first target frequency to shift to a lower frequency" can be used to determine that the patch size corresponding to the first layer antenna needs to be increased; if the second target frequency is found to be lower than its ideal operating frequency, the rule in the perturbation matrix that "decreasing the second dielectric constant can cause the second target frequency to shift to a higher frequency" can be used to determine that the second dielectric constant needs to be decreased. Through this guidance, the parameters can be precisely adjusted and quickly approach the optimal value.

[0108] As a feasible implementation method, based on the antenna simulation model built above, the disturbance matrix obtained from the simulation, and the constructed objective function, the values ​​of each target parameter to be calibrated are initialized (the initial values ​​are the measured initial values ​​of the physical object or typical values ​​of the material). Guided by the disturbance matrix, small adjustments are made to each target parameter, and the corresponding performance degradation parameters and the total frequency offset of the two layers are obtained through simulation. The objective function value is then calculated. Iterative optimization algorithms such as gradient descent or genetic algorithm are used to repeat the process of parameter adjustment, simulation calculation, and objective function value solving, gradually optimizing the objective function value until the objective function value reaches the preset minimum value (i.e., the performance degradation parameters are within a reasonable range and the isolation penalty meets the requirements). The corresponding target parameter values ​​at this time are determined as the optimal calibration parameter values ​​of the target parameters.

[0109] It should be understood that guiding parameter adjustment direction through the perturbation matrix can avoid blindness in the iterative optimization process and significantly shorten the optimization cycle. Combined with the dual constraints of the objective function (performance degradation + isolation penalty), it can ensure that the antenna corresponding to the optimal calibration parameter value not only meets the dual-frequency performance requirements but also effectively suppresses inter-layer frequency interference, making it suitable for the actual application scenarios of intelligent connected vehicles. At the same time, this method does not rely on actual vehicle testing and dedicated anechoic chamber equipment; calibration can be completed solely through simulation modeling and iterative optimization, solving the pain points of complex, costly, and time-consuming calibration in existing technologies and improving calibration efficiency and accuracy.

[0110] In some embodiments, in order to simplify the complexity of parameter calibration of dual-layer dual-frequency satellite navigation antennas, reduce the computational load of iterative optimization, improve calibration efficiency, and ensure that the calibration accuracy meets the actual usage requirements, the target parameters to be calibrated can be screened and determined based on the degree of influence of key parameters in the dual-layer dual-frequency satellite navigation antenna on antenna performance, and redundant parameters with a weak impact on performance can be eliminated, so as to achieve the goal of "accurate calibration and efficient optimization".

[0111] As a feasible implementation method, such as Figure 4 As shown, step S301 above can be specifically implemented as follows: S401. Determine the key parameters of the dual-layer dual-frequency satellite navigation antenna.

[0112] Among them, the key parameter is the parameter whose value perturbation will cause the first target frequency and / or the second target frequency to shift.

[0113] As a feasible implementation method, based on the simulation model of a dual-layer dual-frequency satellite navigation antenna, all potential structural parameters and dielectric parameters of the antenna are subjected to disturbance tests one by one. The simulation obtains the first target frequency offset, the second target frequency offset, and the changes in antenna performance degradation parameters (return loss, peak gain, and maximum axial ratio) after each parameter disturbance. By quantitatively analyzing the frequency offset influence coefficient and performance influence weight of each parameter, parameters with influence coefficients and weights reaching preset thresholds are selected and identified as key parameters to ensure that key parameters can comprehensively cover the core factors affecting the dual-frequency performance of the antenna.

[0114] In the exemplary embodiment, a complete simulation model of a dual-layer dual-frequency satellite navigation antenna is built. The equivalent dielectric constant of the feeding network, the distance between feeding points, the first dielectric constant (first layer antenna substrate), the patch size of the first layer antenna, the second dielectric constant (second layer antenna substrate), the patch size of the second layer antenna, and the interlayer spacing of the first and second layer antennas are selected as potential parameters. Small perturbations of the same amplitude are applied to each potential parameter, and batch simulations are performed to obtain the frequency offset and performance change corresponding to each parameter. A frequency offset influence threshold and a performance influence weight threshold are set. All parameters that meet the criteria of "frequency offset greater than or equal to the preset offset threshold" or "performance influence weight greater than or equal to the preset weight threshold" are identified as key parameters. Finally, the above 7 types of core key parameters are selected.

[0115] For example, key parameters include: the equivalent dielectric constant corresponding to the feed network, the distance between feed points, the first dielectric constant, the patch size corresponding to the first layer antenna, the second dielectric constant, the patch size corresponding to the second layer antenna, and the interlayer spacing between the first layer antenna and the second layer antenna.

[0116] Among them, the feed point distance refers to the spacing between the feed excitation points of the dual-layer dual-frequency satellite navigation antenna relative to the center of the antenna radiating patch or relative to the edge of the substrate, and is used to characterize the geometric offset between the feed position and the radiating body.

[0117] S402. Select the parameter with a disturbance sensitivity greater than the preset disturbance sensitivity from the key parameters and use it as the target parameter.

[0118] As a feasible implementation method, the perturbation sensitivity of key parameters is determined based on the fluctuation of the performance degradation parameters of the first-layer antenna at the ideal operating frequency and the performance degradation parameters of the second-layer antenna at the ideal operating frequency under the perturbation of the key parameter values.

[0119] The fluctuation characteristics include both amplitude and intensity. Amplitude is determined by the ratio of the maximum difference between the performance degradation parameter before and after the key parameter disturbance to its initial value. This ratio quantifies the overall range of change in the performance degradation parameter; a larger ratio indicates a larger amplitude and a more significant impact of the key parameter on performance. Intensity is determined by the rate of change of the performance degradation parameter corresponding to a unit disturbance in the key parameter. A larger rate of change indicates a more intense fluctuation and a higher sensitivity of the key parameter to performance.

[0120] The preset disturbance sensitivity is used to characterize the minimum sensitivity standard for a key parameter to be judged as a target parameter. The value of the preset disturbance sensitivity is set based on the performance requirements, calibration accuracy requirements and engineering experience of the dual-layer dual-frequency satellite navigation antenna. If the disturbance sensitivity of a key parameter exceeds the preset disturbance sensitivity, it means that its impact on antenna performance meets the calibration requirements and needs to be calibrated as a target parameter; if it is lower than the preset disturbance sensitivity, it means that its impact is weak and can be eliminated as a redundant parameter.

[0121] In the exemplary embodiment, key parameters are sampled and generated using hierarchical sampling technology, performance response values ​​are obtained based on finite element simulation, the perturbation sensitivity is quantified by calculating the first-order sensitivity index, and finally the target parameters are selected by combining the preset perturbation sensitivity.

[0122] For example, a hierarchical sampling technique is used to generate 500 sets of samples for key parameters. The sampling range of each key parameter is set in combination with the material properties and structural design requirements of the dual-layer dual-frequency satellite navigation antenna. Based on the simulation model of the dual-layer dual-frequency satellite navigation antenna, finite element simulations are run in batches to obtain the performance degradation parameter response values ​​corresponding to each set of samples, namely the specific values ​​of return loss (S11), peak gain (Gain), and maximum axial ratio (AR). The disturbance sensitivity of each key parameter is quantified by calculating the first-order sensitivity index. A disturbance sensitivity judgment threshold (i.e., preset disturbance sensitivity) is set, and the target parameters are screened in combination with the judgment rules of the first-order sensitivity index.

[0123] Specifically, the sampling range of key parameters can refer to the following ranges: dielectric constant of the first layer ∈ [7,20], patch corresponding to the first layer antenna ∈ [27.5,30.5] mm, patch width corresponding to the first layer antenna ∈ [27.5,30.5] mm, feed point distance ∈ [-5,5] mm (the sampling range of other key parameters is set according to the corresponding material and structural characteristics).

[0124] Secondly, based on the dual-layer dual-frequency satellite navigation antenna simulation model built above, finite element simulations were run in batches on the above 500 sets of key parameter samples to obtain the performance degradation parameter response values ​​corresponding to each set of samples, including the specific values ​​of return loss (S11), peak gain (Gain) and maximum axial ratio (AR), to provide data support for the subsequent quantitative calculation of disturbance sensitivity.

[0125] Furthermore, the perturbation sensitivity of each key parameter is quantified by calculating the first-order sensitivity index. The calculation of perturbation sensitivity follows the formula below: ; in, Used to characterize perturbation sensitivity; Used to characterize the response values ​​of performance degradation parameters; Key parameters used to characterize the i-th input; The variance of the conditional expectation is used to reflect the fluctuation range of the performance degradation parameter response value when the key parameter changes; The total variance of the response value of the performance degradation parameter is used to characterize the degree of fluctuation of the response value of the performance degradation parameter, and both are calculated by Monte Carlo integration.

[0126] Finally, a preset perturbation sensitivity (i.e., perturbation sensitivity threshold) is set to 0.3. Based on the first-order sensitivity index judgment rules, key parameters are categorized and filtered to determine target parameters. Specific judgment rules can be found below: When When =0, it is considered right Completely unaffected, considered an irrelevant variable, and therefore removed; when 0 < When the value is less than 0.3, the key parameter is considered to contribute less than 30% to the response variance and is therefore a minor variable, thus being determined as the target parameter; when 0.3 ≤ When <1, the key parameter is considered to contribute more than 30% to the response variance, thus belonging to an important variable and being determined as the target parameter; when When =1, The changes were entirely due to It is determined separately, is a unique variable, and is given priority in being identified as the target parameter.

[0127] For example, the results of the perturbation sensitivity analysis of a key parameter provided in this application are shown in Table 1. The table intuitively presents the perturbation sensitivity, average perturbation sensitivity and screening results of each key parameter under different performance degradation parameters.

[0128] Table 1 Disturbance Sensitivity

[0129] As shown in Table 1, the larger the disturbance sensitivity value, the more significant the impact of the key parameter on antenna performance degradation. The values ​​range from 0 to 1. Specifically, the first and second dielectric constants have the highest disturbance sensitivity to return loss (some values ​​> 0.8), while the patch size has a relatively high disturbance sensitivity to peak gain (some values ​​> 0.5). It should be noted that although the average disturbance sensitivity of the equivalent dielectric constant and feed point distance is less than 0.3, based on practical engineering verification, a ±5% disturbance in these parameters results in a frequency shift of less than 0.5 MHz, far less than the preset engineering tolerance of ±3 MHz. Therefore, the impact on the antenna coupling matrix is ​​negligible. Based on the parameter sensitivity analysis results, the equivalent dielectric constant and feed point distance, being secondary variables, can be eliminated, further simplifying the calibration process for the target parameters.

[0130] In addition, it should be noted that if the values ​​of patch length and patch width are inconsistent and their impact on antenna performance differ, the patch length and patch width need to be subjected to separate sensitivity analyses to avoid deviations in the screening results due to combined analysis and to ensure the accuracy of target parameter screening.

[0131] In some embodiments, a physical dual-layer dual-frequency satellite navigation antenna can be scanned to construct a simulation model of the dual-layer dual-frequency satellite navigation antenna.

[0132] As a feasible implementation method, such as Figure 5 As shown, the simulation model can be constructed in the following way: S501. Based on the scanning information of the dual-layer dual-frequency satellite navigation antenna, an initial simulation model of the dual-layer dual-frequency satellite navigation antenna is constructed.

[0133] As a feasible implementation method, the scanning information includes: the overall shape and size of the dual-layer dual-frequency satellite navigation antenna, the outline and geometric dimensions of the upper and lower radiating patches, the dimensions of the patch convex structure, the shape and thickness of the upper and lower ceramic dielectric substrates, the antenna layer spacing, the position coordinates of the feed point relative to the center origin, as well as the measured and identified information such as the substrate material type and the initial electromagnetic parameters of the dielectric material.

[0134] In an exemplary embodiment, the complete structure and geometric parameters of the physical antenna can be obtained by non-destructive disassembly, high-precision three-dimensional scanning and precise dimensional measurement. Electromagnetic parameters are preset in combination with material properties. Then, geometric modeling, material assignment, boundary condition setting and equivalent modeling of the feed network are completed in finite element simulation software, thereby constructing the initial simulation model of the dual-layer dual-frequency satellite navigation antenna.

[0135] For example, modeling can be performed as follows: Select a dual-layer, dual-frequency satellite navigation antenna from a supplier, with dimensions of 50mm × 50mm × 10mm. Use a 3D scanner to obtain the patch outline, with a scanning accuracy of ±0.02mm.

[0136] A physical sample of a dual-layer, dual-frequency satellite navigation antenna from a supplier was selected. The overall dimensions of the antenna are 50mm × 50mm × 10mm. A 3D scanner was used to perform high-precision scanning and acquisition of the antenna patch outline, with a scanning accuracy of ±0.02mm. Figure 6 As shown, the antenna was first disassembled and its structure observed and measured non-destructively: the upper patch antenna has a rectangular structure, and the measured base length is as follows: =29mm, Width: =29mm; The lower patch antenna is a 45° chamfered rectangular structure, with the actual measured base length as follows: =40mm, Width: =40mm, and at the center of each of the four sides of the rectangular patch, there is a convex rectangular structure of the same specification, with the dimensions of the convex rectangle being 1.029mm in length and 2.058mm in width.

[0137] like Figure 7 As shown, the measured dimensions of the upper ceramic substrate are 35mm×35mm×3.886mm and the dimensions of the lower ceramic substrate are 45mm×45mm×5.826mm; the upper and lower ceramic substrates are placed on the PCB board.

[0138] Based on visual inspection, the ceramic substrate is made of white alumina. The initial relative permittivity is set to 10.0, and the loss tangent is set to 0.002. The interlayer spacing between the upper and lower antennas is measured to be 0.1 ± 0.05 mm using a dial indicator. The antenna adopts a coaxial differential feeding method, and the positions of the two feeding points (upper antenna feeding terminal and lower antenna feeding terminal) relative to the antenna center origin (X=0, Y=0) are (-5, 0) and (0, -5), respectively.

[0139] Secondly, such as Figure 8 As shown, an initial simulation model of the antenna was built in the finite element simulation software: geometric models of the upper ceramic dielectric substrate (35mm×35mm×3.886mm) and the lower ceramic dielectric substrate (45mm×45mm×5.826mm) were established according to the measured dimensions; structural models of the upper patch antenna (29mm×29mm×0.05mm) and the lower patch antenna (40mm×40mm×0.05mm) were also established. The first dielectric constant of the upper ceramic substrate and the second dielectric constant of the lower ceramic substrate were set as variable parameters to be optimized, with the initial value of the first dielectric constant set to 10 and the initial value of the second node constant set to 9.8; the antenna patch material was assigned as copper, and the outer boundary of the model was set as a metal shield.

[0140] Next, the feed network is modeled using an equivalent circuit model composed of lumped-parameter components. The model includes a feed inductor L=68nH, decoupling capacitors C1=100nF and C2=100pF in parallel, a 100pF DC blocking capacitor, an ESD diode, and a TVS diode. The dielectric region surrounding the upper and lower antenna feed terminals is equivalent to a uniform dielectric block, with the equivalent dielectric constant ranging from 1.0 to 3.0, and an initial value of 2.0, used to simulate the non-ideal transition region between the coaxial connector and the ceramic substrate. The equivalent dielectric constant is used to adjust the local electric field distribution of the feed network and optimize the port impedance matching accuracy. This parameter is set as an independent optimization variable, which can be adjusted synchronously with structural parameters such as the substrate dielectric constant during subsequent parameter calibration and iterative optimization.

[0141] Electromagnetic simulations were performed on the initial simulation model. The initial simulation results showed that the upper antenna resonant point was 1.568 GHz, which deviated from the ideal operating frequency by 7 MHz; the lower antenna resonant point was 1.215 GHz, which deviated from the ideal operating frequency by 12 MHz; at the same time, the antenna port return loss S11 was only -8 dB, which was far from meeting the performance requirements for normal antenna operation. It is necessary to correct the values ​​of each key parameter through subsequent parameter calibration and optimization iterations so that the antenna resonant frequency, impedance matching and radiation performance meet the design standards.

[0142] S502. Decouple the first-layer antenna and the second-layer antenna in the initial simulation model to obtain the intermediate simulation model.

[0143] As a feasible implementation method, decoupling is used to isolate inter-layer frequency interference, weaken and isolate the electromagnetic coupling and crosstalk between the upper and lower antenna layers, suppress inter-layer frequency interference, eliminate the problem of mutual pull between the operating frequency bands of the two antenna layers and the superposition of resonant frequency shifts, and facilitate the individual analysis of the influence of the structural parameters of each antenna layer on its respective target frequency.

[0144] In an exemplary embodiment, by adding an electromagnetic isolation structure, setting an electromagnetic absorption boundary, and adjusting the interlayer dielectric arrangement between the first and second antenna layers in the initial simulation model, electromagnetic decoupling of the upper and lower antenna layers is achieved, and an intermediate simulation model is constructed. The decoupled intermediate simulation model can independently characterize the inherent resonant characteristics of a single-layer antenna without being disturbed by the electromagnetic coupling of the other antenna layer.

[0145] For example, an electromagnetic isolation layer is added between the upper and lower ceramic substrates and the radiating patch in the initial simulation model. At the same time, radiation absorption boundary conditions are set in the interlayer region of the model to cut off the electromagnetic wave coupling path in the interlayer space. The original target parameters such as antenna geometry, substrate dielectric constant, and feeding structure are kept unchanged. Decoupling is completed only through interlayer electromagnetic shielding and boundary condition optimization, thereby obtaining an intermediate simulation model without interlayer coupling interference.

[0146] S503. Using the intermediate simulation model and residual structure, a simulation model is obtained.

[0147] The residual structure is used to represent the frequency offset difference of the target frequency offset when the parameter values ​​of the target parameters are the same in the initial simulation model and the intermediate simulation model.

[0148] For example, select each target parameter to be calibrated and apply the same set of small parameter perturbations, and substitute them into the initial simulation model and the intermediate simulation model to carry out electromagnetic simulations; extract the target frequency offsets corresponding to the first layer antenna and the second layer antenna under the initial simulation model and the intermediate simulation model respectively; calculate the offset difference of each target frequency between the initial simulation model and the intermediate simulation model under the same parameter perturbation conditions, and this frequency offset difference is the residual component brought about by interlayer coupling; quantize and fit the frequency offset residuals under multiple sets of parameter perturbations, construct a mapping relationship, and form a residual structure that can characterize the influence of interlayer coupling.

[0149] As a feasible approach, the residual structure is coupled and superimposed onto the intermediate simulation model to compensate for the frequency offset error caused by interlayer electromagnetic coupling, restore the actual electromagnetic response characteristics of the dual-layer antenna under real vehicle installation conditions, and obtain the final simulation model that takes into account both the inherent characteristics of a single layer and the interlayer coupling effect.

[0150] It should be understood that the simulation model not only retains the dominant law of each target parameter on the single-frequency resonance characteristics after decoupling from the intermediate simulation model, but also introduces the frequency offset correction amount brought about by interlayer coupling through the residual structure. Compared with the simple decoupling model or the initial coupling model, the simulation accuracy is higher and closer to the actual working state of the physical antenna. It can provide high-precision model support for the subsequent establishment of parameter perturbation matrix, optimization of objective function and accurate calibration of antenna parameters.

[0151] In some embodiments, the decoupling of the first-layer antenna and the second-layer antenna can also be achieved by changing the metal boundary material of the antenna.

[0152] As a feasible implementation method, step S502 can be specifically implemented as follows: switch the metal boundary of the second-layer antenna or the first-layer antenna to a perfect electrical conductor boundary or an RLC boundary, so as to decouple the first-layer antenna and the second-layer antenna in the initial simulation model and obtain an intermediate simulation model.

[0153] A Perfect Electric Conductor (PEC) boundary is an ideal electromagnetic boundary whose core characteristic is that the tangential component of the electric field and the normal component of the magnetic field on the boundary surface are zero. It can achieve complete reflection and shielding of electromagnetic waves, effectively blocking the electromagnetic energy coupling between two antenna layers. The antenna structure set as a PEC boundary no longer generates a radiation field and does not receive external electromagnetic signals, thus achieving complete isolation from the other antenna layer.

[0154] The RLC boundary, or lumped-parameter resistance-inductor-capacitor (RLC) boundary, is a non-ideal electromagnetic boundary. By rationally designing RLC parameters to construct a boundary structure with specific impedance characteristics, selective suppression and attenuation of electromagnetic waves can be achieved, rather than complete shielding. Its core function is to suppress the radiation of one antenna layer while allowing another antenna layer to establish an effective radiation field distribution, satisfying the requirements for independent parameter optimization and calibration of single-layer antennas, and balancing decoupling effect with simulation effectiveness.

[0155] In the exemplary embodiment, the core principle of boundary switching is: for the level to which the target parameter to be simulated and calibrated belongs, the metal boundary of the other antenna is switched to the corresponding type of decoupling boundary, so that the antenna of the level to be calibrated is in an approximately independent electromagnetic environment, eliminating the coupling interference of the other antenna and ensuring the accuracy of target parameter simulation and calibration.

[0156] It should be understood that decoupling the upper and lower antenna layers by switching the metal boundary type of the second-layer antenna is a flexible and efficient decoupling method. It requires no modification to the initial antenna geometry or material properties; interlayer coupling can be severed simply by adjusting boundary conditions. This simplifies the decoupling process while preserving the antenna's structural characteristics to the greatest extent possible. This ensures that the intermediate simulation model accurately reflects the inherent electromagnetic response of a single-layer antenna, providing a reliable model foundation for subsequent independent calibration of single-layer antenna parameters. Furthermore, the flexible switching between the two boundary types can accommodate the need for separate calibration of the upper and lower antenna layers, resolving coupling interference issues during independent simulation of a single-layer antenna.

[0157] As a feasible implementation method, when the target parameters being simulated are at the first layer antenna, the metal structure setting of the second layer antenna is switched to the boundary of a perfect electrical conductor.

[0158] In an exemplary embodiment, such as Figure 9 As shown, when simulating and calibrating the target parameters of the first-layer antenna (such as the first dielectric constant and the patch size corresponding to the first-layer antenna), the material properties of all metal structures of the second-layer antenna (including the lower patch and feed line) are set to PEC boundaries. At this time, the PEC boundaries create a mirror shielding effect on the electric field generated by the first-layer antenna, effectively blocking the electromagnetic coupling path between the two antenna layers. This results in an interlayer coupling energy attenuation of over 50dB, ensuring that the first-layer antenna is in a completely independent electromagnetic environment, unaffected by interference from the second-layer antenna. Under these conditions, the focus can be on simulating the relationship between the perturbation of the first-layer antenna's target parameters and the corresponding target frequency offset, accurately completing the parameter calibration of the first-layer antenna and laying the foundation for subsequent calibration of the second-layer antenna and overall parameter optimization.

[0159] As another feasible implementation method, such as Figure 10As shown, when the target parameters being simulated are at the second layer antenna, the metal structure setting of the first layer antenna is switched to the lumped parameter resistance-inductance-capacitance impedance boundary (RLC boundary).

[0160] It should be noted that the first-layer antenna is the upper layer antenna of the second-layer antenna. If the metal structure corresponding to the upper-layer antenna is set as the PEC boundary, it will completely shield the lower-layer antenna, making it impossible for the lower-layer antenna to establish an effective radiation field distribution, and thus impossible to carry out independent simulation and optimization of the target parameters of the lower-layer antenna. However, the RLC impedance boundary, through a specific high-impedance design, can suppress the radiation of the upper-layer antenna while providing an approximately independent electromagnetic environment for the lower-layer antenna, ensuring that the lower-layer antenna can work normally and achieve parameter convergence.

[0161] In the exemplary embodiment, the RLC boundary here is designed as a parallel resonant high impedance surface (HIS). Its core design goal is to make the upper antenna exhibit high impedance characteristics at the operating frequency bands (1.176 GHz and 1.575 GHz) of the dual-layer GNSS antenna, thereby significantly attenuating the radiated power of the upper antenna, reducing coupling interference to the lower antenna, and meeting the requirement for independent calibration of the lower antenna.

[0162] For example, the specific parameter selection and verification process of the RLC boundary is as follows: At the two operating frequency bands (1.176GHz and 1.575GHz) of the dual-layer GNSS antenna, the upper antenna is made to exhibit high impedance characteristics, requiring the surface impedance |Zs|>500Ω, which is much greater than the free space wave impedance Z0=377Ω, thereby attenuating the radiated power of the upper antenna by more than 20dB compared with its actual material state, ensuring that the coupling disturbance of the upper antenna to the lower antenna is controlled within a very small range.

[0163] The RLC boundary is constructed using a parallel LC resonance model, and the resonant frequency is selected. =1.4GHz (this frequency lies between two operating frequencies, ensuring high impedance characteristics in both bands), and based on the resonant frequency formula, inductor L = 20nH and capacitor C = 0.645pF are selected. The formula for calculating the parallel resonant frequency is: ; At the resonant frequency f=f0, the RLC parallel circuit exhibits ideal high impedance characteristics, and the surface impedance |Zs|→∞.

[0164] When the operating frequency deviates from the resonant frequency (f≠f0), the surface impedance calculation formula is: ; Calculations have verified that at the lower antenna operating frequency band f1=1.176GHz, the surface impedance |Zs|≈502Ω>377Ω; at f2=1.575GHz, the surface impedance |Zs|≈761Ω>377Ω, both of which meet the preset high impedance design target and can effectively attenuate the radiated power of the upper antenna.

[0165] By simulating and monitoring the change in the resonant frequency of the lower antenna, it was confirmed that when the upper antenna is in the RLC boundary state, the frequency offset of the lower antenna compared to when the upper antenna is in the real material state is <0.5MHz, which is much smaller than the preset engineering tolerance of ±3MHz. This verifies that the weak coupling assumption is valid and ensures that the lower antenna can complete parameter optimization and calibration in a nearly independent electromagnetic environment.

[0166] To further explain, through the aforementioned RLC boundary design, the dielectric constant of the lower antenna obtained during the optimization process is 9.78. This can be considered approximately equivalent to the independent optimal value of the lower antenna under a completely isolated reference state where "the upper antenna is set to real material + the lower antenna is set to PEC boundary". This equivalence is the core physical basis for using the "PEC isolated state" as the frequency reference during subsequent pre-compensation calculations, ensuring the accuracy of the pre-compensation calculations and providing a guarantee for the optimal calibration of the final target parameters.

[0167] In some embodiments, after constructing a simulation model, the perturbation matrix is ​​constructed by simulating the correspondence between the parameter value perturbation and the target frequency offset through the simulation model.

[0168] As a feasible implementation method, such as Figure 11 As shown, step S302 above can be specifically implemented as follows: S1101. Simulate the first and second perturbation matrices using a simulation model.

[0169] As a feasible implementation method, the first perturbation matrix is ​​used to represent the correspondence between the parameter perturbation of the target parameters and the target frequency offset, under the premise of isolating the inter-layer frequency interference between the first and second layer antennas. The core purpose is to quantify the independent influence of the small perturbations of each target parameter on the antenna target frequency under decoupled conditions, eliminate the superposition effect of inter-layer coupling interference, and provide a basic quantitative reference for subsequent parameter calibration and pre-compensation. It is the core foundation for constructing the coupling matrix and realizing linear compensation.

[0170] In the exemplary embodiment, the parameter perturbation is used to characterize the small offset of the target parameter to be calibrated relative to its reference value. It is set in the form of a relative perturbation, and the value range is usually ±1% to ±5%. This ensures that the perturbation response is identifiable and that the target frequency response is within the linear range. It avoids nonlinear distortion due to excessive perturbation amplitude or excessive simulation error due to insufficient perturbation amplitude, and provides a prerequisite for subsequent calculation of matrix elements using the central difference method. The reference value of the target parameter can be selected from the initial value measured by the actual object, the typical parameter value of the material, or the initial calibration value after previous optimization, which fits the actual structure and material characteristics of the antenna. The target parameters to be calibrated mainly include five sets of core parameters: the first dielectric constant, the second dielectric constant, the patch size corresponding to the first layer antenna, the patch size corresponding to the second layer antenna, and the interlayer spacing.

[0171] For example, taking the first dielectric constant as the target parameter, its reference value is set to 9.65 (based on actual measurement and initial simulation optimization), and the parameter value perturbation is set to ±5%. Then, the parameter value corresponding to the positive perturbation is 9.65×1.05=10.13, and the parameter value corresponding to the negative perturbation is 9.65×0.95=9.17. Similarly, if the target parameter is the patch size of the first layer antenna (reference value 29mm), the parameter values ​​corresponding to the ±5% perturbation are 30.45mm and 27.55mm, respectively. Through this small perturbation, the independent influence trend of parameter changes on the target frequency can be accurately captured, and other target parameters remain unchanged during the perturbation process to ensure the uniqueness of the influence law.

[0172] It should be understood that the parameter perturbation amounts in the first perturbation matrix need to be set and perturbed individually for each target parameter. During the perturbation process, boundary switching (such as setting non-target layer antennas as PEC boundaries) is required to isolate inter-layer frequency interference, ensuring that the perturbation effect of each parameter can be quantified individually. This avoids the superposition of effects caused by simultaneous perturbation of multiple parameters, thereby ensuring the uniqueness and accuracy of the correspondence between the parameter perturbation amounts represented by the first perturbation matrix and the target frequency offset. This provides reliable support for subsequent independent analysis of the impact of each parameter on single-frequency characteristics and calculation of coupling matrix elements. Simultaneously, the rationality of the micro-perturbation step size needs to be verified experimentally to ensure the linearity of the frequency response, thus guaranteeing the accuracy of matrix calculations.

[0173] In the exemplary embodiment, the target frequency offset is used to characterize the difference between the actual operating frequency (first target frequency, second target frequency) of the antenna and the corresponding ideal operating frequency after the target parameters are perturbed. It is the core indicator for quantifying the degree of influence of the target parameter perturbation on the antenna frequency characteristics. It can be divided into positive offset (actual frequency is higher than ideal frequency) and negative offset (actual frequency is lower than ideal frequency), and the unit is usually GHz or MHz. In the first perturbation matrix, the target frequency offset is only caused by the perturbation of the target parameters themselves and does not include the additional offset caused by interlayer coupling.

[0174] For example, taking the ideal operating frequency of the first-layer antenna as 1.575 GHz and the ideal operating frequency of the second-layer antenna as 1.176 GHz, when a +5% perturbation (first dielectric constant = 10.13) is applied to the first dielectric constant and inter-layer interference is isolated, the simulation shows that the actual operating frequency (first target frequency) of the first-layer antenna is 1.571 GHz, and the target frequency offset is 1.571 GHz - 1.575 GHz = -0.004 GHz (i.e., -4 MHz). When a -5% perturbation (first dielectric constant = 9.17) is applied, the actual operating frequency is 1.579 GHz, and the target frequency offset is +0.004 GHz (i.e., +4 MHz). At the same time, since inter-layer interference has been isolated, this parameter perturbation will not cause the target frequency of the second-layer antenna to shift, intuitively reflecting the correspondence between the parameter perturbation and the frequency offset under decoupling conditions.

[0175] It should be understood that the magnitude of the target frequency offset directly reflects the degree of influence of the target parameter disturbance on the antenna resonance characteristics. The larger the offset, the higher the sensitivity of the parameter to the antenna frequency characteristics; conversely, the smaller the offset, the weaker the influence of the parameter on the frequency characteristics. By quantifying this correspondence, the first disturbance matrix can clearly present the independent influence weight of each target parameter on the first and second target frequencies.

[0176] As a feasible implementation method, the second perturbation matrix is ​​used to represent the correspondence between the parameter value perturbation of the target parameters and the inter-layer frequency interference. The core function of the second perturbation matrix is ​​to quantify the change in inter-layer electromagnetic coupling strength caused by small perturbations of each target parameter in the undecoupled state, thereby characterizing the strength of inter-layer frequency interference. This provides support for subsequent compensation of frequency offset errors caused by inter-layer coupling and improvement of calibration accuracy. It complements the first perturbation matrix, together covering the parameter influence laws under both decoupled and undecoupled operating conditions.

[0177] In the exemplary embodiment, the definition and setting principles of the parameter value perturbation are completely consistent with those of the first perturbation matrix. This perturbation is used to characterize the small offset of the target parameter to be calibrated relative to its reference value, and also adopts a relative perturbation form of ±1% to ±5%, ensuring that the parameter perturbation standards of the two matrices are unified, facilitating subsequent data comparison, fusion, and the construction of the coupling matrix. The target parameters here are consistent with the first perturbation matrix, including five sets of core parameters: the first dielectric constant, the patch size corresponding to the first layer antenna, the second dielectric constant, the patch size corresponding to the second layer antenna, and the interlayer spacing. Furthermore, the perturbation amplitude is consistent with that of the first perturbation matrix (e.g., both are ±5%).

[0178] For example, still using the first dielectric constant as the target parameter, its baseline value is set to 9.65 (based on actual antenna measurements and initial simulation optimization). Calculated according to a relative perturbation of ±5%, the parameter value corresponding to the positive perturbation is 9.65 × 1.05 = 10.13, and the parameter value corresponding to the negative perturbation is 9.65 × 0.95 = 9.17. Unlike the parameter perturbation of the first perturbation matrix, the parameter perturbation process of the second perturbation matrix is ​​carried out in the undecoupled initial simulation model. Instead of isolating inter-layer interference through boundary switching (such as PEC boundaries, RLC boundaries), it completely simulates the inter-layer coupling environment under the actual working state of the antenna, thereby accurately capturing the inter-layer frequency interference changes caused by the target parameter perturbation. This ensures that the second perturbation matrix can truly and accurately reflect the influence law of inter-layer coupling, providing reliable data support for subsequent quantification of the additional frequency offset caused by inter-layer coupling and construction of the coupling matrix.

[0179] It should be understood that the parameter perturbation amounts in the second perturbation matrix are completely consistent with those in the first perturbation matrix. The core purpose is to ensure the comparability and correlation between the two matrices, enabling subsequent precise separation of the independent impact of the target parameter's own perturbation on the antenna frequency (characterized by the first perturbation matrix) and the additional impact of inter-layer coupling (characterized by the second perturbation matrix) through difference calculation. This effectively avoids the superposition of errors caused by inconsistent perturbation standards, ensuring the accuracy of subsequent parameter calibration and coupling compensation. Simultaneously, the principle of individual perturbation of target parameters also applies to the second perturbation matrix. That is, a perturbation amount is applied individually to each target parameter, while keeping other target parameters unchanged during the perturbation process. This ensures that the quantization result of inter-layer frequency interference is caused only by the perturbation of a single target parameter, avoiding the superposition of effects caused by simultaneous perturbation of multiple parameters, and ensuring the uniqueness and accuracy of the matrix data.

[0180] In the exemplary embodiment, inter-layer frequency interference is used to characterize the additional offset of the target frequency of the other antenna caused by the change in electromagnetic coupling strength between the two antenna layers due to the perturbation of the target parameters. It is the core indicator for quantifying the inter-layer coupling effect. Its magnitude is positively correlated with the amount of parameter perturbation and the sensitivity of the parameters to coupling. The unit is consistent with the target frequency offset (GHz or MHz). This additional offset is the core reason for the difference between the actual frequency offset of the antenna and the frequency offset in the decoupled state. It is also the key source of error that needs to be corrected in subsequent pre-compensation.

[0181] It should be noted that by using the second perturbation matrix for perturbation compensation, the nonlinear coupling problem between the two-layer antennas can be transformed into a linear problem for processing. Subsequently, the collaborative pre-compensation of the two-layer parameters can be achieved through matrix inversion, which effectively avoids the oscillation phenomenon that occurs during the parameter iterative optimization process, improves the convergence of parameter optimization from local optima to global convergence, and significantly reduces the number of iterations, improving the efficiency and accuracy of antenna parameter calibration, laying the foundation for subsequent antenna performance optimization.

[0182] It should be understood that the core significance of quantifying inter-layer frequency interference lies in providing a basis for error correction for subsequent pre-compensation. The second perturbation matrix quantifies this interference through the system, which can clearly identify the degree of influence of each target parameter on inter-layer coupling, provide cross-coupling coefficients for the construction of the coupling matrix, and provide a reference for subsequent optimization of inter-layer isolation design and further suppression of coupling interference, so as to ensure the stable dual-frequency operation performance of the antenna after final calibration.

[0183] S1102. Determine the perturbation matrix based on the first perturbation matrix and the second perturbation matrix.

[0184] As a feasible implementation method, the parameter-frequency independent mapping relationship represented by the first perturbation matrix in the decoupled state is fused with the interlayer frequency interference coupling relationship represented by the second perturbation matrix in the undecoupled state. The difference analysis and linear fitting are then performed to comprehensively remove the inherent frequency offset contribution of the target parameter itself and the additional frequency offset contribution of interlayer electromagnetic coupling, and an overall perturbation matrix that can simultaneously take into account the inherent characteristics of a single layer and the coupling effect of a double layer is obtained.

[0185] In the exemplary embodiment, the first perturbation matrix is ​​used as the reference matrix to characterize the independent influence components of the perturbation of each target parameter on the target frequency of the two antenna layers under the condition of complete decoupling and no inter-layer interference. Then, the inter-layer frequency interference components under the actual working condition are introduced by the second perturbation matrix. The additional frequency offset caused purely by inter-layer coupling is separated by the subtraction operation of each matrix element. Then, the independent influence components and the coupled additional components are linearly superimposed and uniformly incorporated into the same matrix dimension to form the final overall perturbation matrix.

[0186] It should be understood that relying solely on the first perturbation matrix can only describe the ideal decoupling condition and cannot reflect the interlayer electromagnetic crosstalk present in a real antenna; relying solely on the second perturbation matrix can only reflect the total frequency offset including coupling and cannot separate the inherent influence of the parameters themselves from the influence of coupling interference. This application constructs an overall perturbation matrix by fusing the first and second perturbation matrices, which can achieve the decoupling, separation, and unified representation of independent frequency offset components and interlayer coupled frequency offset components; at the same time, based on the previous micro-perturbation linearity verification results, it ensures that the perturbation matrix is ​​effective throughout the linear interval, providing an accurate and complete matrix model foundation for subsequent matrix inversion, solving for the co-correction of the double-layer dielectric constant, and parameter pre-compensation iterative convergence.

[0187] In some embodiments, when the target parameter is the dielectric constant, the first perturbation matrix and the second perturbation matrix are determined based on the first layer antenna independent optimal dielectric constant and the second layer antenna approximately independent optimal dielectric constant obtained by hierarchical virtual decoupling calibration, combined with the interlayer decoupling rules of PEC boundary and RLC boundary, through pre-selected parameter value perturbation, multi-stage boundary gradual transition and frequency drift quantization processing.

[0188] As a feasible implementation method, such as Figure 12 As shown, step S1101 above can be specifically implemented as follows: S1201. For each of the preselected parameter value perturbations in at least one preselected parameter value perturbation of the target parameter, the preselected parameter value perturbation is input into the initial simulation model and the intermediate simulation model respectively, so as to obtain the target frequency offset output by the intermediate simulation model response and the frequency offset difference output by the residual structure response.

[0189] The pre-selected parameter value perturbation is used to characterize the preset change amount of the target parameter deviating from its baseline value. It is an input variable used to trigger the frequency response of the initial simulation model and the intermediate simulation model, and then obtain the residual structure output (frequency difference between the two models). Its value needs to be reasonably set in combination with the engineering application range and inter-layer coupling characteristics of the target parameter to ensure that the additional frequency offset caused by inter-layer coupling can be effectively quantified.

[0190] The intermediate simulation model achieves interlayer decoupling between the first-layer antenna and the second-layer antenna through boundary switching (PEC boundary or RLC boundary). The initial simulation model is in an undecoupled state and is used to simulate the interlayer coupling effect under real working conditions.

[0191] As a feasible implementation method, the process of determining the target frequency offset of the intermediate simulation model response output includes: setting the intermediate simulation model to a decoupled state by switching the boundary (PEC boundary or RLC boundary) to eliminate the interlayer coupling interference between the first-layer antenna and the second-layer antenna; then inputting the perturbation amount of the pre-selected parameter value into the decoupled model, running the simulation and recording the target frequency data output by the model, and comparing it with the model reference frequency to obtain the target frequency offset of the intermediate simulation model response output. This offset only reflects the inherent influence of the perturbation of the target parameters themselves and does not include any additional errors caused by interlayer coupling.

[0192] As a feasible implementation method, the frequency offset difference in the output of the residual structure response is essentially the difference in the target frequency offset between the initial simulation model and the intermediate simulation model under the same pre-selected parameter value perturbation. It is used to quantify the additional frequency offset error caused by interlayer coupling, provide the core basis for subsequent pre-compensation, and is also a key data source for constructing the second perturbation matrix and solving the dielectric constant correction.

[0193] In the exemplary embodiment, the frequency offset difference in the residual structural response output is used to characterize the difference in target frequency offset between the initial simulation model (undecoupled, including interlayer coupling) and the intermediate simulation model (decoupled, without interlayer coupling) under the same pre-selected parameter value perturbation. This difference is entirely caused by the interlayer electromagnetic coupling effect and does not include the inherent frequency offset component caused by the perturbation of the target parameters themselves. Its core function is to quantify the additional influence of interlayer coupling on the target frequency offset, providing an error correction basis for subsequent construction of the coupling matrix and realization of two-layer parameter collaborative pre-compensation; at the same time, the frequency offset difference can verify the rationality of the perturbation step size, ensure the linearity of the frequency response, and guarantee the accuracy of the coupling matrix calculation.

[0194] To accurately obtain the frequency offset difference and ensure numerical stability, this application employs a three-stage approach to complete the boundary gradual transition and pre-compensation operation, simultaneously monitoring frequency changes to quantify the frequency offset difference corresponding to the residual structure. The specific stage division can be referenced in the following steps (the upper boundary is always the real material, consistent with the coupling matrix construction conditions, ensuring that the frequency drift measurement reflects the real interlayer coupling effect; the main operation line is the gradual transition of the lower boundary from PEC to the real material, relying on the bidirectional characteristics of the interlayer coupling matrix to simultaneously solve the double-layer dielectric constant correction): like Figure 13As shown, in Stage 1 (PEC isolation state, corresponding to the intermediate simulation model working condition): the second-layer antenna is set as the PEC boundary to achieve complete isolation between the first-layer antenna and the second-layer antenna. At this time, the model is in a decoupled state. The perturbation amount of the pre-selected parameter values ​​is input into the model to obtain the target frequency offset output by the intermediate simulation model response (only reflecting the inherent influence of the perturbation of the target parameters themselves, without inter-layer coupling interference). At the same time, this state is used as the reference state for pre-compensation (defined as the "upper layer real material + lower layer PEC" completely isolated state). The first target frequency and the second target frequency are recorded at this time. This reference state is not the "second-layer antenna independent working state" (this state is not achievable), but provides a calculable electromagnetic reference point, laying the foundation for subsequent frequency drift calculation and frequency offset difference quantification.

[0195] like Figure 14 As shown, Stage 2 (RLC transition state, transitional undecoupled condition): The second-layer antenna PEC boundary is changed to an RLC boundary, introducing weak coupling. At this time, the model is in a weakly coupled undecoupled state. The same pre-selected parameter value perturbation is input into the model, and the simulation is run to evaluate the frequency drift. , ,in = - (Initial change in the first target frequency) = - (Initial change in the second target frequency) This frequency drift is used to verify the linearity of the frequency response and does not participate in the core calculation of the subsequent frequency offset difference. Here, the RLC boundary is designed as a high-impedance surface (|Zs|>500Ω), which suppresses the radiation of the first-layer antenna by >20dB. It is only used for transitional verification and does not represent the real physical state. Its core function is to construct a mapping chain with monotonically increasing coupling strength, which facilitates linearization analysis.

[0196] like Figure 15 As shown, Stage 3 (real boundary state, corresponding to the initial simulation model operating condition): The RLC boundary of the second-layer antenna is restored to the real material of the lower layer. At this time, the model is in a completely undecoupled state (simulating the real operating condition). The same pre-selected parameter value perturbation is input into the model, and the total frequency drift is measured. , ,in = - (First target frequency offset) = - (Second target frequency offset). The total frequency drift is the difference between the target frequency offset output by the initial simulation model response (including the inherent frequency offset of the target parameters themselves and the additional frequency offset from interlayer coupling) and the target frequency offset output by the intermediate simulation model in stage 1 (only the inherent frequency offset). This difference is the frequency offset gap output by the residual structure response. This gap includes the mixed effects of "the lower layer being activated from the PEC shielding state to the real radiation state" and "interlayer electromagnetic coupling", which is uniformly processed by the perturbation matrix through linear approximation.

[0197] During the three-stage boundary asymptotic transition process, the upper boundary is always maintained as the real material, and the RLC weak radiation boundary from the independent calibration of the lower layer is not adopted. The reason is that the pre-compensation stage needs to quantify the interlayer coupling effect under the real working condition and complete the parameter correction. The RLC boundary is only an optimization auxiliary tool for independent calibration and does not correspond to the real physical condition. If the upper layer still maintains the RLC boundary, the pre-compensation can only correct the incremental deviation from the RLC to the real state, and cannot characterize the overall frequency offset effect from complete isolation to real coupling, which will result in incomplete compensation and poor numerical stability.

[0198] It can be seen that, as Figure 16 As shown in the experimental results, when jumping directly from the PEC stage to the true boundary without compensation, the sudden change in coupling strength will cause the error in the calculation of parameter correction to exceed 40% (e.g., the total frequency offsets for each of the 15 iterations are 13.3, 15, 28.4, 22.1, 18.5, 25.3, 19.8, 16.2, 21.5, 14.8, 20.1, 17.6, 15.3, 12.9, and 5.8, respectively). The optimization is prone to getting trapped in local optima, and the frequency offset is still too large after iterations or even oscillates and diverges. This application adds an intermediate transition state of RLC and establishes a monotonic coupling strength mapping relationship to linearize the frequency drift change. After adding matrix pre-compensation, it can reach stability in only 3 iterations (e.g., the first two are 0.8 and 0.3, and the subsequent ones are 0.05), without divergence and oscillation, ensuring the numerical stability of matrix inversion.

[0199] It should be understood that this application sets the pre-compensation reference state as the real material of the first-layer antenna + the PEC of the second-layer antenna. This is because the independent calibration of the second-layer antenna can only use the RLC high-impedance boundary of the first-layer antenna, avoiding the PEC completely shielding the second-layer antenna. Although the dielectric constant is optimized, there will still be residual coupling error, which needs to be uniformly corrected by pre-compensation. At the same time, this reference can be used as a calculable electromagnetic reference point. The frequency drift can be fully covered by the comprehensive influence of the boundary state switching of the second-layer antenna and the interlayer coupling. The parameters of the two layers can be synchronously corrected by using the linear approximation of the coupling matrix. After pre-compensation, the frequency offset and the number of iterations meet the engineering tolerance. The problem of integrated antenna parameter calibration is solved by realizing engineering approximation. At the same time, the first-layer antenna must always maintain the real material boundary during the pre-compensation process. If the RLC boundary is used, only local incremental correction can be achieved, and the global coupling effect cannot be characterized. However, through the three-stage gradual transition from PEC zero coupling, RLC near zero coupling to real full coupling, the additional frequency offset caused by interlayer coupling can be accurately quantified, ensuring the quantization accuracy and numerical stability of pre-compensation.

[0200] Optionally, further linearization verification of the transition frequency drift is required, and a linearity threshold should be preset for criterion verification. The specific steps are as follows: Determine frequency drift: RLC transition stage frequency drift: =-1.2MHz, =-0.8MHz; Total frequency drift in the true boundary phase: =-5.6MHz, =-5.2MHz.

[0201] Define linearity evaluation index ( ): ; The calculated linearity η = 1.0 and the deviation is 0%, which is less than the set threshold η_thr = 10%. This verifies that the frequency response has good linearity within the ±5% perturbation range. The use of the central difference method to solve the coupling matrix elements is engineering reasonable, thus ensuring that the frequency offset difference of the residual structure response output can be accurately used for subsequent matrix construction and parameter correction.

[0202] If the linearity η exceeds the linearity threshold, it indicates that the perturbation step size is too large and the nonlinear effect is prominent. The relative perturbation needs to be reduced from ±5% to ±2% or ±1%, and the simulation and frequency offset difference quantization should be carried out again until the linearity requirement is met, so as to ensure the reliability of the residual structure response output frequency offset difference.

[0203] S1202. Using the parameter value perturbation and the target frequency offset output by the intermediate simulation model response, the first perturbation matrix is ​​generated.

[0204] The first perturbation matrix is ​​used to measure the frequency offset of the second-layer antenna from the PEC boundary to the RLC boundary.

[0205] As a feasible implementation method, by combining the first perturbation matrix with the decoupling characteristics of the intermediate simulation model, the frequency response quantization relationship of the first-layer antenna and the second-layer antenna under the isolation condition can be established, the performance degradation quantization expression of the first-layer antenna can be obtained, and then the reference frequency offset characteristics and the optimal reference value of dielectric constant of the first-layer antenna under the uncoupled reference can be determined.

[0206] S1203. Using the parameter value perturbation and the frequency offset difference of the residual structure response output, a second perturbation matrix is ​​generated.

[0207] As a feasible implementation method, the perturbation of each pre-selected parameter value is used as input. The frequency offset difference of the residual structural response output obtained by comparing the initial simulation model (undecoupled, simulating the real coupling condition) and the intermediate simulation model (decoupled, without interlayer coupling) is used to first quantify the interlayer frequency interference component under different parameter perturbations. This frequency offset difference is entirely caused by interlayer electromagnetic coupling and does not contain the inherent frequency offset of the parameter perturbation itself, which can accurately reflect the additional influence of interlayer coupling. Then, based on the quantized interlayer frequency interference component and the target frequency offset of each layer, a second perturbation matrix that can comprehensively characterize the interlayer coupling characteristics is obtained by fitting and solving.

[0208] For example, before constructing the second perturbation matrix, it is necessary to verify the rationality of the perturbation step size selection to ensure that the target parameters are within the set perturbation range, and that there is a good linear relationship between the parameter value perturbation, the frequency difference between the initial and intermediate models under the same perturbation (residual structure output), the inter-layer frequency interference, and the target frequency offset. This provides a premise for the subsequent calculation of the elements of the second perturbation matrix using the central difference method, and at the same time, it intuitively characterizes the changing law of the inter-layer coupling effect.

[0209] The verification process is as follows: Figure 17 As shown, with the first dielectric constant To verify the object, a pre-selected parameter perturbation (such as ±5% micro-perturbation) is applied to it, and three key data points are obtained: the baseline point (without perturbation), =9.65, second target frequency =1.176GHz); +5% perturbation point (positive perturbation, =10.13, =1.178GHz); -5% perturbation point (negative perturbation, =9.17, =1.174GHz). Connecting the three key data points above forms a broken line. The slope of this broken line is the secant slope, which can intuitively reflect the overall change law of the frequency offset of the second-layer antenna (including the effect of interlayer coupling) after the first dielectric constant perturbation is combined with the frequency offset difference of the residual structure. At the same time, a tangent is drawn at the reference point (marked by a dashed line in the figure), and the tangent slope is calculated. The slope of the tangent line and the slope of the secant line ((1.178-1.174) / (10.13-9.17)=0.004167) highly coincide, which verifies that within the range of ±5% perturbation, the perturbation of the target parameter and the frequency response (including the influence of interlayer coupling) have a good linear relationship, thus proving the engineering rationality of using a ±5% perturbation step size and calculating the elements of the second perturbation matrix by the central difference method.

[0210] Optionally, if the deviation between the line connecting the three points and the tangent is too large (i.e., nonlinearity > 10%), it indicates that the current perturbation step size exceeds the linear range of the frequency response, which will cause the calculation deviation of the second perturbation matrix elements. In this case, the perturbation step size needs to be reduced to ±2% or ±1%, and the parameter perturbation and verification should be carried out again until the linearity requirement is met, so as to ensure the calculation accuracy of the second perturbation matrix.

[0211] After the perturbation step size verification was passed, based on the dual-layer parameter locking, relative perturbations of δ=±5% were applied to the first and second dielectric constants respectively. For each set of parameter perturbation conditions, the initial simulation model and intermediate simulation model were substituted into the model to solve for the frequency offset difference of the corresponding residual structure response output. Simultaneously, the target frequency offset and inter-layer coupling coefficient of each antenna layer were extracted, and the second perturbation matrix was constructed based on this. The specific form of the second perturbation matrix is ​​as follows: ; in, The change in the first target frequency difference between the initial simulation model and the intermediate simulation model under the same disturbance amount is the self-coupling coefficient when the first dielectric constant changes by one unit value. The change in the first target frequency difference between the initial simulation model and the intermediate simulation model under the same disturbance amount (cross-coupling coefficient) is the amount of change in the first target frequency difference between the initial simulation model and the intermediate simulation model for each unit change in the second dielectric constant. The frequency difference output by the residual structure directly characterizes the interlayer coupling effect. The change in the second target frequency difference between the initial simulation model and the intermediate simulation model under the same disturbance amount is the cross-coupling coefficient when the first dielectric constant changes by a unit value. The frequency difference output by the residual structure directly characterizes the interlayer coupling effect. The change in the second target frequency difference between the initial simulation model and the intermediate simulation model under the same disturbance amount is the self-coupling coefficient when the second dielectric constant changes by one unit value.

[0212] All matrix elements in the second perturbation matrix ( All calculations are performed using the central difference method combined with the residual structure frequency offset difference to ensure the accuracy and reliability of the results: ; in, The difference in target frequency between the i-th layer antenna in the initial simulation model and the intermediate simulation model under the same perturbation (i=1 corresponds to the first layer antenna, whose ideal operating frequency is 1.575GHz; i=2 corresponds to the second layer antenna, whose ideal operating frequency is 1.176GHz). Let be the dielectric constant corresponding to the j-th layer antenna (j=1 corresponds to the first dielectric constant, j=2 corresponds to the second dielectric constant); The absolute disturbance of the target parameter is calculated as follows: =0.05 (i.e., 5% relative disturbance).

[0213] It should be understood that the calculation process uses multiple sets of parameter perturbations and the corresponding frequency difference between the two models under the same perturbation as the data source to ensure that the matrix elements can truly map the correlation between parameter perturbations and the frequency difference between the initial and intermediate models, as well as the interlayer coupling.

[0214] In some embodiments, a reference value (e.g., 11) for the first dielectric constant and a reference value (e.g., 9.78, which is the independent optimal value under the weak radiation state of RLC and is equivalent to the optimal value under the isolated state of PEC through engineering approximation) are determined by the first perturbation matrix and the objective function. Then, the first reference dielectric constant and the second reference dielectric constant are pre-compensated and corrected by the second perturbation matrix (characterizing the frequency difference between the initial and intermediate models under the same perturbation amount) to finally determine the optimal calibration parameter value of the dielectric constant.

[0215] The first perturbation matrix and objective function are used to determine the reference value of dielectric constant when there is no interlayer coupling interference, ensuring the rationality of the reference parameter itself; while the second perturbation matrix is ​​used to quantify the additional frequency offset error caused by interlayer coupling. By pre-compensating and correcting the reference value, the negative impact of interlayer coupling is offset, and finally the optimal calibration value that fits the actual working condition is obtained.

[0216] As a feasible implementation method, the frequency offset difference (residual structure output) between the initial simulation model and the intermediate simulation model under the same perturbation is the core basis for simultaneously solving the dielectric constant correction of the upper and lower layers. The specific solution process and principle are as follows: The frequency offset difference (residual structure output) is essentially the frequency offset difference between the initial simulation model (including interlayer coupling) and the intermediate simulation model (without interlayer coupling) under the same preselected parameter value perturbation. It is used to quantify the additional frequency offset error caused by interlayer coupling. Combining this frequency difference with the total frequency drift data, a system of linear equations is constructed, and the dielectric constant correction of the upper and lower layers is solved simultaneously.

[0217] In the exemplary embodiment, by combining the frequency difference (coupled additional frequency offset) obtained by the above quantization with the total frequency drift data (actual frequency offset under real operating conditions), a linear relationship between the "dielectric constant correction amount" and the "total frequency drift" can be established, and then a system of linear equations can be constructed. By solving the system of equations simultaneously, the correction amount of the dielectric constant of the upper and lower layers can be obtained, thereby achieving accurate compensation for the reference value.

[0218] The application logic of frequency offset difference is: to , Simultaneously, as input, the output, after inverse matrix operations, consists of corrected values ​​for the first and second dielectric constants. This simultaneous solution process ensures that, considering interlayer bidirectional coupling, the parameters of both layers are synergistically optimized, avoiding the accumulation of cross-coupling errors caused by independent corrections.

[0219] Based on experimental simulation data, the specific values ​​and physical meanings of each element of the constructed second perturbation matrix (coupling matrix C, representing the frequency difference between the initial and intermediate models under the same perturbation amount) are as follows: ; The specific meanings of each matrix element (each corresponding to the frequency difference between the initial and intermediate models under the same perturbation): =5.6364: For every unit increase in the upper layer dielectric constant, the difference between the upper layer target frequency of the initial and intermediate models increases by 5.6364 MHz (self-coupling coefficient) under the same perturbation. =3.3742: For every unit increase in the lower layer dielectric constant, the difference between the upper layer target frequency of the initial and intermediate models increases by 3.3742 MHz (cross-coupling coefficient) under the same perturbation. =3.2727: For every unit increase in the upper layer dielectric constant, the difference between the lower layer target frequency of the initial and intermediate models increases by 3.2727 MHz (cross-coupling coefficient) under the same perturbation. =6.3395: For every unit increase in the lower layer dielectric constant, the difference between the target lower layer frequency of the initial and intermediate models increases by 6.3395 MHz (self-coupling coefficient) under the same perturbation.

[0220] Further clarifying the construction logic of the linear equation system: the elements of the second perturbation matrix are essentially "the coupling-induced frequency offset change corresponding to a unit perturbation of the dielectric constant" (i.e., the first-order partial derivative), and the dielectric constant correction vector is "the parameter adjustment amount required to offset the coupling-induced frequency offset." Their product precisely corresponds to "the total frequency offset that the correction amount can offset." This total frequency offset must be equal in magnitude and opposite in direction to the total frequency drift vector under actual operating conditions to achieve frequency offset compensation. Therefore, through... This relationship can transform the "frequency offset compensation requirement" into the "dielectric constant correction problem", providing theoretical support for subsequent matrix inversion operations.

[0221] In constructing a system of linear equations In this case, the dielectric constant correction can be obtained by solving the linear equations through matrix inversion. The specific steps are as follows: Calculate the determinant: det(C) = (5.6364 × 6.3395) - (3.2727 × 3.3742) = 24.6892; Find the adjoint matrix: ; Calculate the inverse matrix: ; Solve for the dielectric constant correction: ; After obtaining the correction amount, the corresponding correction amount is subtracted from the previously determined reference value of the dielectric constant (a negative correction amount is equivalent to a slight reduction in the reference value). This ensures that the corrected dielectric constant can offset the additional frequency offset caused by interlayer coupling, enabling the dual-layer antenna to reach the ideal resonant frequency under actual coupling conditions. Finally, the optimal calibration value of the first dielectric constant is obtained: =11-0.7268=10.2732; Optimal calibration value of the second dielectric constant: =9.78-0.4408=9.3392.

[0222] It should be understood that the second perturbation matrix is ​​full, and the actual operating frequency offset of the first target simultaneously affects both the first and second dielectric constants. Similarly, the actual operating frequency offset of the second target also simultaneously affects both dielectric constants. Therefore, even if the main operating principle is "second-layer antenna boundary transition," the first dielectric constant must be corrected synchronously; otherwise, the interlayer coupling effect cannot be accurately compensated. Furthermore, the second dielectric constant of 9.78 is the independent optimal value in the RLC weak radiation state. Through engineering approximation, it is equivalent to the optimal value in the PEC isolated state. Pre-compensation corrects it to 9.3392, with the core purpose of offsetting the interlayer coupling effect "from the isolated state to the real coupled state," enabling the dual-layer antenna to simultaneously reach the target resonant frequency under real material conditions.

[0223] By adopting this boundary progressive transition combined with pre-compensation correction strategy, the frequency offset of the upper antenna can be compensated from -5.6MHz to -1.2MHz, with a compensation efficiency of 77%. This effectively avoids parameter oscillations caused by boundary switching, ensures the stability and accuracy of the optimal calibration value of the double-layer dielectric constant, and ultimately optimizes the performance of the double-layer antenna to meet the positioning accuracy requirements of the vehicle-mounted GNSS receiver.

[0224] In some embodiments, when the target parameter is the patch size, the element values ​​in the second perturbation matrix are all 0 by default; in this case, there is no need to construct the second perturbation matrix, and the simulation can be directly performed based on the simulation model to obtain the first perturbation matrix.

[0225] As a feasible implementation method, such as Figure 18 As shown, the first perturbation matrix can be constructed in the following way: S1801. For each of the preselected parameter value perturbations in at least one preselected parameter value perturbation of the target parameter, input the preselected parameter value perturbation into the intermediate simulation model to obtain the target frequency offset output by the intermediate simulation model in response to the preselected parameter value perturbation.

[0226] In the exemplary embodiment, the intermediate simulation model is a simulation model for the decoupled, non-interlayer coupling condition. It only introduces the perturbation of the pre-selected parameter value corresponding to the patch size. Under the premise of eliminating interlayer coupling interference, it separately characterizes the inherent correspondence between the perturbation of the patch size parameter itself and the target frequency offset.

[0227] S1802. The first perturbation matrix is ​​generated by using the parameter value perturbation amount and the target frequency offset output by the pre-selected parameter value perturbation amount of the intermediate simulation model response.

[0228] In the exemplary embodiment, the first perturbation matrix is ​​used to quantitatively characterize the mapping relationship between a single perturbation of the patch size parameter and the first-order partial derivative of the corresponding target frequency offset. Since the interlayer coupling effect has been eliminated and the second perturbation matrix is ​​set to zero, the first perturbation matrix can independently characterize the frequency response characteristics of the patch size parameter. Subsequently, parameter calibration, frequency offset compensation and structural size optimization can be carried out directly based on the first perturbation matrix, saving the complex calculation process of solving the two-layer coupling matrix simultaneously.

[0229] It should be understood that the patch size is a structural geometric parameter, which only has an inherent offset effect on the frequency of this layer and will not generate electromagnetic cross-coupling disturbances between layers. There is no mutual interference frequency offset contribution between different layers. Therefore, all elements in the second perturbation matrix can be set to 0 by default. There is no need to solve for the residual frequency difference through the initial simulation model and the intermediate simulation model. The parameter perturbation characteristics can be characterized by relying solely on the decoupled intermediate simulation model.

[0230] In some embodiments, such as Figure 19As shown, the process of constructing the objective function includes: S1901. The first optimization objective is to minimize the weighted sum of return loss, peak gain, and maximum axial ratio at the ideal operating frequency of the first-layer antenna.

[0231] The first optimization objective is to achieve the optimal overall performance of the first-layer antenna by quantifying the comprehensive degradation of the three key performance indicators mentioned above, thus laying a stable foundation for subsequent lower-layer parameter optimization and ultimately ensuring the positioning accuracy, receiving sensitivity, and multipath suppression capability of the vehicle-mounted GNSS receiver. All optimization steps revolve around "minimizing the weighted sum value" to ensure that each operation serves this core objective.

[0232] As a feasible implementation method, the target parameters to be calibrated for the first-layer antenna are clearly defined, and the parameter value range and first perturbation matrix for each target parameter are defined, while other irrelevant parameters are fixed; a performance degradation quantification expression is defined, and the weighting coefficients are determined and their rationality is verified by combining the analytic hierarchy process (AHP) with the receiver sensitivity budget; convergence criteria and initial target parameter values ​​are set; the particle swarm optimization algorithm is used for iterative optimization until the convergence requirements are met, the target parameter values ​​for the first-layer antenna are locked, and the performance is verified.

[0233] It should be understood that return loss determines the matching effect between the antenna and the RF link, directly affecting positioning accuracy; peak gain is related to receiver sensitivity, affecting signal reception capability; and axial ratio is related to multipath suppression capability, adapting to complex automotive scenarios. By quantifying the overall performance degradation of these three factors through weighted summation, the importance differences of each indicator can be taken into account, achieving optimal overall performance of the upper-layer antenna and laying a stable foundation for subsequent lower-layer parameter optimization.

[0234] In an exemplary embodiment, the parameter values ​​of the target parameters of the first layer antenna, determined based on the first optimization objective, can be determined in the following way: the first dielectric constant and the patch size corresponding to the first layer antenna are defined, and the value range of each target parameter is delineated. At the same time, other irrelevant parameters are fixed to define reasonable boundaries for subsequent optimization calculations. Irrelevant parameters such as the second dielectric constant, the patch size corresponding to the second layer antenna, and the interlayer spacing are fixed, and adjustments are made only for the first dielectric constant and the patch size corresponding to the first layer antenna to avoid irrelevant parameters interfering with the optimization accuracy and improve optimization efficiency. Simultaneously, a first perturbation matrix is ​​generated by combining the intermediate simulation model under decoupling conditions. The first perturbation matrix is ​​used to quantitatively characterize the mapping relationship between the single perturbation amount of the first dielectric constant and the patch size corresponding to the first layer antenna and the corresponding target frequency offset, providing data support for subsequent parameter optimization. For example, the value range of the target parameters can be the first dielectric constant ∈ [7.0, 20.0] and the patch size ∈ [27.5, 30.5].

[0235] A performance degradation quantification objective function is defined, and weight coefficients are determined by combining the analytic hierarchy process (AHP) with the sensitivity budget of the vehicle-mounted GNSS receiver. The rationality of the weight configuration is verified. The upper-level parameters are optimized using the particle swarm optimization algorithm, and the performance degradation quantification expression (i.e., the objective function) is defined as follows: Fitness1=w1·| |+w2·Gain1+w3·AR1; Among them, return loss Determine the basic matching between the antenna and the RF link, based on the RF link budget. For every 1dB deterioration, the system noise figure NF increases by 0.5dB, directly leading to a 0.5dB decrease in the carrier-to-noise ratio (C / N0), which has the greatest impact on positioning accuracy. Therefore, it is assigned the highest weight w1=0.5. Peak gain affects receiver sensitivity. For every 1dB decrease in peak gain, it is equivalent to a 1dB decrease in signal power, and C / N0 decreases by 1dB simultaneously. However, this impact can be partially compensated by the receiver's internal AGC (Automatic Gain Control), so its impact is secondary, and it is assigned a weight w2=0.3. Maximum axial ratio affects multipath suppression capability. For every 1dB increase in maximum axial ratio, the positioning error increases by about 0.5m in multipath environments such as urban canyons, but the impact is small in open sky scenarios. Therefore, it has the lowest weight, and is assigned a weight w3=0.2. To verify the rationality of the weight coefficients, 1000 sets of random parameters were generated through random sampling. The convergence success rate of optimization under each weight configuration was calculated. The verification results show that when w1∈[0.45,0.55], w2∈[0.25,0.35], and w3∈[0.15,0.25], the convergence success rate is >85%. Therefore, w1=0.5, w2=0.3, and w3=0.2 are determined as the optimal weight center points, balancing convergence efficiency and optimization accuracy.

[0236] Set the convergence criterion, initial optimization variable vector, and simulation operation parameters to provide clear standards and operational basis for iterative optimization: The convergence criterion is set to | - | < 3MHz and S11 < -15dB, where, | - | Used to characterize the deviation between the actual operating frequency and the rational operating frequency of the first-layer antenna; since the phase-locked loop capture bandwidth of GNSS receivers is usually ±5MHz, setting a 3MHz tolerance can reserve a 2MHz safety margin to avoid the frequency offset exceeding the phase-locked loop capture range, which would cause the receiver to be unable to lock the signal normally.

[0237] The simulation operation of the above process can be performed by referring to the following steps: ① In the simulation software, select the optimization module (Optimetrics) → SetupOptimization to launch the optimization settings interface.

[0238] ② Enter the following formula in the objective function (CostFunction) field: Cost=0.5×abs(dB(S(1,1)))+0.3×GainTotal+0.2×AxialRatio, which is consistent with the performance degradation quantification expression defined above, to achieve the quantitative calculation of the degree of performance degradation.

[0239] ③ Set specific performance requirements in the Optimization Goals column: S11 < -15dB (weight 1.0, prioritizing impedance matching), GainTotal > 3.5dBi (weight 0.6, taking gain performance into account), AxialRatio < 3dB (weight 0.4, meeting polarization purity requirements).

[0240] ④ In the Variables section, check the first dielectric constant and the patch size corresponding to the first layer antenna to ensure that the parameter values ​​are optimized only for the preset target parameters of the first layer antenna.

[0241] Finally, the particle swarm optimization algorithm was used for iterative optimization until the convergence requirement was met. The optimal parameters for the upper layer were then locked, and the performance was verified. Key parameters for the particle swarm optimization algorithm were set: the population size was 30, and the maximum number of iterations was 15. This configuration ensures optimization accuracy while avoiding inefficiency caused by excessive iterations. After 8 iterations, the optimization process reached the preset convergence criterion. At this point, all target parameters for the upper layer were locked, yielding the optimal parameters: the first dielectric constant was 11, and the patch size corresponding to the first-layer antenna was 29mm. Performance verification of these optimal parameters showed that at the ideal operating frequency of 1.575GHz, the first-layer antenna achieved a return loss S11 of -19.2dB, a peak gain of 4.1dBi, and an axial ratio of 2.3dB. All performance indicators met the requirements for use with the vehicle-mounted GNSS receiver, and the optimization effect was satisfactory.

[0242] S1902. The second optimization objective is to minimize the weighted sum of return loss, peak gain, and maximum axial ratio at the ideal operating frequency of the second-layer antenna.

[0243] The core logic of the second optimization objective is the same as that of the first optimization objective. Both are to achieve the optimal overall performance of the corresponding antenna layer by quantifying the comprehensive degradation of the three key performance indicators. The difference is that the second optimization objective focuses on the second layer antenna. Its ideal operating frequency, optimization parameters and suitable scenarios are all in line with the design requirements of the second layer antenna. At the same time, it is necessary to take into account the inter-layer synergy with the already optimized first layer antenna, so as to lay the foundation for the overall performance of the dual-layer antenna to meet the standards. All optimization operations are carried out around the core objective of "minimizing the weighted sum value".

[0244] As a feasible implementation method, to efficiently and accurately achieve the second optimization objective while ensuring the consistency of the dual-layer antenna optimization strategy and reducing optimization complexity, the same global optimization strategy as the first-layer antenna is adopted. The specific implementation process is as follows: The performance degradation quantification objective function and weight configuration are kept completely consistent with the first layer, ensuring that the performance evaluation system of the two antenna layers is unified and comparable, avoiding deviations in the collaborative performance of the two layers due to differences in evaluation standards; the optimized and locked target parameters (first dielectric constant, first-layer antenna patch size) and interlayer spacing parameters of the first-layer antenna are fixed to prevent disturbances to the already achieved upper-layer parameters, ensuring the stability of the upper-layer performance. The optimization process eliminates interference from irrelevant parameters on lower-level optimization. It focuses on optimizing the core parameters of the second-layer antenna—the second dielectric constant, the patch length, and the patch width—within a specific range, confining optimization to improve lower-level performance and increase efficiency. Referencing the frequency tolerance and impedance matching indices of the first-layer antenna, it sets convergence criteria consistent with those of the first layer, clearly defining the standard for minimizing the weighted sum. Through iterative optimization using a global optimization algorithm, it continues until the convergence criteria are met, ultimately determining the optimal parameters for the second-layer antenna. A comprehensive performance check of the optimal parameters is then performed to ensure that the parameters meet the overall usage requirements of the vehicle-mounted GNSS receiver.

[0245] In the exemplary embodiment, a performance degradation quantification expression for the second-layer antenna is constructed; based on the performance degradation quantification expression and the first interference matrix, the parameter values ​​of the target parameters of the second-layer antenna are determined when the second optimization objective is satisfied.

[0246] For example, the performance degradation quantification expression for the second-layer antenna can satisfy the following formula: Fitness2=w1·| |+w2·Gain2+w3·AR2; in, Gain2 and AR2 correspond to the return loss, peak gain, and maximum axial ratio at the ideal operating frequency of the second-layer antenna, and are used to quantify the degree of overall performance degradation of the second-layer antenna.

[0247] Furthermore, a first perturbation matrix is ​​generated by combining the intermediate simulation model under decoupling conditions. This matrix quantitatively represents the mapping relationship between the first-order partial derivatives of the second dielectric constant, the single perturbation of the second-layer antenna patch size, and the corresponding target frequency offset, providing data support for parameter optimization. Based on this, the optimized first-layer parameters and interlayer spacing are fixed, and the second-layer optimization variables are defined (second dielectric constant ∈ [7.0, 20.0], patch length and width ∈ [39.5, 42.0]), and the convergence criterion is set as | - With |<3MHz and S11<-15dB, a genetic algorithm was used for iterative optimization. After 12 generations of iterations and convergence, the optimal target parameters for the second-layer antenna were determined to be: second dielectric constant = 9.78 and patch size = 40mm. At this point, the performance degradation quantization value was minimized, and all performance parameters of the second-layer antenna met the design requirements, thus successfully achieving the second optimization objective.

[0248] Among them, | - | Used to characterize the deviation between the actual operating frequency and the rational operating frequency of the second-layer antenna.

[0249] S1903. Minimize the isolation penalty as the third optimization objective.

[0250] It should be understood that while the first and second optimization objectives have achieved optimal performance for the first and second antenna layers respectively, inter-layer electromagnetic coupling exists during dual-layer antenna operation. Insufficient inter-layer isolation can lead to coupling interference, causing resonant frequency shifts and performance degradation in both antenna layers. This, in turn, affects the overall operational stability of the dual-layer antenna and the positioning accuracy of the vehicle-mounted GNSS receiver. The isolation penalty quantifies the severity of inter-layer coupling interference; a larger penalty indicates stronger inter-layer coupling and poorer isolation, resulting in more significant damage to the overall performance of the dual-layer antenna. Therefore, minimizing the isolation penalty is the third optimization objective. Its core purpose is to suppress inter-layer coupling interference, compensate for the shortcomings of the individual optimizations in the first two stages, achieve optimal overall performance of the dual-layer antenna, and ensure that the overall usage requirements of the vehicle-mounted GNSS receiver are met.

[0251] As a feasible implementation method, the isolation penalty is determined based on the interlayer isolation. The optimal parameters of the dual-layer antenna obtained by optimizing the first and second optimization objectives are used as the basis. The interlayer isolation is optimized by fine-tuning the interlayer spacing, and the convergence judgment criteria are set to control coupling interference. After the convergence requirements are met, full parameter fine-tuning is performed, taking into account the performance of each layer and the interlayer isolation, so as to achieve the overall optimality of the dual-layer antenna and achieve the goal of minimizing the isolation penalty.

[0252] In the exemplary embodiment, based on the optimal parameters locked in the first two stages (first layer: first dielectric constant = 11, patch size = 29mm; second layer: second dielectric constant = 9.78, patch size = 40mm), the interlayer spacing d is finely adjusted with interlayer isolation as the core; the convergence criterion is set as the resonant frequency offset of both antenna layers < 2MHz and the interlayer coupling coefficient k < 0.05.

[0253] The coupling coefficient can be calculated using the following formula: ; Among them, in the formula for calculating the coupling coefficient, Used to characterize the transmission coefficient between two antenna layers, reflecting the strength of interlayer coupling. The larger the absolute value, the stronger the interlayer coupling and the greater the isolation penalty; Used to characterize the return loss of each antenna layer and reflect the performance stability of a single layer. The smaller the absolute value, the better the performance of a single layer, which can reduce interference with inter-layer coupling.

[0254] It should be understood that after fine-tuning to ensure the interlayer spacing meets the convergence requirements, all parameters of the dual-layer antenna are fine-tuned to ensure that both layers of the antenna meet the constraints at their respective ideal operating frequencies (e.g., ...). With a peak gain of <-15dB, peak gain >3.5dBi, and maximum axial ratio <3dB, the interlayer isolation penalty is minimized, the overall performance of the dual-layer antenna is optimal, and the third optimization objective is successfully achieved.

[0255] S1904. Construct the objective function using the first, second, and third optimization objectives.

[0256] As a feasible implementation method, the objective function must comply with the constraints.

[0257] The constraints are used to limit the range of return loss, peak gain, and maximum axial ratio at the ideal operating frequency of the first-layer antenna, as well as the range of return loss, peak gain, and maximum axial ratio at the ideal operating frequency of the second-layer antenna.

[0258] It should be understood that in the process of optimizing the overall objective function, it is necessary to avoid excessive optimization of a single optimization objective, which could lead to the deterioration of other performance indicators. By setting upper and lower bound constraints on the key electrical performance indicators of the two antenna layers, the parameter optimization is limited to a reasonable range that is engineering-feasible, device-feasible, and system-adaptable. This ensures that the basic electromagnetic performance of a single antenna layer meets the standards in the first and second optimization objectives, and also defines a feasible region for minimizing the inter-layer isolation penalty in the third optimization objective. This makes the optimal solution of the overall objective function both physically feasible and practical for the engineering of vehicle-mounted GNSS systems.

[0259] As a feasible implementation method, the weighted summation terms of the first-layer antenna performance, the weighted summation terms of the second-layer antenna performance, and the inter-layer isolation penalty term are weighted and fused to construct a global comprehensive objective function. The return loss, peak gain, and axial ratio of the two-layer antennas are preset threshold values ​​as inequality constraints, and the frequency offset and inter-layer coupling coefficient are used as coupling constraints. A global optimization algorithm is used to synchronously iterate and find the optimal solution within the multi-constraint boundary, taking into account both the independent performance of the single-layer antenna and the inter-layer isolation characteristics of the two layers. The solution yields the globally optimal structural parameters that take into account the three optimization objectives.

[0260] In an exemplary embodiment, such as Figure 20 As shown, step S1904 above can be specifically implemented as follows: S2001. Determine the first dynamic weight corresponding to the first optimization objective and the second dynamic weight corresponding to the second optimization objective.

[0261] The sum of the first dynamic weight and the second dynamic weight is 1.

[0262] Specifically, the initial first dynamic weight is set to w. 10 =0.5, the second dynamic weight is w 20 =0.5, achieving equal initial optimization weight allocation for the two antenna layers; define the first frequency difference. = - (i.e., the deviation between the current simulated operating frequency and the ideal operating frequency of the upper-layer antenna), the second frequency difference. = - (i.e., the deviation between the current simulated operating frequency and the ideal operating frequency of the lower-layer antenna); set the maximum allowable frequency offset thresholds for the first-layer and second-layer antennas. All frequencies are 3MHz, and the maximum allowable frequency offset threshold is combined with the GNSS receiver's phase-locked loop capture bandwidth setting to reserve sufficient safety margin.

[0263] In the exemplary embodiment, the first dynamic weight and the second dynamic weight satisfy the following condition: the first frequency difference is greater than the maximum allowable frequency offset threshold of the first layer antenna (i.e., | When the frequency difference is greater than 3MHz, the first dynamic weighting exponential decay occurs; when the second frequency difference is greater than the maximum allowable frequency offset threshold of the second-layer antenna (i.e., |>3MHz), ... In the case of |>3MHz), the second dynamic weighted exponent decays. For example, the first frequency difference| |=5.6MHz, second frequency difference| |=5.2MHz, all exceeding the preset frequency offset threshold, with the first frequency difference error being larger. The algorithm adaptively increases the optimization weight of the first-layer antenna, adaptively adjusting the first dynamic weight and the second dynamic weight to w. 10 =0.65, w 20 =0.35, ensuring that optimized resources are tilted towards levels with larger frequency offset errors, thereby improving convergence efficiency.

[0264] It should be understood that the dynamic weight adaptive adjustment process is implemented in a closed loop through finite element simulation software. After each iteration, the software automatically reads the actual frequency offset values ​​of the upper and lower antenna layers and updates the first and second dynamic weight parameters in real time according to the preset exponential formula to ensure the real-time performance and accuracy of the weight adjustment.

[0265] S2002. Construct an objective function based on the first optimization objective, the second optimization objective, the third optimization objective, the first dynamic weight, and the second dynamic weight.

[0266] Specifically, when the first frequency difference is greater than the maximum allowable frequency offset threshold of the first layer antenna, the first dynamic weighting exponent decays.

[0267] When the second frequency difference is greater than the maximum allowable frequency offset threshold of the second layer antenna, the second dynamic weighting exponential decays.

[0268] As a feasible implementation method, the quantitative representation of the three optimization objectives is clearly defined. The first optimization objective (first-layer antenna performance) and the second optimization objective (second-layer antenna performance) are quantized as the absolute values ​​of the return loss of the first-layer antenna and the return loss of the second-layer antenna, respectively. The third optimization objective (minimizing the isolation penalty) is quantized as the absolute value of the inter-layer coupling frequency offset. Secondly, a first dynamic weight and a second dynamic weight (the sum of the two is 1) are introduced to weight the quantization terms of the first-layer antenna and the second-layer antenna performance, respectively. At the same time, a coupling weight coefficient is introduced to weight the quantization term of the isolation penalty. Finally, the three are integrated to construct an objective function. The weight attenuation rule is embedded synchronously, and the first dynamic weight attenuates exponentially when the first frequency difference exceeds the maximum allowable frequency offset threshold of the upper layer, and the second dynamic weight attenuates exponentially when the second frequency difference exceeds the maximum allowable frequency offset threshold of the lower layer, forming an adaptively adjustable dynamic objective function.

[0269] In an exemplary embodiment, the objective function can satisfy the following formula: Fitness _total =w 10 ·|S 11_Upper |+w 20 ·|S 11_Lower |+α·|Δ f _coupling |; Among them, |S 11_Upper | represents the absolute value of the return loss of the first-layer antenna, corresponding to the performance degradation index of the upper-layer antenna in the first optimization objective; |S 11_Lower | represents the absolute value of the return loss of the second-layer antenna, corresponding to the performance degradation index of the lower-layer antenna in the second optimization objective; α is the inter-layer coupling frequency offset weighting coefficient, |Δ f _coupling | represents the absolute value of frequency offset caused by interlayer coupling, corresponding to the quantitative index of isolation penalty in the third optimization objective, thus achieving the synergistic integration of the three optimization objectives.

[0270] After the objective function is constructed, iterative optimization is started, with the gradient descent step size set to 0.02. In each iteration, the dynamic weights and objective function values ​​are updated synchronously. After 12 iterations, the overall optimization process reaches final convergence.

[0271] The final parameter calibration results are as follows: Upper antenna: resonant frequency f1 = 1.5738 GHz (frequency deviation -1.25 MHz), return loss S11 = -18.5 dB, peak gain 4.05 dBi, maximum axial ratio 2.2 dB; Lower antenna: resonant frequency f2 = 1.1748 GHz (frequency deviation -1.2 MHz), return loss S11 = -17.1 dB, peak gain 3.75 dBi, maximum axial ratio 2.0 dB; interlayer isolation at 1.575 GHz S21 = 32 dB, which meets the requirement of S21 > 30 dB in engineering design.

[0272] The final convergence result described above was obtained through objective function-driven, adaptive, and collaborative fine-tuning after correcting the first and second dielectric constants. The frequency offset of the optimized dual-layer dual-frequency satellite navigation antenna is less than 3MHz, fully verifying the effectiveness of the joint pre-compensation strategy and the dynamic weight optimization strategy. Simultaneously, it achieves the synergistic attainment of the first, second, and third optimization objectives, ensuring that the overall performance of the dual-layer antenna meets the engineering requirements for dual-layer dual-frequency satellite navigation antenna receivers.

[0273] In some embodiments, after determining the optimal calibration parameter value of the target parameter, it is necessary to complete the simulation model correction and vehicle adaptation optimization based on the optimal parameter to ensure that the dual-layer dual-frequency satellite navigation antenna and the vehicle system are adapted in a coordinated manner and to give full play to the optimal performance of the antenna.

[0274] As a feasible approach, the optimal calibration parameter values ​​of the target parameters are determined, the initial simulation model is corrected, and an optimized simulation model of the dual-layer dual-frequency satellite navigation antenna is obtained. Based on the optimized simulation model, the vehicle is optimized.

[0275] In the exemplary embodiment, the final calibrated optimal calibration parameter values ​​are substituted one by one into the initial finite element simulation model to correct key parameters such as dielectric constant, patch size, and interlayer spacing in the model. At the same time, the electromagnetic coupling parameters in the model are calibrated to complete the correction of the initial simulation model and obtain the optimized simulation model of the dual-layer dual-frequency satellite navigation antenna.

[0276] In the exemplary embodiment, considering the installation location requirements of the dual-layer dual-frequency satellite navigation antenna receiver, the optimal installation area for the dual-layer dual-frequency satellite navigation antenna on the vehicle is determined (prioritizing locations with unobstructed roofs and far away from strong electromagnetic interference sources such as vehicle radar and power amplifiers); the structural design of the vehicle roof mounting surface is optimized, and an installation interface matching the size of the dual-layer dual-frequency satellite navigation antenna is reserved to ensure that the dual-layer dual-frequency satellite navigation antenna is firmly installed and closely fitted to the vehicle body, reducing electromagnetic reflection interference caused by installation gaps; at the same time, the wiring design of the vehicle radio frequency link is optimized to shorten the transmission distance between the antenna and the receiver, reduce signal transmission loss, and ensure that the optimal performance of the optimized simulation model can be realized in actual vehicle scenarios.

[0277] It should be understood that after vehicle optimization design, the modified optimized simulation model and the vehicle 3D model were jointly simulated and verified. The results showed that in the vehicle installation scenario, the resonant frequency deviation of the two antenna layers was still less than 3MHz. The return loss, gain, axial ratio and interlayer isolation all met the design requirements. The positioning accuracy error in the complex vehicle environment was controlled within the preset range. The coordinated adaptation of the dual-layer dual-frequency satellite navigation antenna and the vehicle system was realized, verifying the feasibility and effectiveness of this implementation method.

[0278] Please see Figure 21 , Figure 21 This is a schematic diagram of the structure of an electronic device disclosed in an embodiment of this application. The electronic device 2100 may include a processor 2101 and a memory 2102. The processor 2101 and the memory 2102 are communicatively connected. The memory 2102 is used to store programs, and the processor 2101 is used to execute the programs, specifically the relevant steps described in the embodiment of the parameter calibration method for a dual-layer dual-frequency satellite navigation antenna.

[0279] Specifically, the program may include program code, which includes computer-executable instructions. Memory 2102 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device. Processor 2101 may be a central processing unit (CPU), a microcontroller unit (MCU), or an application-specific integrated circuit (ASIC).

[0280] This application also provides a computer-readable storage medium storing at least one executable instruction. When the executable instruction is executed on the parameter calibration device of the dual-layer dual-frequency satellite navigation antenna, the parameter calibration device of the dual-layer dual-frequency satellite navigation antenna performs the parameter calibration method of the dual-layer dual-frequency satellite navigation antenna in any of the above method embodiments.

[0281] This application provides a computer program product that can be executed by the processor 2101 of the electronic device 2100 to complete the parameter calibration method for the dual-layer dual-frequency satellite navigation antenna in the above embodiments.

[0282] It should be understood that the application of this application is not limited to the examples above. Those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims. Those skilled in the art can understand that implementing all or part of the processes of the above embodiments and making equivalent changes according to the claims of this application still fall within the scope of this application.

Claims

1. A parameter calibration method for a dual-layer dual-frequency satellite navigation antenna, characterized in that, The parameter calibration method for the dual-layer dual-frequency satellite navigation antenna includes: Determine the target parameters to be calibrated for the dual-layer dual-frequency satellite navigation antenna; wherein, perturbation of the parameter values ​​of the target parameters will cause a shift in the first target frequency and / or the second target frequency; the first target frequency is the actual operating frequency of the first layer antenna in the dual-layer dual-frequency satellite navigation antenna; the second target frequency is the actual operating frequency of the second layer antenna in the dual-layer dual-frequency satellite navigation antenna. Using the simulation model of the dual-layer dual-frequency satellite navigation antenna, the perturbation matrix of the target parameters is simulated; wherein, the perturbation matrix is ​​used to represent the correspondence between the parameter value perturbation amount of the target parameters and the target frequency offset; the target frequency offset is the frequency offset of the first target frequency and / or the second target frequency; Based on the perturbation matrix and the objective function, the optimal calibration parameter values ​​of the target parameters are determined; wherein, the objective function is used to minimize the performance degradation parameter at the ideal operating frequency of the first layer antenna and the performance degradation parameter at the ideal operating frequency of the second layer antenna; the performance degradation parameter at the ideal operating frequency of the first layer antenna and the performance degradation parameter at the ideal operating frequency of the second layer antenna change with the change of the first target frequency and / or the second target frequency.

2. The parameter calibration method for a dual-layer dual-frequency satellite navigation antenna according to claim 1, characterized in that, The simulation model using the dual-layer dual-frequency satellite navigation antenna, simulating the perturbation matrix of the target parameters, includes: The simulation model is used to simulate a first perturbation matrix and a second perturbation matrix; wherein, the first perturbation matrix is ​​used to represent the correspondence between the parameter value perturbation of the target parameter and the target frequency offset under the premise of isolating the inter-layer frequency interference between the first layer antenna and the second layer antenna; the second perturbation matrix is ​​used to represent the correspondence between the parameter value perturbation of the target parameter and the inter-layer frequency interference; The perturbation matrix is ​​determined based on the first perturbation matrix and the second perturbation matrix.

3. The parameter calibration method for a dual-layer dual-frequency satellite navigation antenna according to claim 2, characterized in that, The process of constructing the simulation model; Based on the scanning information of the dual-layer dual-frequency satellite navigation antenna, an initial simulation model of the dual-layer dual-frequency satellite navigation antenna is constructed. The first-layer antenna and the second-layer antenna in the initial simulation model are decoupled to obtain an intermediate simulation model; the decoupling is used to isolate the inter-layer frequency interference. The simulation model is obtained using the intermediate simulation model and the residual structure; wherein, the residual structure is used to represent the frequency offset difference of the target frequency offset when the parameter values ​​of the target parameters are the same in the initial simulation model and the intermediate simulation model.

4. The parameter calibration method for a dual-layer dual-frequency satellite navigation antenna according to claim 3, characterized in that, The target parameters include the first dielectric constant corresponding to the first layer antenna and the second dielectric constant corresponding to the second layer antenna; The simulation of the first perturbation matrix and the second perturbation matrix using the simulation model includes: For each of the preselected parameter value perturbations in at least one preselected parameter value perturbation of the target parameter, the preselected parameter value perturbation is input into the initial simulation model and the intermediate simulation model respectively, so as to obtain the target frequency offset output by the intermediate simulation model response and the frequency offset difference output by the residual structure response; The first perturbation matrix is ​​generated by using the parameter value perturbation amount and the target frequency offset output by the intermediate simulation model response; The second perturbation matrix is ​​generated using the perturbation amount of the parameter value and the frequency offset difference output by the residual structure response.

5. The parameter calibration method for a dual-layer dual-frequency satellite navigation antenna according to claim 3, characterized in that, The target parameters include: the patch size corresponding to the first layer antenna, the patch size corresponding to the second layer antenna, and the interlayer spacing between the first layer antenna and the second layer antenna; the element values ​​in the second perturbation matrix are all 0 by default; Using the aforementioned simulation model, the first perturbation matrix is ​​simulated, including: For each of the preselected parameter value perturbations in at least one preselected parameter value perturbation of the target parameter, the preselected parameter value perturbation is input into the intermediate simulation model to obtain the target frequency offset output by the intermediate simulation model in response to the preselected parameter value perturbation. The first perturbation matrix is ​​generated by using the parameter value perturbation amount and the target frequency offset output by the intermediate simulation model in response to the preselected parameter value perturbation amount.

6. The parameter calibration method for a dual-layer dual-frequency satellite navigation antenna according to any one of claims 1-5, characterized in that, The objective function is also used to minimize the isolation penalty; Wherein, the isolation penalty is the product of the penalty coefficient and the total frequency offset of the two layers; the penalty coefficient is used to suppress the deterioration of the isolation effect of the inter-layer frequency interference; the total frequency offset of the two layers is the sum of the first frequency difference and the second frequency difference; the first frequency difference is the frequency difference between the first target frequency and the ideal operating frequency of the first layer antenna; the second frequency difference is the frequency difference between the second target frequency and the ideal operating frequency of the second layer antenna.

7. The parameter calibration method for a dual-layer dual-frequency satellite navigation antenna according to claim 6, characterized in that, The performance degradation parameters include at least one of the following: return loss, peak gain, and maximum axial ratio.

8. The parameter calibration method for a dual-layer dual-frequency satellite navigation antenna according to claim 7, characterized in that, The performance degradation parameters include the return loss, the peak gain, and the maximum axial ratio; The process of constructing the objective function includes: The first optimization objective is to minimize the weighted sum of the return loss, peak gain, and maximum axial ratio at the ideal operating frequency of the first-layer antenna. The second optimization objective is to minimize the weighted sum of the return loss, peak gain, and maximum axial ratio at the ideal operating frequency of the second-layer antenna. Minimizing the isolation penalty is the third optimization objective; The objective function is constructed using the first optimization objective, the second optimization objective, and the third optimization objective.

9. The parameter calibration method for a dual-layer dual-frequency satellite navigation antenna according to claim 8, characterized in that, The step of constructing the objective function using the first optimization objective, the second optimization objective, and the third optimization objective includes: Determine the first dynamic weight corresponding to the first optimization objective and the second dynamic weight corresponding to the second optimization objective; the sum of the first dynamic weight and the second dynamic weight is 1; The objective function is constructed based on the first optimization objective, the second optimization objective, the third optimization objective, the first dynamic weight, and the second dynamic weight; Wherein, when the first frequency difference is greater than the maximum allowable frequency offset threshold of the first layer antenna, the first dynamic weighting exponential decays; When the second frequency difference is greater than the maximum allowable frequency offset threshold of the second layer antenna, the second dynamic weighting exponential decays.

10. The parameter calibration method for a dual-layer dual-frequency satellite navigation antenna according to claim 8, characterized in that, The objective function complies with the constraints. The constraints are used to constrain the range of return loss, peak gain and maximum axial ratio at the ideal operating frequency of the first-layer antenna, and the range of return loss, peak gain and maximum axial ratio at the ideal operating frequency of the second-layer antenna.

11. The parameter calibration method for a dual-layer dual-frequency satellite navigation antenna according to claim 7, characterized in that, The determination of the target parameters to be calibrated for the dual-layer dual-frequency satellite navigation antenna includes: Determine the key parameters of the dual-layer dual-frequency satellite navigation antenna; wherein, the key parameters are those whose parameter values, when disturbed, will cause a shift in the first target frequency and / or the second target frequency; Select the parameter whose perturbation sensitivity is greater than the preset perturbation sensitivity from the key parameters, and use it as the target parameter; The disturbance sensitivity of the key parameter is determined based on the fluctuation of the performance degradation parameters of the first-layer antenna at the ideal operating frequency and the performance degradation parameters of the second-layer antenna at the ideal operating frequency under the disturbance of the parameter value of the key parameter; the fluctuation includes the fluctuation amplitude and the degree of fluctuation.

12. The parameter calibration method for a dual-layer dual-frequency satellite navigation antenna according to claim 7, characterized in that, The process of decoupling the first-layer antenna and the second-layer antenna in the initial simulation model to obtain the intermediate simulation model includes: The metal boundary of the second-layer antenna or the first-layer antenna is switched to a perfect electrical conductor boundary or a lumped parameter resistance-inductance-capacitance impedance boundary to decouple the first-layer antenna and the second-layer antenna in the initial simulation model, thereby obtaining the intermediate simulation model. When the target parameters being simulated are at the first layer antenna, the metal structure of the second layer antenna is switched to the boundary of the perfect electrical conductor. When the target parameters being simulated are at the second layer antenna, the metal structure of the first layer antenna is switched to the lumped parameter resistance-inductance-capacitance impedance boundary. The first layer antenna is the upper layer antenna of the second layer antenna.

13. The parameter calibration method for a dual-layer dual-frequency satellite navigation antenna according to claim 3, characterized in that, The method further includes: After determining the optimal calibration parameter values ​​of the target parameters, the initial simulation model is corrected to obtain the optimized simulation model of the dual-layer dual-frequency satellite navigation antenna; Based on the optimized simulation model, the vehicle is then optimized in design.

14. A parameter calibration device for a dual-layer dual-frequency satellite navigation antenna, characterized in that, The parameter calibration device for the dual-layer dual-frequency satellite navigation antenna includes: a first determination module, a simulation module, and a second determination module; The first determining module is used to determine the target parameters to be calibrated for the dual-layer dual-frequency satellite navigation antenna; wherein, perturbation of the parameter values ​​of the target parameters will cause a shift in the first target frequency and / or the second target frequency; the first target frequency is the actual operating frequency of the first layer antenna in the dual-layer dual-frequency satellite navigation antenna; the second target frequency is the actual operating frequency of the second layer antenna in the dual-layer dual-frequency satellite navigation antenna. The simulation module is used to simulate the perturbation matrix of the target parameters using the simulation model of the dual-layer dual-frequency satellite navigation antenna; wherein, the perturbation matrix is ​​used to represent the correspondence between the parameter value perturbation amount of the target parameters and the target frequency offset; the target frequency offset is the frequency offset of the first target frequency and / or the second target frequency; The second determining module is used to determine the optimal calibration parameter value of the target parameter based on the perturbation matrix and the objective function; wherein, the objective function is used to minimize the performance degradation parameter at the ideal operating frequency of the first layer antenna and the performance degradation parameter at the ideal operating frequency of the second layer antenna; the performance degradation parameter at the ideal operating frequency of the first layer antenna and the performance degradation parameter at the ideal operating frequency of the second layer antenna change with the change of the first target frequency and / or the second target frequency.

15. A vehicle, characterized in that, Includes the vehicle body and the dual-layer dual-frequency satellite navigation antenna whose parameters are calibrated using the parameter calibration device for the dual-layer dual-frequency satellite navigation antenna as described in claim 14.

Images