Filtering antenna design method of naked mole algorithm based on SHAP

Abstract

The invention discloses a filtering antenna design method of a naked mole algorithm based on SHAP, and relates to the crossing field of a radio frequency front end of a wireless communication system and a computer algorithm. According to the method, the SHAP value is utilized to quantify the influence of the antenna structure parameters on the antenna performance, the search strategy in the NMRA is dynamically adjusted, and the exploration capability of the algorithm in a high-dimensional space is enhanced. Firstly, the SHAP value of each antenna parameter is initialized, and then the NMRA algorithm stage is entered. In the working stage, SHAP weighted differential evolution is carried out, and in the breeding stage, all breeders update their own states according to a certain probability, and breeders with poor performance are eliminated; self-adaptive variation is carried out according to the probability pm so as to prevent from falling into local optimum; and finally, evaluating the new population, updating the current optimal solution, and outputting the optimal solution after the maximum number of iterations is reached. The method is mainly applied to the fields of short-wave satellite communication, radars and the like, the limitation that a uniform search strategy is adopted for all features in the high-dimensional optimization problem of a traditional NMRA algorithm is overcome, and it is guaranteed that the performance of a filtering antenna is good.

Description

Technical Field

[0001] This invention relates to the intersection of radio frequency front-end and computer algorithms in wireless communication systems, and in particular to a method for designing filtered antennas based on the naked mole-mole algorithm using SHAP. Background Technology

[0002] In recent years, metaheuristic algorithms have demonstrated significant advantages in the field of antenna optimization. For example, algorithms such as Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Differential Evolution (DE), and Grey Wolf Optimization (GWO) can perform global exploration in complex search spaces by simulating natural phenomena or swarm intelligence, and have been successfully applied to various antenna design problems. Among them, the Naked Mole-Rat Algorithm (NMRA) is a novel metaheuristic algorithm that simulates the social behavior and reproductive mechanisms of naked mole rats. With its powerful global exploration and local exploitation capabilities, it shows great potential in antenna optimization. Research shows that NMRA performs excellently in handling high-dimensional, multimodal optimization problems, effectively balancing exploration and exploitation, and avoiding premature convergence.

[0003] Meanwhile, the application of interpretable machine learning techniques in antenna design is gradually gaining attention. SHAP (SHapley Additive exPlanations), as a game theory-based interpretive framework, can quantify the contribution of each feature to the model output, providing an intuitive and reliable explanation for black-box models. Although SHAP was initially used for interpretability analysis of machine learning models, it has great potential in feature importance assessment and decision guidance in optimization problems.

[0004] Current antenna optimization methods are mainly divided into three categories: traditional mathematical optimization methods, intelligent optimization algorithms, and hybrid optimization frameworks. Traditional mathematical optimization methods, such as gradient descent and conjugate gradient methods, rely on the convexity and derivative information of the problem, often making it difficult to find the global optimum for complex antenna optimization problems that are non-convex or non-differentiable. Intelligent optimization algorithms do not rely on gradient information and have strong global search capabilities, but they also have some inherent drawbacks. For example, genetic algorithms are prone to premature convergence; particle swarm optimization algorithms are prone to getting trapped in local optima; and while the naked mole-rat algorithm has advantages in exploration capabilities, its convergence speed is slow when dealing with high-dimensional problems. Therefore, providing an efficient and reliable new method for antenna performance optimization and solving the problem of getting trapped in local optima in high-dimensional antenna optimization remains a challenge that needs to be addressed. Summary of the Invention

[0005] In view of this, this invention proposes a filter antenna design method based on the naked mole-mole algorithm using SHAP. This method effectively overcomes the limitation of the traditional NMRA algorithm in using a uniform search strategy for all features in high-dimensional optimization problems, ensuring that the designed filter antenna has good performance.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] A method for designing a filtered antenna based on the naked mole-mouse algorithm using SHAP, comprising the following steps:

[0008] Step 1: Initialize the reproduction probability pb, the mutation probability pm, and the naked mole-rat population, where the number of breeding rats B = n / 5 and the number of working rats W = nB; where n is the size of the naked mole-rat population.

[0009] Step 2: Connect the electromagnetic simulation software with the antenna model;

[0010] Step 3: Set the maximum number of iterations (max-iteration) for the naked mole-mole algorithm and guide the iteration based on SHAP; complete the determination of the filter antenna parameters.

[0011] Furthermore, the formula for initializing the naked mole-rat population in step 1 is as follows:

[0012]

[0013] in , , It is the i-th solution in the j-th dimension, where D is the number of antenna structure parameters to be optimized. and These are the lower and upper bounds of the j-th antenna structure parameter to be optimized, respectively. yes Uniformly random numbers within.

[0014] Furthermore, the specific process of step 3 is as follows:

[0015] Step 301: For each individual in the current naked mole-rat population, calculate its corresponding SHAP value;

[0016] Step 302: Calculate the objective function value corresponding to the current naked mole-rat population, update and record the global optimal solution;

[0017] Step 303: Sort the antenna bandwidth of each individual in the current naked mole rat population, and record the B individuals with the largest bandwidth as breeding rats in the current naked mole rat population, and the remaining W individuals as working rats in the current naked mole rat population;

[0018] Step 304: Enter the worker stage, determine the weights based on the SHAP value, and update the worker mouse;

[0019] Step 305: Enter the breeding stage. Determine whether to move to the optimal position based on the breeding probability pb, and update the breeding mice.

[0020] Step 306: Enter the mutation phase and perform SHAP adaptive mutation operation on the updated working mice and breeding mice;

[0021] Step 307: Determine whether the maximum number of iterations (max-iteration) has been reached. If it has, calculate the objective function value corresponding to the current naked mole-rat population, and output the global optimal solution after updating and recording it. If it has not been reached, repeat steps 301 to 307.

[0022] Furthermore, the specific process of step 301 is as follows:

[0023]

[0024]

[0025]

[0026] in Let represent the complete set of features consisting of the D dimensions of the i-th individual in the current naked mole-rat population. Indicates the total set of features Delete from middle The resulting set for Any feature subset, including the empty set, The symbol '!' indicates the number of elements in the set, and '!' indicates factorial. for marginal contribution For feature subset The predicted contribution value is given by the random forest model; and This represents the D dimensions of the i-th individual in the current naked mole-rat population. The maximum and minimum values ​​among the values; This is the SHAP value of the j-th dimension solution for the i-th individual in the current naked mole-rat population.

[0027] Furthermore, the specific process of step 302 is as follows:

[0028] Step 3021: For each individual in the current naked mole-rat population, based on the D-dimensional antenna structure parameters, simulate the corresponding antenna model using electromagnetic simulation software to obtain its corresponding smoothness. And bandwidth (BW):

[0029]

[0030] in, and These are the sets of local maximum and local minimum values ​​of the S11 curve within the target frequency band;

[0031] Step 3022: If the current naked mole-rat population contains For individuals whose bandwidth B is not 0, the maximum bandwidth value is selected from those individuals and recorded as the objective function value of the current naked mole-rat population; otherwise, the objective function value of the current naked mole-rat population is recorded as... This refers to the penalty mechanism; the globally optimal solution is the solution obtained by excluding all iterations. The maximum objective function value other than [the objective function value].

[0032] Furthermore, the specific process of step 304 is as follows:

[0033]

[0034]

[0035] in, In the t-th iteration, the first... The SHAP value corresponding to the j-th dimension solution of the working mouse. In the t-th iteration, the first... The j-th dimension solution of a working mouse. This indicates its corresponding updated value. and This represents the j-th dimension solution of two random working mice in the t-th iteration. It is a scaling factor based on LEVY flight, ranging from... Control the update step size. .

[0036] Furthermore, the specific process of step 305 is as follows:

[0037]

[0038]

[0039] in, In the t-th iteration, the first... The SHAP value corresponding to the j-th dimension solution of a breeding mouse. In the t-th iteration, the first... The j-th dimension solution of a breeding mouse. This indicates its corresponding updated value. Let be the j-th dimension solution of the individual corresponding to the current global optimal solution. For the t-th iteration The random probability of generating the j-th dimension solution of a breeding mouse. Let be a random variable that follows a uniform distribution and has a range of [-1, 1]. It is a scaling factor based on LEVY flight, ranging from... Control the update step size. .

[0040] Furthermore, the SHAP adaptive mutation probability in step 306 is as follows:

[0041] .

[0042] Due to the adoption of the above technical solution, the beneficial effects of this invention compared with the prior art are as follows:

[0043] 1. This invention introduces the SHAP (SHapley Additive exPlanations) interpretability analysis method. By quantifying the contribution of each feature parameter to the optimization objective, the search strategy is dynamically adjusted: for high SHAP feature values, a local fine-grained search is used to deeply explore potential high-quality solutions; for low SHAP feature values, the global exploration capability is enhanced to avoid getting trapped in local optima. This mechanism effectively overcomes the limitation of the traditional NMRA algorithm in using a uniform search strategy for all features in high-dimensional optimization problems.

[0044] 2. This invention sets the weight function of the naked mole-mole algorithm based on the SHAP value. If the current parameter is highly important, its small change may cause large fluctuations in the optimization objective. If a large step size is used, it may skip the local region where the optimal solution is located, and the optimization objective will be difficult to converge. Therefore, if the parameter is highly important, its SHAP value should be large and its weight value should be small. The optimization step size should be small during the working phase to avoid destroying the already found better solution due to excessive adjustment.

[0045] 3. This invention designs a customized objective function for the filter antenna based on bandwidth, ensuring that the designed filter antenna has good performance. Attached Figure Description

[0046] Figure 1 This is an overall flowchart of a filtered antenna design method based on the SHAP naked mole-mole algorithm in an embodiment of the present invention.

[0047] Figure 2 This is a schematic diagram comparing the optimization results of SG-NMRA with those of traditional algorithms when maximizing bandwidth is the optimization objective in an embodiment of the present invention.

[0048] Figure 3This is a schematic diagram comparing the iteration curves of SG-NMRA with traditional algorithms in an embodiment of the present invention. Detailed Implementation

[0049] The invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0050] This paper presents a SHAP-based naked mole-rat algorithm for designing filtered antennas. Since there are many antenna structural parameters, and each parameter has a different impact on the overall antenna performance, we use SHAP values ​​to quantify the impact of each design parameter on antenna performance, dynamically adjusting the search strategy in the NMRA (Near Marrow-Rat Algorithm) to enhance the algorithm's exploration capability in high-dimensional spaces. Then, the NMRA algorithm is implemented. Weights are determined based on SHAP values. If a parameter is highly important, even small changes can lead to significant fluctuations in the optimization objective; therefore, smaller weights are used to avoid skipping the local region where the optimal solution is located. In the working phase, SHAP-weighted differential evolution is performed. In the breeding phase, all breeders update their states with a certain probability, which helps to eliminate poorly performing breeders. Then, adaptive mutation is performed with probability pm to prevent getting trapped in local optima. Finally, the new population is evaluated, the current optimal solution is updated, and after reaching the maximum number of iterations, the optimal solution is output. Figure 1 As shown, the specific steps include:

[0051] Step 1: Initialize the reproductive probability pb, mutation probability pm, and initialize the naked mole-rat population, where the number of breeding rats is B = n / 5, and the number of working rats is W = nB; where n is the naked mole-rat population size, and n is a multiple of 5; pb, pm ∈ .

[0052] Step 2: Connect the electromagnetic simulation software with the antenna model;

[0053] Step 3: Set the maximum number of iterations (max-iteration) for the naked mole-mole algorithm and guide the iteration based on SHAP; complete the determination of the filter antenna parameters.

[0054] Furthermore, the formula for initializing the naked mole-rat population in step 1 is as follows:

[0055]

[0056] in , , It is the i-th solution in the j-th dimension, where D is the number of antenna structure parameters to be optimized. and These are the lower and upper bounds of the j-th antenna structure parameter to be optimized, respectively. yes Uniformly random numbers within.

[0057] Furthermore, the specific process of step 3 is as follows:

[0058] Step 301: For each individual in the current naked mole-rat population, calculate its corresponding SHAP value;

[0059] Step 302: Calculate the objective function value corresponding to the current naked mole-rat population, update and record the global optimal solution;

[0060] Step 303: Sort the antenna bandwidth of each individual in the current naked mole rat population, and record the B individuals with the largest bandwidth as breeding rats in the current naked mole rat population, and the remaining W individuals as working rats in the current naked mole rat population;

[0061] Step 304: Enter the worker stage, determine the weights based on the SHAP value, and update the worker mouse;

[0062] Step 305: Enter the breeding stage. Determine whether to move to the optimal position based on the breeding probability pb, and update the breeding mice.

[0063] Step 306: Enter the mutation phase and perform SHAP adaptive mutation operation on the updated working mice and breeding mice;

[0064] Step 307: Determine whether the maximum number of iterations (max-iteration) has been reached. If it has, calculate the objective function value corresponding to the current naked mole-rat population, and output the global optimal solution after updating and recording it. If it has not been reached, repeat steps 301 to 307.

[0065] Furthermore, the specific process of step 301 is as follows:

[0066]

[0067]

[0068]

[0069] in Let represent the complete set of features consisting of the D dimensions of the i-th individual in the current naked mole-rat population. Indicates the total set of features Delete from middle The resulting set for Any feature subset, including the empty set, The symbol '!' indicates the number of elements in the set, and '!' indicates factorial. for marginal contribution For feature subset The predicted contribution value is given by the random forest model. Specifically, in this embodiment, a TreeExplainer package is called to implement it. and This represents the D dimensions of the i-th individual in the current naked mole-rat population. The maximum and minimum values ​​among the values; This is the SHAP value of the j-th dimension solution for the i-th individual in the current naked mole-rat population.

[0070] Furthermore, the specific process of step 302 is as follows:

[0071] Step 3021: For each individual in the current naked mole-rat population, based on the D-dimensional antenna structure parameters, simulate the corresponding antenna model using electromagnetic simulation software to obtain its corresponding smoothness. And bandwidth (BW):

[0072]

[0073] in, and These are the sets of local maximum and local minimum values ​​of the S11 curve within the target frequency band;

[0074] Step 3022: If the current naked mole-rat population contains For individuals whose bandwidth B is not 0, the maximum bandwidth value is selected from those individuals and recorded as the objective function value of the current naked mole-rat population; otherwise, the objective function value of the current naked mole-rat population is recorded as... This refers to the penalty mechanism; the globally optimal solution is the solution obtained by excluding all iterations. The maximum objective function value other than [the objective function value].

[0075] Furthermore, the specific process of step 304 is as follows:

[0076]

[0077]

[0078] in, In the t-th iteration, the first... The SHAP value corresponding to the j-th dimension solution of the working mouse. In the t-th iteration, the first... The j-th dimension solution of a working mouse. This indicates its corresponding updated value. and This represents the j-th dimension solution of two random working mice in the t-th iteration. It is a scaling factor based on LEVY flight, ranging from... Control the update step size. . These are parameter weights obtained from the SHAP value, used for adaptive step size adjustment.

[0079] Furthermore, the specific process of step 305 is as follows:

[0080]

[0081]

[0082] in, In the t-th iteration, the first... The SHAP value corresponding to the j-th dimension solution of a breeding mouse. In the t-th iteration, the first... The j-th dimension solution of a breeding mouse. This indicates its corresponding updated value. Let be the j-th dimension solution of the individual corresponding to the current global optimal solution. For the t-th iteration The random probability of generating the j-th dimension solution of a breeding mouse. Let be a random variable that follows a uniform distribution and has a range of [-1, 1]. It is a scaling factor based on LEVY flight, ranging from... Control the update step size. .

[0083] Furthermore, the SHAP adaptive mutation probability in step 306 is as follows:

[0084] .

[0085] The SHAP adaptive mutation operation is the same as the mutation operation in the naked mole-mole algorithm, targeting the updated working and breeding mice, and the mutation probability is... .

[0086] To comprehensively evaluate the algorithm performance, this embodiment optimizes three structural parameters of the filter antenna and selects six standard algorithms. The proposed SG-NMRA algorithm is then compared with the traditional Naked Mole-Rat Algorithm (NMRA), Gray Wolf Optimization Algorithm (GWO), Artificial Rabbit Algorithm (ARO, 2022), Coati Optimization Algorithm (CoatiOA, 2023), Egret Flock Optimization Algorithm (ESOA, 2022), and Golden Jackal Optimization Algorithm (GJO, 2022). Figure 2 The optimization results of each algorithm are shown when the optimization objective is to maximize bandwidth. It can be seen that SG-NMRA can achieve a wider bandwidth compared with the other algorithms. Figure 3 It demonstrates that SG-NMRA converges faster in terms of bandwidth optimization compared to the other six algorithms, and avoids getting stuck in local optima.

[0087] Those skilled in the art will recognize that the described embodiments are intended to help readers understand the principles of the invention and should be understood as not limiting the scope of protection of the invention to the described embodiments. Various modifications and variations can be made to the invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the invention should be included within the scope of the claims of the invention.

Claims

1. A method for designing a filtered antenna based on the SHAP naked mole-mouse algorithm, characterized in that, Includes the following steps: Step 1: Initialize the reproduction probability pb, the mutation probability pm, and the naked mole-rat population, where the number of breeding rats B = n / 5 and the number of working rats W = nB; where n is the size of the naked mole-rat population. Step 2: Connect the electromagnetic simulation software with the antenna model; Step 3: Set the maximum number of iterations (max-iteration) for the naked mole-mole algorithm and guide the iteration based on SHAP; complete the determination of the filter antenna parameters.

2. The method for designing a filtered antenna based on the SHAP naked mole-mouse algorithm according to claim 1, characterized in that, The formula for initializing the naked mole-rat population in step 1 is as follows: ； in , , It is the i-th solution in the j-th dimension, where D is the number of antenna structure parameters to be optimized. and These are the lower and upper bounds of the j-th antenna structure parameter to be optimized, respectively. yes Uniformly random numbers within.

3. The method for designing a filtered antenna based on the SHAP naked mole-mouse algorithm according to claim 2, characterized in that, The specific process of step 3 is as follows: Step 301: For each individual in the current naked mole-rat population, calculate its corresponding SHAP value; Step 302: Calculate the objective function value corresponding to the current naked mole-rat population, update and record the global optimal solution; Step 303: Sort the antenna bandwidth of each individual in the current naked mole rat population, and record the B individuals with the largest bandwidth as breeding rats in the current naked mole rat population, and the remaining W individuals as working rats in the current naked mole rat population; Step 304: Enter the worker stage, determine the weights based on the SHAP value, and update the worker mouse; Step 305: Enter the breeding stage. Determine whether to move to the optimal position based on the breeding probability pb, and update the breeding mice. Step 306: Enter the mutation phase and perform SHAP adaptive mutation operation on the updated working mice and breeding mice; Step 307: Determine whether the maximum number of iterations (max-iteration) has been reached. If it has, calculate the objective function value corresponding to the current naked mole-rat population, and output the global optimal solution after updating and recording it. If it has not been reached, repeat steps 301 to 307.

4. The method for designing a filtered antenna based on the SHAP naked mole-mouse algorithm according to claim 3, characterized in that, The specific process of step 301 is as follows: ； ； ； in Let represent the complete set of features consisting of the D dimensions of the i-th individual in the current naked mole-rat population. Indicates the total set of features Delete from middle The resulting set for Any feature subset, including the empty set, The symbol '!' indicates the number of elements in the set, and '!' indicates factorial. for marginal contribution For feature subset The predicted contribution value is given by the random forest model; and This represents the D dimensions of the i-th individual in the current naked mole-rat population. The maximum and minimum values ​​among the values; This is the SHAP value of the j-th dimension solution for the i-th individual in the current naked mole-rat population.

5. The method for designing a filtered antenna based on the SHAP naked mole-mouse algorithm according to claim 3, characterized in that, The specific process of step 302 is as follows: Step 3021: For each individual in the current naked mole-rat population, based on the D-dimensional antenna structure parameters, simulate the corresponding antenna model using electromagnetic simulation software to obtain its corresponding smoothness. And bandwidth (BW): ； in, and These are the sets of local maximum and local minimum values ​​of the S11 curve within the target frequency band; Step 3022: If the current naked mole-rat population contains For individuals whose bandwidth B is not 0, the maximum bandwidth value is selected from those individuals and recorded as the objective function value of the current naked mole-rat population; otherwise, the objective function value of the current naked mole-rat population is recorded as... This refers to the penalty mechanism; the globally optimal solution is the solution obtained by excluding all iterations. The maximum objective function value other than [the objective function value].

6. The method for designing a filtered antenna based on the SHAP naked mole-mouse algorithm according to claim 4, characterized in that, The specific process of step 304 is as follows: ； ； in, In the t-th iteration, the first... The SHAP value corresponding to the j-th dimension solution of the working mouse. In the t-th iteration, the first... The j-th dimension solution of a working mouse. This indicates its corresponding updated value. and This represents the j-th dimension solution of two random working mice in the t-th iteration. It is a scaling factor based on LEVY flight, ranging from... Control the update step size. .

7. The method for designing a filtered antenna based on the SHAP naked mole-mouse algorithm according to claim 4, characterized in that, The specific process of step 305 is as follows: ； ； in, In the t-th iteration, the first... The SHAP value corresponding to the j-th dimension solution of a breeding mouse. In the t-th iteration, the first... The j-th dimension solution of a breeding mouse. This indicates its corresponding updated value. Let be the j-th dimension solution of the individual corresponding to the current global optimal solution. For the t-th iteration The random probability of generating the j-th dimension solution of a breeding mouse. Let be a random variable that follows a uniform distribution and has a range of [-1, 1]. It is a scaling factor based on LEVY flight, ranging from... Control the update step size. .

8. The method for designing a filtered antenna based on the SHAP naked mole-mouse algorithm according to claim 4, characterized in that, The SHAP adaptive mutation probability in step 306 is as follows: 。

Images