Direction of arrival estimation method fusing improved genetic algorithm and MUSIC

Abstract

The invention discloses a direction-of-arrival estimation method fusing an improved genetic algorithm and MUSIC, and relates to the technical field of underwater target DOA estimation, and the method comprises the steps: S1, initialization population optimization, S2, fitness evaluation and factor generation, S3, fitness adjustment and genetic manipulation, and S4, underwater direction-of-arrival estimation: employing an ant colony algorithm to optimize the initialization population of the genetic algorithm, and carrying out the optimization of the initial population of the genetic algorithm. A high-quality starting point is established for genetic search, an adjustment factor is adaptively generated for each individual by using a particle swarm algorithm, dynamic correction based on an MUSIC algorithm estimation result is realized, the algorithm exploration and development capability is more accurately balanced, and finally a high-precision underwater direction-of-arrival estimation result is output after improved genetic manipulation iterative convergence. According to the method, the direction-of-arrival estimation precision and the algorithm stability in a low signal-to-noise ratio and multi-target coexistence scene in an underwater complex environment are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of underwater target DOA estimation technology, specifically to a wave arrival direction estimation method that integrates an improved genetic algorithm and MUSIC. Background Technology

[0002] Currently, underwater direction-of-arrival estimation is a core area of ​​underwater acoustic signal processing. The MUSIC algorithm, with its high resolution, has become a classic foundational method in this field, driving early technological development. As the complexity of underwater environments increases, algorithm fusion has become an evolutionary trend. The global search advantage of genetic algorithms has been introduced into relevant research, and after improvement, its adaptability has been further optimized. The fusion of the two empowers underwater direction-of-arrival estimation technology, promoting its continuous development towards a direction that is more adaptable to complex scenarios and more reliable and efficient.

[0003] Existing technologies, such as the invention patent application with publication number CN115186697B, disclose an underwater direction-of-arrival (DOA) estimation preprocessing method. This method involves improving the BERT model, inputting a linear array signal, and outputting denoised data and sound velocity estimates to enhance DOA estimation performance under low signal-to-noise ratio (SNR) and uncertain sound velocity conditions. Another existing technology, such as the invention patent application with publication number CN114035151B, discloses a DOA estimation method, device, and system. This method includes: establishing a received signal matrix based on a uniform linear array antenna, obtaining a low-noise, low-dimensional matrix using a low-rank matrix estimation algorithm, updating the LMS step size and weight vector, calculating the radiation pattern and spatial spectrum, and achieving accurate DOA estimation under low SNR conditions.

[0004] As can be seen from the above solutions, the current method for DOA estimation that integrates improved genetic algorithms and MUSIC still has significant limitations. In complex and variable underwater environments, the initial population of traditional genetic algorithms is often randomly generated, resulting in insufficient diversity. This leads to low search efficiency in the early stages of the algorithm and a tendency to get trapped in local optima. Furthermore, existing methods mostly adopt fixed fitness evaluation and selection strategies, which do not make full use of the MUSIC spatial spectrum under low signal-to-noise ratio conditions and fail to dynamically adjust the selection pressure according to individual differences. This limits the algorithm's ability to balance global exploration and local exploitation. In addition, when facing multi-peaked and highly nonlinear DOA estimation problems, the convergence speed and estimation accuracy of single genetic operation strategies still have room for improvement. Especially in scenarios with significant noise interference and dense target angles, the robustness and resolution of existing methods cannot meet the requirements of high-precision estimation. Summary of the Invention

[0005] To address the aforementioned technical shortcomings, the present invention aims to provide a direction-of-arrival estimation method that integrates an improved genetic algorithm with MUSIC.

[0006] To solve the above technical problems, the present invention adopts the following technical solution: The present invention provides a direction-of-arrival estimation method that integrates improved genetic algorithm and MUSIC, including S1. Initial population optimization: ant colony algorithm is used to optimize the initial population of genetic algorithm.

[0007] Preferably, the optimization of the initial population of the genetic algorithm using the ant colony algorithm is specifically performed as follows: S1-1: Constructing the path of the initial population.

[0008] S1-2: Update the pheromone concentration on the initial group path.

[0009] Preferably, the method for constructing the initialization path of the ant colony is as follows: a probabilistic model based on the ant colony algorithm is used to select a path for each ant, wherein the ants... From the current city When selecting a city The probability of it being the next city visited is ,in Ants The city you are currently in , It is a positive integer greater than 2. This indicates the ant's number. , It is a positive integer greater than 2. Indicates the candidate cities to be selected. , Ants From the city you are currently in Cities that are directly accessible and have already been visited in this round A collection of cities outside of China It is a heuristic message with a value of , The current city where the ant is located and candidate cities to be selected The distance between, This indicates the current city where the ant is located. and candidate cities to be selected The pheromone concentration values ​​between them and To control the positive parameters of the relative weights of pheromones and heuristic information, For set Excluding candidate cities Any city other than [the city mentioned above].

[0010] Each ant starts from the starting city and iteratively selects the next city according to the above probability formula until all cities have been visited once, resulting in a complete path composed of the order in which the cities were visited. The set of complete paths constructed by all ants constitutes the initial population.

[0011] Preferably, the method for updating the pheromone concentration on the initialization group path is as follows: updating the pheromone concentration on the initialization group path using a global pheromone update formula, wherein the global pheromone update formula is: ,in , The total number of ants It is the evaporation rate of pheromones. , It is the first The pheromone increment added by an ant along the corresponding edges it traverses on the path it constructs in this round. Indicates the first The total length of the path constructed by an ant, i.e., the sequence of cities it has visited in this round. The sum of the lengths of all sides in the equation.

[0012] S2. Fitness Assessment and Factor Generation: The initial fitness value of each individual in the initial population is calculated using the MUSIC algorithm, and the adjustment factor of each individual is generated using the particle swarm optimization algorithm.

[0013] Preferably, the method for calculating the original fitness value of each individual in the initial population using the MUSIC algorithm is as follows: Each individual in the initial population is decoded into its corresponding direction-of-arrival (DOA) angle value; the DOA angle value of each individual is substituted into the spatial spectrum function value obtained by the MUSIC algorithm; and this value is used as the original fitness value of each individual. .

[0014] Preferably, the specific method for generating the adjustment factor for each individual using the particle swarm optimization algorithm is as follows: the original fitness value of each individual is optimized and calculated using the particle swarm optimization algorithm to generate the adjustment factor for each individual. Among them, adjustment factor The final value is constrained by the preset lower bound of the parameters. and parameter upper bound Within the closed interval formed, the following must be satisfied: .

[0015] S3. Fitness Adjustment and Genetic Operations: The original fitness values ​​of the corresponding individuals are corrected using the adjustment factor to obtain the adjusted fitness values. Based on the adjusted fitness values, the selection, crossover, and mutation operations of the genetic algorithm are performed to generate a new generation of population.

[0016] Preferably, the method for correcting the original fitness value of the corresponding individual using the adjustment factor to obtain the adjusted fitness value is as follows: The adjustment factor for each individual is... Its corresponding original fitness value Through formula The adjusted fitness value of the individual is obtained through calculation. .

[0017] Preferably, the method for generating a new generation population is as follows: the selection probability of each individual is obtained by dividing the adjusted fitness value of each individual by the sum of the adjusted fitness values ​​of all individuals in the current population, and a roulette wheel selection mechanism is used to select several individuals to form the parent population based on the selection probability of each individual.

[0018] Each individual in the parent population undergoes a simulated binary crossover operation with a preset crossover probability. An intermediate population is generated by exchanging some gene values ​​among individuals in the parent population. Each individual in the intermediate population undergoes a polynomial mutation operation with a preset mutation probability, and its encoded value is randomly perturbed according to a polynomial distribution to obtain the offspring population. The offspring population is then used as the new generation population.

[0019] S4. Underwater direction of arrival estimation: Repeat steps S2 to S3 until the termination condition is met, and output the direction of arrival estimation results of the underwater target based on the final population.

[0020] Preferably, the termination condition is: when the number of iterations of the genetic algorithm reaches the preset maximum number of generations or when the adjusted fitness value of the individual with the highest fitness value in each generation of the population reaches the preset satisfaction threshold, the termination condition is met and the algorithm iteration stops.

[0021] Preferably, the method for estimating the direction of arrival of the underwater target based on the final population output is as follows: after the termination condition is met, select the individual with the highest adjusted fitness value in the final population, decode it into the corresponding direction of arrival angle value, and output the direction of arrival angle value as the final estimation result of the underwater target's direction of arrival.

[0022] The beneficial effects of the present invention are as follows: (1) The first part of the present invention: by introducing the ant colony algorithm to optimize the initial population of the genetic algorithm, the quality of the initial distribution of the population is effectively improved. The initial path set constructed by the ant colony algorithm based on pheromone feedback gives the population a more reasonable structure and guidance, providing a better starting point for subsequent genetic operations, thereby improving the overall search efficiency and global coverage of the algorithm from the source.

[0023] (2) The second part of the present invention: The adjustment factor is dynamically generated by the particle swarm algorithm and applied to the correction of the original fitness value calculated based on the MUSIC algorithm, thereby realizing the adaptive control of selection pressure. This mechanism gives the algorithm a more intelligent ability to balance global exploration and local development, significantly enhances the guidance and stability of the search process in complex scenarios such as low signal-to-noise ratio, and effectively improves the accuracy and robustness of the convergence process.

[0024] (3) The third part of the present invention: The entire fusion improvement scheme is finally coordinated with the high-resolution MUSIC algorithm to form an efficient and reliable underwater direction of arrival estimation process. This method gives full play to the powerful search capability of the hybrid intelligent optimization algorithm, ensuring that the direction of arrival estimation results can be output stably and accurately in complex acoustic environments, effectively improving the practicality and anti-interference performance of the overall system. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 This is a schematic diagram of the method flow of the present invention. Detailed Implementation

[0027] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] Reference Figure 1 As shown, this invention provides a direction-of-arrival estimation method that integrates an improved genetic algorithm and MUSIC, including S1. Initial population optimization: Optimizing the initial population of the genetic algorithm using an ant colony algorithm.

[0029] In a specific embodiment of the present invention, the optimization of the initial population of the genetic algorithm using the ant colony algorithm is specifically performed as follows: S1-1: Constructing the path of the initial population.

[0030] S1-2: Update the pheromone concentration on the initial group path.

[0031] In a specific embodiment of the present invention, the method for constructing the initialization path of the ant colony is as follows: a probabilistic model based on the ant colony algorithm is used to select a path for each ant, wherein the ant... From the current city When selecting a city The probability of it being the next city visited is ,in Ants The city you are currently in , It is a positive integer greater than 2. This indicates the ant's number. , It is a positive integer greater than 2. Indicates the candidate cities to be selected. , Ants From the city you are currently in Cities that are directly accessible and have already been visited in this round A collection of cities outside of China It is a heuristic message with a value of , The current city where the ant is located and candidate cities to be selected The distance between, This indicates the current city where the ant is located. and candidate cities to be selected The pheromone concentration values ​​between them and To control the positive parameters of the relative weights of pheromones and heuristic information, For set Excluding candidate cities Any city other than [the city mentioned above].

[0032] Each ant starts from the starting city and iteratively selects the next city according to the above probability formula until all cities have been visited once, resulting in a complete path composed of the order in which the cities were visited. The set of complete paths constructed by all ants constitutes the initial population.

[0033] It should be noted that the visited city sequence in this round refers to the ant dynamically maintaining a list of visited cities to record its movement trajectory. This list is initially empty. After the ant starts from a certain starting city, that city is added to the list. Each time a new city is selected according to a probability formula, it is appended to the end of the list, thereby updating the list content in real time. This list is directly used to determine the optional subsequent cities at each step, that is, to exclude cities already existing in the list from the cities that can be directly reached from the current city. When the list contains all cities, it means that a complete path has been constructed, and the order recorded in the list is the ant's final path.

[0034] In a specific embodiment of the present invention, the method for updating the pheromone concentration on the initialization group path is as follows: the pheromone concentration on the initialization group path is updated using a global pheromone update formula, wherein the global pheromone update formula is: ,in , The total number of ants It is the evaporation rate of pheromones. , It is the first The pheromone increment added by an ant along the corresponding edges it traverses on the path it constructs in this round. Indicates the first The total length of the path constructed by an ant, i.e., the sequence of cities it has visited in this round. The sum of the lengths of all sides in the equation.

[0035] S2. Fitness Assessment and Factor Generation: The initial fitness value of each individual in the initial population is calculated using the MUSIC algorithm, and the adjustment factor of each individual is generated using the particle swarm optimization algorithm.

[0036] In a specific embodiment of the present invention, the method for calculating the original fitness value of each individual in the initial population using the MUSIC algorithm is as follows: each individual in the initial population is decoded into its corresponding direction-of-arrival (DOA) angle value; the DOA angle value of each individual is substituted into the spatial spectrum function value obtained by the MUSIC algorithm, and this value is used as the original fitness value of each individual. .

[0037] In one specific embodiment, each individual in the initial population is decoded into a corresponding direction-of-arrival (DOA) angle value. The specific decoding method is as follows: according to the genetic algorithm encoding rules adopted, the solution space coordinates represented by each individual are directly converted into DOA angle values. When real number encoding is used to define the individuals of the genetic algorithm, each real number value that makes up the individual itself represents a DOA angle to be estimated. This decoding process establishes the correspondence between the genetic algorithm optimization variables and the physical estimation parameters of the MUSIC algorithm.

[0038] For example, the initial fitness value of each individual, such as [45.2], [92.7], and [123.5], if real-number encoding is used and the target to be estimated is a single azimuth angle, the initial population may contain three individuals. During decoding, the real-number value of each individual is directly used as the direction-of-arrival (DOA) value, that is, converted into physical angles of 45.2°, 92.7°, and 123.5° respectively. These DOA values ​​are then input into the spatial spectrum function of the MUSIC algorithm. Perform calculations, where The spectral function value is obtained from the direction of arrival angle. , and These are the original fitness values ​​of the corresponding individuals.

[0039] In a specific embodiment of the present invention, the method for generating the adjustment factor for each individual using the particle swarm optimization algorithm is as follows: the original fitness value of each individual is optimized and calculated using the particle swarm optimization algorithm to generate the adjustment factor for each individual. Among them, adjustment factor The final value is constrained by the preset lower bound of the parameters. and parameter upper bound Within the closed interval formed, the following must be satisfied: .

[0040] It should be noted that the particle swarm optimization algorithm is a swarm intelligence optimization algorithm in the prior art. It simulates the social behavior of flocks of birds or schools of fish and searches for the optimal solution by iteratively updating the position and velocity of particles in the solution space. In this invention, the algorithm is used to perform optimization calculations in a preset search space based on the original fitness value of each individual, thereby dynamically generating an optimal adjustment factor for each individual. This algorithm is relatively mature and will not be described in detail here.

[0041] It should be noted that the lower bound of the preset parameters... and parameter upper bound The closed interval formed As the search space of the particle swarm optimization algorithm, it is used to ensure that the value of the adjustment factor is within a reasonable range, so as to avoid the algorithm performance being unstable due to the value being too large or too small.

[0042] S3. Fitness Adjustment and Genetic Operations: The original fitness values ​​of the corresponding individuals are corrected using the adjustment factor to obtain the adjusted fitness values. Based on the adjusted fitness values, the selection, crossover, and mutation operations of the genetic algorithm are performed to generate a new generation of population.

[0043] In a specific embodiment of the present invention, the method for correcting the original fitness value of the corresponding individual using the adjustment factor to obtain the adjusted fitness value is as follows: The adjustment factor for each individual is... Its corresponding original fitness value Through formula The adjusted fitness value of the individual is obtained through calculation. .

[0044] For example, if an individual's initial fitness value is 0.75, and its adjustment factor obtained through particle swarm optimization is 1.2, then according to the formula... The calculated fitness value of this individual is 0.90.

[0045] In a specific embodiment of the present invention, the method for generating a new generation population is as follows: the selection probability of each individual is obtained by dividing the adjusted fitness value of each individual by the sum of the adjusted fitness values ​​of all individuals in the current population, and a roulette wheel selection mechanism is used to select several individuals based on the selection probability of each individual to form the parent population.

[0046] Each individual in the parent population undergoes a simulated binary crossover operation with a preset crossover probability. An intermediate population is generated by exchanging some gene values ​​among individuals in the parent population. Each individual in the intermediate population undergoes a polynomial mutation operation with a preset mutation probability, and its encoded value is randomly perturbed according to a polynomial distribution to obtain the offspring population. The offspring population is then used as the new generation population.

[0047] In one specific embodiment, a roulette wheel selection mechanism is used to select several individuals based on the selection probability of each individual. The specific selection method is as follows: Based on the selection probability of each individual, a probability accumulation interval from 0 to 1 is constructed, where the length of the sub-interval corresponding to each individual is proportional to its selection probability. By generating a random number uniformly distributed in the range [0,1), the individual whose probability sub-interval the random number falls into is selected. The above process of generating random numbers and matching is repeated until the same number of individuals as the parent population size are selected and the selection ends. This selection process is based on the selection probability of each individual. The higher the probability value of an individual, the greater the probability that it will be selected by the random number, thereby ensuring that individuals with better fitness have a higher chance of being retained in the parent population.

[0048] The simulated binary crossover operation is the crossover operator used in existing real-number encoded genetic algorithms. It controls the proximity of the offspring and parent values ​​through a distribution index and calculates the gene values ​​of the two parent individuals based on a specific mathematical formula to generate new values ​​for the offspring, thereby achieving directional search and effective development in the solution space.

[0049] It should be noted that the intermediate population refers to the set of individuals in the intermediate state of the parent population after crossover but before mutation.

[0050] The polynomial mutation operation is a mutation operator in the existing real-number encoded genetic algorithm. Based on the probability model of the polynomial distribution, it applies a random perturbation of controllable magnitude to the individual encoded value. This mechanism can generate small perturbations with a high probability to achieve local fine search, while introducing large perturbations with a low probability to promote global exploration. This effectively maintains population diversity, avoids premature convergence, and significantly enhances the algorithm's global search capability and robustness in complex solution spaces.

[0051] It should be noted that simulated binary crossover will be performed on individuals in the parent population according to a preset crossover probability. By exchanging some gene values ​​of two parent individuals, new individuals will be generated. The purpose is to effectively combine the strengths of different individuals by exchanging high-quality gene fragments of parent individuals, guiding the search direction and enabling the algorithm to explore advantageous regions in the solution space more specifically, thereby significantly accelerating the speed of approaching the optimal solution. After crossover, a multinomial mutation operation will be performed on the individuals according to a preset mutation probability. That is, the encoded value of the individuals will be randomly perturbed according to a multinomial distribution. The purpose is to introduce new genetic diversity into the population, help the algorithm escape from possible local optima, and enhance the algorithm's global exploration capability in the solution space. These two operations work together to ensure that the genetic algorithm can take into account both depth mining and breadth exploration during the iteration process, thereby effectively improving the convergence speed, accuracy, and robustness of underwater target direction of arrival estimation.

[0052] S4. Underwater direction of arrival estimation: Repeat steps S2 to S3 until the termination condition is met, and output the direction of arrival estimation results of the underwater target based on the final population.

[0053] In a specific embodiment of the present invention, the termination condition is: when the number of iterations of the genetic algorithm reaches the preset maximum number of generations or when the adjusted fitness value of the individual with the highest fitness value in each generation of the population reaches the preset satisfaction threshold, the termination condition is met and the algorithm iteration stops.

[0054] It should be noted that setting the maximum number of iterations of the genetic algorithm as the termination condition is to prevent the algorithm from getting stuck in an infinite loop or running meaninglessly for a long time when it cannot find a better solution, and to ensure that computing resources are used effectively within a controllable time.

[0055] In a specific embodiment of the present invention, the method for estimating the direction of arrival of an underwater target based on the final population output is as follows: after the termination condition is met, select the individual with the highest adjusted fitness value in the final population, decode it into the corresponding direction of arrival angle value, and output the direction of arrival angle value as the final estimation result of the underwater target's direction of arrival.

[0056] It should be noted that during the iteration process of the genetic algorithm, the direction-of-arrival angle value obtained after decoding each individual is a candidate solution to be evaluated. When the genetic algorithm terminates, the direction-of-arrival angle value obtained from the individual with the highest fitness value after adjustment from the final population is the final estimated result.

[0057] The examples described in this invention are not limited to the specific embodiments listed above. The examples are merely illustrative to facilitate understanding of the invention and do not constitute a limitation on the scope of protection of this invention. Any modifications, equivalent substitutions, etc., made within the spirit and principles of this invention should be included within the scope of protection.

[0058] The above description is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined in this specification, they should all fall within the protection scope of the present invention.

Claims

1. A direction-of-arrival estimation method integrating an improved genetic algorithm and MUSIC, characterized in that, Includes the following steps: S1. Initial Population Optimization: Optimize the initial population of the genetic algorithm using the ant colony algorithm; S2. Fitness Assessment and Factor Generation: The initial fitness value of each individual in the initial population is calculated using the MUSIC algorithm, and the adjustment factor of each individual is generated using the particle swarm optimization algorithm. S3. Fitness Adjustment and Genetic Operations: The original fitness values ​​of the corresponding individuals are corrected using the adjustment factors to obtain adjusted fitness values. Based on the adjusted fitness values, selection, crossover, and mutation operations of the genetic algorithm are performed to generate a new generation of population. S4. Underwater direction of arrival estimation: Repeat steps S2 to S3 until the termination condition is met, and output the direction of arrival estimation results of the underwater target based on the final population.

2. The direction-of-arrival estimation method integrating an improved genetic algorithm and MUSIC as described in claim 1, characterized in that, The optimization of the initial population of the genetic algorithm using the ant colony algorithm involves the following steps: S1-1: Path for constructing the initial population; S1-2: Update the pheromone concentration on the initial group path.

3. The direction-of-arrival estimation method integrating an improved genetic algorithm and MUSIC as described in claim 2, characterized in that, The specific method for constructing the initialization group path is as follows: The probabilistic model based on the ant colony algorithm selects a path for each ant, where the ant... From the current city When selecting a city The probability of it being the next city visited is ,in Ants The city you are currently in , It is a positive integer greater than 2. This indicates the ant's number. , It is a positive integer greater than 2. Indicates the candidate cities to be selected. , Ants From the city you are currently in Cities that are directly accessible and have already been visited in this round A collection of cities outside of China It is a heuristic message with a value of , The current city where the ant is located and candidate cities to be selected The distance between, This indicates the current city where the ant is located. and candidate cities to be selected The pheromone concentration values ​​between them and To control the positive parameters of the relative weights of pheromones and heuristic information, For set Excluding candidate cities Any other available city; Each ant starts from the starting city and iteratively selects the next city according to the above probability formula until all cities have been visited once, resulting in a complete path composed of the order in which the cities were visited. The set of complete paths constructed by all ants constitutes the initial population.

4. The direction-of-arrival estimation method integrating an improved genetic algorithm and MUSIC as described in claim 2, characterized in that, The specific method for updating the pheromone concentration on the initialization group path is as follows: The pheromone concentration on the initial population path is updated using a global pheromone update formula, which is as follows: ,in , The total number of ants It is the evaporation rate of pheromones. , It is the first The pheromone increment added by an ant along the corresponding edges it traverses on the path it constructs in this round. Indicates the first The total length of the path constructed by an ant, i.e., the sequence of cities it has visited in this round. The sum of the lengths of all sides in the equation.

5. The direction-of-arrival estimation method integrating an improved genetic algorithm and MUSIC as described in claim 1, characterized in that, The specific method for calculating the initial fitness value of each individual in the initial population using the MUSIC algorithm is as follows: Each individual in the initial population is decoded into its corresponding direction-of-arrival (DOA) value. This DOA value is then substituted into the spatial spectrum function obtained from the MUSIC algorithm, and used as the initial fitness value for each individual. .

6. The direction-of-arrival estimation method integrating an improved genetic algorithm and MUSIC as described in claim 5, characterized in that, The adjustment factor for each individual is generated using the particle swarm optimization algorithm, and the specific adjustment method is as follows: The original fitness value of each individual is optimized using the particle swarm optimization algorithm to generate an adjustment factor for each individual. Among them, adjustment factor The final value is constrained by the preset lower bound of the parameters. and parameter upper bound Within the closed interval formed, the following must be satisfied: .

7. The direction-of-arrival estimation method integrating an improved genetic algorithm and MUSIC as described in claim 6, characterized in that, The method for correcting the original fitness value of the corresponding individual using the adjustment factor to obtain the adjusted fitness value is as follows: Adjustment factors for each individual Its corresponding original fitness value Through formula The adjusted fitness value of the individual is obtained through calculation. .

8. The direction-of-arrival estimation method integrating an improved genetic algorithm and MUSIC as described in claim 7, characterized in that, The specific method for generating a new generation of population is as follows: The selection probability of each individual is obtained by dividing the adjusted fitness value of each individual by the sum of the adjusted fitness values ​​of all individuals in the current population. Based on the selection probability of each individual, a roulette wheel selection mechanism is used to select a number of individuals to form the parent population. Each individual in the parent population undergoes a simulated binary crossover operation with a preset crossover probability. An intermediate population is generated by exchanging some gene values ​​among individuals in the parent population. Each individual in the intermediate population undergoes a polynomial mutation operation with a preset mutation probability, and its encoded value is randomly perturbed according to a polynomial distribution to obtain the offspring population. The offspring population is then used as the new generation population.

9. The direction-of-arrival estimation method integrating an improved genetic algorithm and MUSIC as described in claim 1, characterized in that, The termination condition is as follows: when the number of iterations of the genetic algorithm reaches the preset maximum number of generations or when the adjusted fitness value of the individual with the highest fitness value in each generation of the population reaches the preset satisfaction threshold, the termination condition is met and the algorithm iteration stops.

10. The direction-of-arrival estimation method integrating an improved genetic algorithm and MUSIC as described in claim 9, characterized in that, The specific method for estimating the direction of arrival (DOA) of underwater targets based on the final population output is as follows: After the termination condition is met, select the individual with the highest adjusted fitness value in the final population, decode it into the corresponding direction of arrival angle value, and output the direction of arrival angle value as the final estimate of the underwater target's direction of arrival.

Images