An ai-based duplexer design optimization method and system

Abstract

This invention relates to the field of radio frequency device design optimization technology, and provides an AI-based duplexer design optimization method and system. The method includes: predicting the event frequency characteristics of the future application environment; determining a correction factor based on the magnitude of the exceedance when the event frequency exceeds a threshold; correcting the key structural parameters of the duplexer to be designed based on the correction factor; and completing the duplexer design based on the corrected key structural parameters. Compared to existing methods that rely solely on experience or a single indicator for design, this invention achieves closed-loop optimization from identifying additional environmental factors to adaptive parameter adjustment, exhibiting higher accuracy and foresight, while avoiding the cost increase caused by over-design, and significantly improving the duplexer's adaptability and long-term reliability in complex environments.

Description

Technical Field

[0001] This invention belongs to the field of radio frequency device design optimization technology, and in particular relates to an AI-based duplexer design optimization method and system. Background Technology

[0002] Duplexers, as key components in the radio frequency (RF) front-end, are widely used in wireless communication devices to achieve isolation between transmitted and received signals. In existing technologies, duplexer design is typically based on the intended application scenario. Key structural parameters such as resonant structure parameters, matching parameters, and surface protection parameters are optimized to meet requirements for isolation, insertion loss, and noise performance. These parameters are generally determined through simulation analysis, experimental verification, and historical experience, with trade-offs made between performance and cost to create a design suitable for the target environment.

[0003] However, in the operating environment of low-altitude UAVs at sea, duplexers are constantly exposed to high humidity and high salt spray conditions, making it easy for salt spray particles to adhere to the device surface. Simultaneously, due to factors such as equipment heating, changes in ambient temperature, and airflow, the device surface temperature fluctuates periodically, causing the adhered salt spray to undergo alternating condensation and volatilization processes. This creates a dynamically changing medium environment on the duplexer surface, affecting its electromagnetic properties. Existing technologies typically improve environmental adaptability by applying anti-condensation coatings, hydrophobic coatings, or anti-corrosion treatments, but these designs are mostly based on static environmental conditions and do not provide in-depth analysis of the aforementioned dynamic additional environmental factors.

[0004] In practical applications, when the effects of the additional environmental factors are weak, the performance changes of the duplexer are usually within the normal fluctuation range and are not easily detected. However, as their influence gradually increases, they can lead to implicit degradation phenomena such as decreased isolation and increased noise floor in the receiving link. These changes are usually small and do not trigger alarms, and existing technologies lack effective identification and pre-optimization methods. Therefore, how to identify the impact of these additional environmental factors during the design phase and optimize key structural parameters accordingly to improve the stability of the duplexer in complex environments has become a pressing technical problem in this field. Summary of the Invention

[0005] The purpose of this invention is to provide an AI-based method and system for optimizing duplexer design, aiming to solve the problems mentioned in the background art.

[0006] This invention is implemented as follows: an AI-based duplexer design optimization method, the method comprising:

[0007] If the future application environment of the duplexer to be designed is identified as a preset environment, several historical application samples are selected from the preset duplexer application database. The duplexers corresponding to the samples are consistent with the duplexer to be designed in terms of key structural parameters, and the application environment corresponding to the samples is within the preset similarity range to the future application environment.

[0008] The event frequency characteristics of additional environmental factors under the preset environment, as well as the duplexer isolation change characteristics and the receiver link noise floor change characteristics are extracted from the samples.

[0009] Correlation analysis was performed on the event frequency characteristics, isolation change characteristics, and receiver link noise floor change characteristics to determine the cooperative degradation threshold corresponding to the event frequency characteristics. The threshold is used to characterize that when the event frequency characteristics exceed the threshold, the rate of decrease in duplexer isolation increases and the trend of increase in receiver link noise floor intensifies.

[0010] Predict the event frequency characteristics of the future application environment, and when they exceed the turning point threshold, determine the correction factor based on the magnitude of the exceedance, and correct the key structural parameters of the duplexer to be designed based on the correction factor, and complete the duplexer design based on the corrected key structural parameters.

[0011] As a further limitation of the technical solution of this embodiment of the invention, the preset environment is the operating environment of a low-altitude unmanned aerial vehicle at sea, and salt spray exists in the environment.

[0012] As a further limitation of the technical solution of the present invention, the key structural parameters include surface protection parameters, which include coating thickness, protective layer thickness, and hydrophobic treatment intensity.

[0013] As a further limitation of the technical solution of the present invention, the application environment corresponding to the sample and the future application environment being within a preset similar range specifically means that, within a preset application period after being put into use, at least one or more of the statistical characteristics of environmental humidity, salt spray concentration and temperature change range are within a preset similar range.

[0014] As a further limitation of the technical solution of the present invention, the additional environmental factors specifically refer to the environmental process in which salt spray particles, after adhering to the surface of the duplexer, undergo alternating condensation and volatilization under the influence of temperature changes.

[0015] As a further limitation of the technical solution of the present invention, the event frequency characteristic is calculated as follows: the number of times the salt spray condensation and volatilization alternation event occurs within a preset application period is counted, and the result is calculated based on the number of occurrences and the time length of the preset application period.

[0016] As a further limitation of the technical solution of this invention, the steps of performing correlation analysis on event frequency characteristics, isolation change characteristics, and receiver link noise floor change characteristics to determine the cooperative degradation threshold corresponding to the event frequency characteristics, and using the threshold to characterize the increased rate of duplexer isolation decrease and enhanced trend of receiver link noise floor increase when the event frequency characteristics exceed the threshold, include:

[0017] A sample sequence is obtained by sorting several samples according to their event frequency characteristics from smallest to largest.

[0018] Based on the sample sequence, the trends of isolation change features with event frequency features and the trends of received link noise floor change features with event frequency features are extracted.

[0019] Determine the rate of change of the isolation degree change characteristic and the rate of change of the receive link noise floor change characteristic, respectively;

[0020] If, within a certain range of event frequency characteristics, the rate of change of the isolation change characteristic exceeds the corresponding first preset change threshold, and the rate of change of the received link noise floor change characteristic exceeds the corresponding second preset change threshold, then the range is determined to correspond to a cooperative degradation interval, and an event frequency characteristic is selected from the cooperative degradation interval as the cooperative degradation inflection threshold.

[0021] As a further limitation of the technical solution of this embodiment of the invention, the steps of predicting the event frequency characteristics of the future application environment, determining a correction factor based on the magnitude of the excess when the event frequency exceeds a threshold, and correcting the key structural parameters of the duplexer to be designed based on the correction factor, and completing the duplexer design based on the corrected key structural parameters include:

[0022] Calculate the event frequency characteristics of the duplexer to be designed within a preset application cycle in a future application environment, and determine whether it exceeds the threshold value.

[0023] When the event frequency characteristic does not exceed the turning point threshold, the key structural parameters remain unchanged;

[0024] When the event frequency characteristic exceeds the turning point threshold, a correction factor is determined based on the extent to which the event frequency characteristic exceeds the turning point threshold, and the key structural parameters of the duplexer to be designed are corrected based on the correction factor to obtain the optimized key structural parameters.

[0025] The duplexer design was completed based on the optimized key structural parameters.

[0026] As a further limitation of the technical solution of this embodiment of the invention, when modifying key structural parameters, a preset correction function is used, the correction function including:

[0027] ;

[0028] in, This refers to the revised key structural parameters. This refers to the original key structural parameters. This refers to the maximum upper limit of the value of the key structural parameter. This refers to the event frequency characteristics of the future application environment. This refers to the turning point threshold. This refers to exceeding the allowable range. This refers to the preset control correction strength coefficient, and it satisfies... Greater than 0, This refers to the correction factor.

[0029] An AI-based duplexer design optimization system, the system comprising:

[0030] The sample screening module is used to select several historical application samples from the preset duplexer application database if the future application environment of the duplexer to be designed is identified as a preset environment. The duplexers corresponding to the samples are consistent with the duplexer to be designed in terms of key structural parameters, and the application environment corresponding to the samples is within a preset similarity range to the future application environment.

[0031] The feature extraction module is used to extract event frequency features of additional environmental factors under a preset environment, as well as duplexer isolation change features and receiver link noise floor change features from the sample.

[0032] The trend identification module is used to perform correlation analysis on the event frequency characteristics, isolation change characteristics, and receiver link noise floor change characteristics to determine the cooperative degradation threshold corresponding to the event frequency characteristics. The threshold is used to characterize that when the event frequency characteristics exceed the threshold, the rate of decrease in duplexer isolation increases and the trend of rising receiver link noise floor intensifies.

[0033] The parameter optimization module is used to predict the event frequency characteristics of the future application environment, and when they exceed the turning point threshold, it determines the correction factor based on the magnitude of the excess, and corrects the key structural parameters of the duplexer to be designed based on the correction factor, and completes the duplexer design based on the corrected key structural parameters.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] This invention addresses the additional environmental factor present in the operating environment of low-altitude unmanned aerial vehicles (UAVs) at sea: salt spray particles adhere to the surface of the duplexer, undergoing alternating condensation and volatilization under temperature changes. By statistically modeling the event frequency characteristics of this additional environmental factor and combining it with the synergistic analysis of duplexer isolation variation characteristics and receiver link noise floor variation characteristics, a correlation between environmental effects and performance degradation is established. This identifies hidden degradation boundaries, i.e., synergistic degradation thresholds, that are difficult to detect in existing technologies. Based on this, a correction factor is dynamically determined according to the extent to which the event frequency characteristics in future application environments exceed the threshold, enabling pre-optimization of key structural parameters. This ensures that the duplexer maintains stable performance in practical applications. Compared to existing methods that rely solely on experience or single indicators for design, this invention achieves closed-loop optimization from additional environmental factor identification to adaptive parameter adjustment, exhibiting higher accuracy and foresight. It also avoids the increased costs associated with over-design, significantly improving the duplexer's adaptability and long-term reliability in complex environments. Attached Figure Description

[0036] Figure 1 A flowchart of the method provided in the embodiments of the present invention;

[0037] Figure 2 This is a flowchart illustrating the process of determining the collaborative degradation inflection threshold in the method provided by the embodiments of the present invention;

[0038] Figure 3 This is a flowchart illustrating the correction of key structural parameters in the method provided in this embodiment of the invention;

[0039] Figure 4 The application architecture diagram of the system provided in the embodiments of the present invention. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0041] Figure 1 A flowchart of the method provided by an embodiment of the present invention is shown.

[0042] Specifically, an AI-based duplexer design optimization method includes the following steps:

[0043] Step S100: If the future application environment of the duplexer to be designed is identified as a preset environment, select a number of historical application samples from the preset duplexer application database. The duplexers corresponding to the samples are consistent with the duplexer to be designed in terms of key structural parameters. The key structural parameters include surface protection parameters, which include coating thickness, protective layer thickness, and hydrophobic treatment strength.

[0044] The preset environment is a low-altitude maritime UAV operating environment, which contains salt spray. A unique additional environmental factor exists in the preset environment: after salt spray particles adhere to the surface of the duplexer, they undergo an alternating process of condensation and volatilization under temperature changes.

[0045] The application environment corresponding to the sample and the future application environment are within a preset similarity range; specifically, it means that within a preset application period after the sample is put into use, at least one or more of the statistical characteristics of environmental humidity, salt spray concentration and temperature change range are within a preset similarity range.

[0046] In this embodiment of the invention, based on existing artificial intelligence technology, it is applied to the research and development design stage of a duplexer before production. By analyzing historical operating data, the key structural parameters of the duplexer to be designed are pre-optimized, thereby improving the adaptability and long-term stability of the duplexer in the target application environment. Those skilled in the art know that duplexers have multiple adjustable directions during the design process, such as resonant structure dimensions, matching structure parameters, surface protection parameters, and packaging-related parameters. These parameters are all designable variables in existing structural systems, and their specific values ​​are usually determined by trade-offs between performance indicators, cost, and applicable environment.

[0047] This invention focuses on a specific application scenario: the operating environment of low-altitude unmanned aerial vehicles (UAVs) at sea. In this environment, salt spray particles are prevalent in the air. Furthermore, due to the UAV's compact structure, open shells, heat dissipation design, and airflow erosion during flight, salt spray particles easily penetrate the equipment or adhere to the duplexer surface, thus exposing the duplexer to a high-humidity, high-salt-spray environment for extended periods. Existing technologies typically mitigate these environmental effects by setting surface protection parameters, such as adjusting anti-condensation coating thickness, applying hydrophobic coatings, and performing surface anti-corrosion treatments, to reduce the environmental impact on the device surface. Generally, existing technologies combine experimental verification results and historical application experience, rationally setting the aforementioned key structural parameters within cost and applicability constraints, thereby enabling the duplexer to meet the usage requirements of conventional environments.

[0048] However, those skilled in the art have discovered that in the aforementioned operating environment of low-altitude UAVs at sea, there are additional environmental factors, namely, the alternating condensation and volatilization of salt spray particles after they adhere to the surface of the duplexer under temperature changes. This process occurs because the UAV experiences periodic temperature changes during flight, such as heat generated by equipment operation, changes in ambient temperature differences, and airflow cooling. These factors cause the salt spray adhering to the duplexer surface to undergo phase changes under certain conditions, resulting in a periodic alternating process of liquid condensation and gaseous volatilization on the device surface.

[0049] Furthermore, through analysis of extensive practical application data, those skilled in the art have discovered that, when the impact of this additional environmental factor is low, existing duplexers can largely withstand its influence through their existing key structural parameters, and the performance changes of the duplexer remain within normal fluctuations, making them difficult to detect. However, when the alternating effects of condensation and evaporation increase to a certain extent, it can trigger a latent degradation in duplexer performance, such as an increased rate of isolation decrease and a gradual increase in the noise floor of the receiving link. Because this type of degradation is relatively small, it usually does not trigger system alarms, and existing technologies often attribute it to random noise or other environmental disturbances, thus failing to address it specifically. Essentially, this indicates that the design of existing key structural parameters has an implicit tolerance range, meaning that it can withstand this additional environmental factor to a certain extent, but performance will gradually deteriorate beyond this range.

[0050] Based on the above findings, this invention proposes to identify the degree of influence of the additional environmental factors in the future application environment of the duplexer during the duplexer design stage, and to establish a relationship model between the additional environmental factors and the duplexer performance based on historical sample data. On this basis, key structural parameters are pre-corrected so that the duplexer can maintain stable performance in actual application.

[0051] In step S100, it is first necessary to identify whether the future application environment of the duplexer to be designed is the preset environment. Specifically, the target application environment can be determined based on task planning information, equipment deployment location, operating platform type, and environmental database or meteorological forecast data. For example, when it is identified that the duplexer to be designed will be deployed on a maritime inspection UAV or maritime operation UAV platform within a preset application period, and when it is determined that the area has typical high salt spray characteristics based on historical environmental data or real-time environmental forecast results, its future application environment can be determined to be the preset environment.

[0052] The pre-defined duplexer application database can be constructed from a combination of historical operation records, experimental test data, and simulation data. Specifically, the database includes duplexer structural parameter data (such as surface protection parameters), operating environment data (including humidity, salt spray concentration, temperature changes, etc.), operating cycle data, and corresponding performance monitoring data (such as isolation changes, receiver link noise floor changes, etc.). By uniformly storing and associating the above multi-source data, a sample set for subsequent analysis is formed.

[0053] When screening historical application samples, limiting the duplexers corresponding to the samples to have the same or similar key structural parameters as the duplexer to be designed is to ensure comparability between samples, ensuring that performance changes in subsequent analyses mainly stem from differences in environmental factors rather than structural differences. Simultaneously, limiting the application environment corresponding to the samples to be within a preset similarity range with the future application environment ensures that the selected samples accurately reflect the environmental characteristics of the target application scenario, thereby improving the effectiveness and reliability of the analysis results. In addition to key structural parameters and environmental factors, factors such as operating frequency band, power level, installation location, and packaging form can be further considered to further improve sample matching accuracy.

[0054] By setting stricter screening criteria, the interference of irrelevant variables on the analysis results can be reduced, and the impact of additional environmental factors on duplexer performance can be extracted more accurately. By introducing a preset similarity range, the problem of insufficient sample size caused by overly strict screening criteria can be avoided, thus achieving a balance between data sufficiency and data consistency.

[0055] In summary, the purpose of sample selection is to construct a set of historical data that are comparable in terms of key structural parameters and environmental characteristics, so that subsequent AI-based correlation analysis can accurately extract the relationship between additional environmental factors and duplexer performance changes, providing a reliable basis for determining the collaborative degradation inflection point threshold and correcting key structural parameters.

[0056] Furthermore, the AI-based duplexer design optimization method also includes the following steps:

[0057] Step S200: Extract the event frequency characteristics of additional environmental factors under the preset environment, as well as the duplexer isolation change characteristics and the receiver link noise floor change characteristics from the sample.

[0058] The frequency characteristic of the event is calculated as follows: the number of times the salt spray condensation and volatilization alternation event occurs within a preset application period is counted, and the result is calculated based on the number of occurrences and the length of the preset application period.

[0059] In this embodiment of the invention, the event frequency characteristic is a statistical description of additional environmental factors within a preset application period, reflecting the frequency of alternating condensation and volatilization processes occurring after salt spray particles adhere to the duplexer surface. Specifically, within the preset application period, the number of occurrences of the alternating salt spray condensation and volatilization events is obtained by identifying and counting them, and then normalized in conjunction with the duration of the preset application period to obtain the event frequency characteristic. For example, the specific form of the event frequency characteristic can be "the number of alternating events occurring per unit time" or "the proportion of alternating events in the total time".

[0060] In practical implementation, the alternating salt spray condensation and volatilization events can be identified in various ways. For example, they can be identified based on environmental sensor data (such as humidity and temperature sensors) combined with preset judgment rules; when humidity and temperature meet the conditions for condensation or volatilization, the corresponding event is determined to have occurred. Alternatively, they can be indirectly identified based on equipment surface condition monitoring data (such as changes in surface conductivity and dielectric properties). Furthermore, they can be inferred from environmental change patterns in historical operating data. By continuously recording events in these ways, the basic data for calculating event frequency characteristics can be formed.

[0061] The significance of introducing the event frequency characteristics lies in transforming the originally random and discrete additional environmental effects into quantifiable and comparable statistical indicators, which can be used for unified analysis between different historical samples and provide input variables for subsequent AI-based correlation analysis, enabling the relationship between additional environmental factors and duplexer performance to be effectively modeled.

[0062] The duplexer isolation variation characteristic is used to characterize the change in isolation performance between the transmit and receive channels of the duplexer. It can be obtained by continuously monitoring or periodically testing the isolation parameters of the duplexer during operation. Specifically, the isolation performance between channels can be characterized by measuring the S-parameters (e.g., S21 or S12) in the scattering parameters of the duplexer, and the isolation variation characteristic, such as the mean change, rate of change, or fluctuation amplitude of the isolation, can be obtained by statistically analyzing the changes of this parameter within a preset application period.

[0063] The noise floor variation characteristics of the receiving link are used to characterize the changes in the noise level of the receiving channel, which can be obtained by monitoring the noise power of the receiving link. In practice, the noise floor variation characteristics of the receiving link can be obtained by measuring parameters such as noise power, level noise, or noise figure at the receiving end and statistically analyzing their changes over a preset application period, such as the mean change, trend, or rate of change of the noise floor.

[0064] It should be noted that the isolation and receiver link noise floor mentioned above are mature and commonly used performance indicators in this field. Isolation is typically used to evaluate the duplexer's ability to isolate transmitted and received signals, and is one of the key indicators in RF front-end design; receiver link noise floor reflects the noise level of the receiving system and is closely related to system sensitivity. In practical applications, these indicators are widely used in mobile communication equipment, wireless base stations, satellite communication systems, and radar systems. By monitoring and analyzing them, the performance stability of the equipment in complex environments can be evaluated.

[0065] By extracting the event frequency characteristics, isolation variation characteristics, and receiver link noise floor variation characteristics, the correlation between additional environmental factors and duplexer performance can be constructed, providing basic data support for subsequently determining the collaborative degradation inflection point threshold.

[0066] The preset application period is preferably set during the initial operation phase after the duplexer to be designed or the duplexer corresponding to the historical application sample is actually put into use. The reason is that during this phase, the duplexer has not yet been affected by long-term aging, material fatigue or other cumulative deterioration factors, and can more realistically reflect the direct effect of additional environmental factors on the duplexer performance. This makes the extracted event frequency characteristics and the corresponding isolation change characteristics and receiver link noise floor change characteristics more representative and comparable, which is conducive to improving the accuracy of subsequent correlation analysis and determination of the collaborative deterioration inflection threshold.

[0067] Furthermore, the AI-based duplexer design optimization method also includes the following steps:

[0068] Step S300: Perform correlation analysis on the event frequency characteristics, isolation change characteristics, and receiver link noise floor change characteristics to determine the cooperative degradation threshold corresponding to the event frequency characteristics. The threshold is used to characterize that when the event frequency characteristics exceed the threshold, the rate of decrease in duplexer isolation increases and the trend of increase in receiver link noise floor intensifies.

[0069] Specifically, Figure 2 A flowchart for determining the collaborative degradation inflection point threshold is shown.

[0070] The process involves correlation analysis between event frequency characteristics, isolation variation characteristics, and receiver link noise floor variation characteristics to determine the cooperative degradation threshold corresponding to the event frequency characteristics. This threshold characterizes the increased rate of duplexer isolation decline and the enhanced upward trend of receiver link noise floor when the event frequency characteristics exceed the threshold. Specifically, this includes the following steps:

[0071] Step S301: Sort several samples according to the event frequency characteristics from smallest to largest to obtain a sample sequence;

[0072] Step S302: Based on the sample sequence, extract the trend of the isolation change feature with the event frequency feature and the trend of the received link noise floor change feature with the event frequency feature.

[0073] Step S303: Determine the rate of change of the isolation degree change characteristic and the rate of change of the receiver link noise floor change characteristic, respectively;

[0074] Step S304: Within a certain range of event frequency characteristics, if the rate of change of the isolation change characteristic exceeds the corresponding first preset change threshold, and the rate of change of the receiving link noise floor change characteristic exceeds the corresponding second preset change threshold, then the range is determined to correspond to a cooperative degradation interval, and an event frequency characteristic is selected from the cooperative degradation interval as the cooperative degradation inflection threshold.

[0075] In this embodiment of the invention, step S300 is the core analysis step, aiming to mine the intrinsic correlation between the event frequency characteristics of additional environmental factors and the performance changes of the duplexer based on historical application sample data, and further identify a critical boundary, namely the collaborative degradation turning point threshold. By determining this turning point threshold, the impact of additional environmental factors on the duplexer performance can be distinguished between "imperceptible or negligible normal fluctuation range" and "sensitive range that gradually triggers performance degradation," thereby providing a clear basis for the subsequent correction of key structural parameters. Compared with the prior art, which only monitors or sets based on a single performance indicator, this invention introduces the collaborative analysis of isolation change characteristics and receiver link noise floor change characteristics to identify latent degradation behavior, achieving higher accuracy and foresight.

[0076] In step S301, several historical application samples are sorted according to their event frequency characteristics from smallest to largest to obtain a sample sequence. This step can be implemented using conventional data sorting algorithms, such as quicksort or database-based sorting operations. By arranging the event frequency characteristics in an orderly manner, the samples can be organized according to the degree of influence of additional environmental factors from weakest to strongest, thus providing a foundation for subsequent trend analysis.

[0077] In step S302, based on the sample sequence, the changing trends of isolation variation characteristics and received link noise floor variation characteristics with event frequency characteristics are extracted. Specifically, data fitting or smoothing methods, such as moving average, locally weighted regression (LOESS), or multinomial fitting, can be used to process discrete sample points to obtain continuous trend curves. Through the above processing, the interference caused by fluctuations in individual samples can be effectively reduced, making the overall variation pattern clearer.

[0078] In step S303, the rates of change of the isolation degree variation characteristic and the rates of change of the received link noise floor variation characteristic are determined respectively. Specifically, the variation trend curves obtained in step S302 can be differentially or derivatively calculated to obtain the rate of change of each feature with the event frequency characteristic. In discrete implementations, the approximate rate of change can be calculated by the ratio of differences between adjacent samples; in continuous models, the rate of change can be obtained by differentiating the fitted curve. The purpose of this step is to transform the original performance change into a quantitative indicator of "rate of change" in order to more sensitively capture the escalating trend of performance degradation.

[0079] In step S304, if the rate of change of the isolation change feature exceeds the corresponding first preset change threshold and the rate of change of the received link noise floor change feature exceeds the corresponding second preset change threshold within a certain range of event frequency features, then the range is determined to be a cooperative degradation interval, and an event frequency feature is selected from the cooperative degradation interval as the cooperative degradation inflection threshold.

[0080] Specifically, the first preset change threshold and the second preset change threshold are used to characterize the significant change boundaries of the isolation rate change rate and the receive link noise floor change rate, respectively, and their settings can be obtained based on historical sample data statistics. For example, the mean and standard deviation of the change rate can be statistically analyzed within a low event frequency characteristic range, and the preset change threshold can be set as the standard deviation of the mean-weighted multiple, or empirically set as the upper bound of the normal fluctuation range. Alternatively, the change rate distribution can be modeled using machine learning methods to adaptively determine the threshold.

[0081] After determining the co-degradation range, an event frequency characteristic needs to be selected from this range as the co-degradation inflection threshold. In this embodiment, the initial event frequency characteristic of the co-degradation range can preferably be selected as the inflection threshold to reflect the earliest critical point at which performance begins to degrade. In another embodiment, the inflection threshold can also be selected by combining the predicted characteristics of the future application environment and choosing the event frequency characteristic within the co-degradation range that is closest to the future application environment, in order to improve the pertinence and adaptability of subsequent parameter corrections.

[0082] Through the above steps, the critical boundaries of the synergistic degradation of duplexer performance under the influence of additional environmental factors can be identified in historical data. This provides a basis for determining the correction factor based on the relationship between event frequency characteristics and turning point thresholds, giving the adjustment of key structural parameters a clear basis and direction.

[0083] Furthermore, the AI-based duplexer design optimization method also includes the following steps:

[0084] Step S400: Predict the event frequency characteristics of the future application environment, and when it exceeds the turning point threshold, determine the correction factor according to the magnitude of the excess, and correct the key structural parameters of the duplexer to be designed based on the correction factor, and complete the duplexer design based on the corrected key structural parameters.

[0085] Specifically, Figure 3 A flowchart illustrating the correction of key structural parameters is shown.

[0086] The process involves predicting the event frequency characteristics of the future application environment, determining a correction factor based on the magnitude of the exceedance when the event frequency exceeds a threshold, and then correcting the key structural parameters of the duplexer to be designed based on the correction factor. The duplexer design is then completed based on the corrected key structural parameters, specifically including the following steps:

[0087] Step S401: Calculate the event frequency characteristics of the duplexer to be designed within a preset application cycle in the future application environment, and determine whether it exceeds the turning point threshold.

[0088] Step S402: When the event frequency feature does not exceed the turning threshold, the key structural parameters remain unchanged;

[0089] Step S403: When the event frequency feature exceeds the turning threshold, a correction factor is determined based on the extent of the event frequency feature exceeding the turning threshold, and the key structural parameters of the duplexer to be designed are corrected based on the correction factor to obtain the optimized key structural parameters.

[0090] Step S404: Complete the duplexer design based on the optimized key structural parameters.

[0091] When correcting key structural parameters, a preset correction function is used, which includes:

[0092] ;

[0093] in, This refers to the revised key structural parameters. This refers to the original key structural parameters. This refers to the maximum upper limit of the value of the key structural parameter. This refers to the event frequency characteristics of the future application environment. This refers to the turning point threshold. This refers to exceeding the allowable range. This refers to the preset control correction strength coefficient, and it satisfies... Greater than 0, This refers to the correction factor.

[0094] In this embodiment of the invention, step S400 is the implementation and closing step of the invention. Its core purpose is to apply the cooperative degradation inflection point obtained in step S300 to actual design decisions, thereby realizing the transformation from "pattern discovery" to "design optimization". Specifically, steps S100 to S300 complete the modeling and threshold identification of the relationship between additional environmental factors and duplexer performance, while step S400 further makes targeted corrections to key structural parameters based on the degree of influence of additional environmental factors in the future application environment of the duplexer to be designed, so that the duplexer can still maintain stable performance in actual operation.

[0095] It should be noted that when the predicted event frequency characteristics exceed the aforementioned turning point threshold, it means that the effects of additional environmental factors have entered the synergistic degradation range. At this point, the duplexer performance will no longer be within the normal fluctuation range, but will exhibit an accelerated degradation trend. Therefore, this invention does not simply determine whether the threshold is exceeded, but further determines the correction factor based on the magnitude of the exceedance, thereby achieving "degree-driven" parameter adjustment. This approach is a response to the core research content proposed in step S100, namely, that existing duplexer designs have an implicit "tolerance range," and this invention achieves refined pre-adjustment of key structural parameters by quantifying the degree to which this range is exceeded.

[0096] In step S401, the event frequency characteristics of the duplexer to be designed are calculated within a preset application cycle in the future application environment, and it is determined whether they exceed the transition threshold. This step can be implemented based on environmental prediction data, such as weather forecast data, historical environmental databases, or operational task planning information, to predict changes in humidity, salt spray concentration, and temperature in the future application environment. Combined with the event identification rules in step S200, the alternating events of salt spray condensation and volatilization are estimated to obtain the corresponding event frequency characteristics.

[0097] In step S402, when the event frequency characteristic does not exceed the turning point threshold, the key structural parameters remain unchanged. The significance of this step is to avoid over-design; that is, when additional environmental factors are within acceptable limits, maintaining the original design can meet the usage requirements, thereby avoiding unnecessary cost increases or structural complexity.

[0098] In step S403, when the event frequency characteristic exceeds the threshold, a correction factor is determined based on the extent of the event frequency characteristic exceeding the threshold. The key structural parameters of the duplexer under design are then corrected based on this correction factor to obtain optimized key structural parameters. The significance of using "exceedance magnitude" as the correction criterion lies in expressing the degree of influence of additional environmental factors continuously, allowing parameter adjustments to dynamically change with environmental intensity, thereby avoiding the coarse adjustment problem caused by using only binary judgment. In contrast, if only the judgment method of "whether it exceeds the threshold" is used, it is impossible to distinguish between slight and severe exceedances. By introducing the exceedance magnitude, gradual adjustment can be achieved.

[0099] In other implementations, different comparison methods can be used to determine the correction factor, such as hierarchical adjustment based on piecewise functions, nonlinear amplification based on exponential functions, or direct output of the correction factor through a machine learning model, thereby further improving adaptability.

[0100] In this embodiment, when modifying key structural parameters, a preset modification function is used to adjust them. This modification function is constructed based on the relationship between event frequency characteristics and turning thresholds, and has the advantages of simple form and clear physical meaning, which can intuitively reflect the correspondence between environmental intensity and parameter adjustment. At the same time, by introducing a maximum upper limit limit, this modification function avoids over-adjustment of key structural parameters, thereby ensuring the feasibility of the design.

[0101] It should be noted that the correction function is not limited to a linear form. In other embodiments, a nonlinear function or a piecewise function can also be used. For example, a slow adjustment can be used when the threshold is approached, and an accelerated adjustment can be used when the threshold is far exceeded, so as to further improve the design flexibility.

[0102] To facilitate understanding of the implementation process of this invention, an example is given below. Assume the initial critical structural parameter of a duplexer to be designed is 10, with a corresponding maximum allowable upper limit of 15; the transition threshold determined through step S300 is 5. After predicting the future application environment, the event frequency characteristic is obtained as 8, which exceeds the transition threshold by 3, and the relative proportion is 3 divided by 5 equals 0.6. If the preset control correction strength coefficient is 0.5, then the correction factor is 0.5 multiplied by 0.6 equals 0.3. Based on this correction factor, the initial critical structural parameter is adjusted, i.e., 10 multiplied by 1 plus 0.3 equals 13. Since 13 does not exceed the maximum upper limit of 15, the final optimized critical structural parameter is 13.

[0103] In the above process, the control correction strength coefficient is a key parameter used to adjust the sensitivity of key structural parameters to additional environmental factors. This coefficient can be obtained by fitting historical data, for example, by determining the optimal value through least squares method or regression analysis; or it can be set empirically and optimized while ensuring performance stability. A reasonable setting of this coefficient can achieve a balance between "performance improvement" and "cost control".

[0104] In step S404, the duplexer design is completed based on the optimized key structural parameters. Specifically, the optimized parameters can be input into a duplexer design tool or simulation platform to complete structural modeling, performance simulation, and subsequent production design.

[0105] In summary, this invention, by introducing event frequency characteristics, collaborative degradation thresholds, and correction factors, achieves pre-optimization design of key structural parameters of the duplexer, effectively solving the problem raised in step S100 where the prior art failed to identify the implicit influence of additional environmental factors. This invention not only improves the long-term stability of the duplexer in complex environments but also avoids increased costs due to over-design, demonstrating significant engineering application value.

[0106] This invention is applicable to the design of radio frequency front-ends in marine unmanned aerial vehicle communication equipment, maritime communication systems, and other environments with high humidity and high salinity, and has broad application prospects.

[0107] Furthermore, Figure 4 An application architecture diagram of the system provided in an embodiment of the present invention is shown.

[0108] In another preferred embodiment of the present invention, an AI-based duplexer design optimization system includes:

[0109] The sample screening module 100 is used to screen several historical application samples from the preset duplexer application database if the future application environment of the duplexer to be designed is identified as a preset environment. The duplexers corresponding to the samples are consistent with the duplexer to be designed in terms of key structural parameters, and the application environment corresponding to the samples is within a preset similarity range to the future application environment.

[0110] Furthermore, the AI-based duplexer design optimization system also includes:

[0111] The feature extraction module 200 is used to extract event frequency features of additional environmental factors under a preset environment, as well as duplexer isolation change features and receiver link noise floor change features from the sample.

[0112] Furthermore, the AI-based duplexer design optimization system also includes:

[0113] The trend recognition module 300 is used to perform correlation analysis on the event frequency characteristics, isolation change characteristics, and receiver link noise floor change characteristics to determine the cooperative degradation threshold corresponding to the event frequency characteristics. The threshold is used to characterize that when the event frequency characteristics exceed the threshold, the rate of decrease in duplexer isolation increases and the trend of increase in receiver link noise floor intensifies.

[0114] Furthermore, the AI-based duplexer design optimization system also includes:

[0115] The parameter optimization module 400 is used to predict the event frequency characteristics of the future application environment, and when it exceeds the turning point threshold, it determines the correction factor according to the magnitude of the excess, and corrects the key structural parameters of the duplexer to be designed based on the correction factor, and completes the duplexer design based on the corrected key structural parameters.

[0116] It should be understood that although the steps in the flowcharts of the various embodiments of the present invention are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the various embodiments may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of the sub-steps or stages of other steps.

[0117] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0118] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0119] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

[0120] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An AI-based duplexer design optimization method, characterized by, The method includes: If the future application environment of the duplexer to be designed is identified as a preset environment, several historical application samples are selected from the preset duplexer application database. The duplexers corresponding to the samples are consistent with the duplexer to be designed in terms of key structural parameters, and the application environment corresponding to the samples is within the preset similarity range to the future application environment. The event frequency characteristics of additional environmental factors under the preset environment, as well as the duplexer isolation change characteristics and the receiver link noise floor change characteristics are extracted from the samples. Correlation analysis was performed on the event frequency characteristics, isolation change characteristics, and receiver link noise floor change characteristics to determine the cooperative degradation threshold corresponding to the event frequency characteristics. The threshold is used to characterize that when the event frequency characteristics exceed the threshold, the rate of decrease in duplexer isolation increases and the trend of increase in receiver link noise floor intensifies. Predict the event frequency characteristics of the future application environment, and when they exceed the turning point threshold, determine the correction factor based on the magnitude of the exceedance, and correct the key structural parameters of the duplexer to be designed based on the correction factor, and complete the duplexer design based on the corrected key structural parameters.

2. The AI-based duplexer design optimization method of claim 1, wherein, The preset environment is a low-altitude maritime unmanned aerial vehicle (UAV) operating environment, which contains salt spray. 3.The AI-based duplexer design optimization method of claim 1, wherein, The key structural parameters include surface protection parameters, which include coating thickness, protective layer thickness, and hydrophobic treatment intensity. 4.The AI-based duplexer design optimization method of claim 1, wherein, The application environment corresponding to the sample and the future application environment being within a preset similarity range specifically means that, within a preset application period after being put into use, at least one or more of the statistical characteristics of environmental humidity, salt spray concentration and temperature change range are within a preset similarity range.

5. The AI-based duplexer design optimization method of claim 2, wherein, The additional environmental factors specifically refer to the environmental process in which salt spray particles, after adhering to the surface of the duplexer, undergo alternating condensation and volatilization under temperature changes.

6. The AI-based duplexer design optimization method of claim 5, wherein, The frequency characteristic of the event is calculated as follows: the number of times the salt spray condensation and volatilization alternation event occurs within a preset application period is counted, and the result is calculated based on the number of occurrences and the length of the preset application period.

7. The AI-based duplexer design optimization method of claim 1, wherein, Correlation analysis is performed on event frequency characteristics, isolation variation characteristics, and receiver link noise floor variation characteristics to determine the cooperative degradation threshold corresponding to the event frequency characteristics. The threshold is used to characterize the increased rate of duplexer isolation decrease and the enhanced upward trend of receiver link noise floor when the event frequency characteristics exceed the threshold. The steps include: A sample sequence is obtained by sorting several samples according to their event frequency characteristics from smallest to largest. Based on the sample sequence, the trends of isolation change features with event frequency features and the trends of received link noise floor change features with event frequency features are extracted. Determine the rate of change of the isolation degree change characteristic and the rate of change of the receive link noise floor change characteristic, respectively; If, within a certain range of event frequency characteristics, the rate of change of the isolation change characteristic exceeds the corresponding first preset change threshold, and the rate of change of the received link noise floor change characteristic exceeds the corresponding second preset change threshold, then the range is determined to correspond to a cooperative degradation interval, and an event frequency characteristic is selected from the cooperative degradation interval as the cooperative degradation inflection threshold.

8. The AI-based duplexer design optimization method of claim 1, wherein, The process involves predicting the event frequency characteristics of the future application environment, determining a correction factor based on the magnitude of the exceedance when the event frequency exceeds a threshold, and then correcting the key structural parameters of the duplexer to be designed based on the correction factor. The steps for completing the duplexer design based on the corrected key structural parameters include: Calculate the event frequency characteristics of the duplexer to be designed within a preset application cycle in a future application environment, and determine whether it exceeds the threshold value. When the event frequency characteristic does not exceed the turning point threshold, the key structural parameters remain unchanged; When the event frequency characteristic exceeds the turning point threshold, a correction factor is determined based on the extent to which the event frequency characteristic exceeds the turning point threshold, and the key structural parameters of the duplexer to be designed are corrected based on the correction factor to obtain the optimized key structural parameters. The duplexer design was completed based on the optimized key structural parameters.

9. The AI-based duplexer design optimization method according to claim 8, characterized in that, When correcting key structural parameters, a preset correction function is used, which includes: ； in, This refers to the revised key structural parameters. This refers to the original key structural parameters. This refers to the maximum upper limit of the value of the key structural parameter. This refers to the event frequency characteristics of the future application environment. This refers to the turning point threshold. This refers to exceeding the allowable range. This refers to the preset control correction strength coefficient, and it satisfies... Greater than 0, This refers to the correction factor.

10. An AI-based duplexer design optimization system, characterized in that, The system includes: The sample screening module is used to select several historical application samples from the preset duplexer application database if the future application environment of the duplexer to be designed is identified as a preset environment. The duplexers corresponding to the samples are consistent with the duplexer to be designed in terms of key structural parameters, and the application environment corresponding to the samples is within a preset similarity range to the future application environment. The feature extraction module is used to extract event frequency features of additional environmental factors under a preset environment, as well as duplexer isolation change features and receiver link noise floor change features from the sample. The trend identification module is used to perform correlation analysis on the event frequency characteristics, isolation change characteristics, and receiver link noise floor change characteristics to determine the cooperative degradation threshold corresponding to the event frequency characteristics. The threshold is used to characterize that when the event frequency characteristics exceed the threshold, the rate of decrease in duplexer isolation increases and the trend of rising receiver link noise floor intensifies. The parameter optimization module is used to predict the event frequency characteristics of the future application environment, and when they exceed the turning point threshold, it determines the correction factor based on the magnitude of the excess, and corrects the key structural parameters of the duplexer to be designed based on the correction factor, and completes the duplexer design based on the corrected key structural parameters.

Images