A method for partial fault tolerant channel adaptive beam keeping

Abstract

This invention discloses an adaptive beam-maintaining method for de-localized fault channels, which acquires fault channel information of the radar array through internal and external radar monitoring. An evolutionary objective is established, and the fault channel information is used to create a first-generation population. The gene values ​​corresponding to the fault channels of all individuals in the first-generation population are set to zero. A fitness function is established based on array characteristics to evaluate the fitness of the first-generation population. A selection zone is set, and parent and parent individuals are randomly selected within the selection zone. Gene breakpoints are also randomly selected, and gene segments of the parent and parent individuals are exchanged to generate a pair of offspring individuals. Gene mutation factors, maximum gene mutation ranges, and gene mutation ranges are set to generate a corresponding number of offspring individuals. The offspring population and the parent population are then evaluated for fitness and ranked together, retaining the specified number of individuals for the population. The evolutionary process is repeated until a specified number of evolutions is reached or the beam result corresponding to the best individual fully conforms to the evolutionary objective.

Description

Technical Field

[0001] This invention relates to antenna and microwave technology, and in particular to an adaptive beam-holding method for de-localized faulting channels. Background Technology

[0002] Phased array radars, due to their unique advantages such as multiple functions, high mobility, short reaction time, high data rate, strong anti-jamming capability, and high reliability, have been fully utilized in various military and civilian fields. In the field of weather radar, phased array radars can significantly reduce radar scanning time, greatly improving the early warning capability for short-term, small-scale convective weather. This makes phased array radars widely favored in the meteorological field.

[0003] However, meteorological detection requires knowledge of the radar's beamwidth, which necessitates maintaining a stable beamwidth and relatively high sidelobe suppression during operation. While the high reliability of phased array radars allows them to operate stably even with partial channel failures, damage to these channels inevitably leads to beam degradation. How to eliminate the limitations imposed on meteorological applications by this beam degradation under partial channel failure conditions is a pressing issue that needs to be addressed.

[0004] Patent application number 202011044199.X discloses a beamforming simulation design method based on a genetic algorithm. This method achieves a 40° wide beam cosecant square beamforming pattern by iteratively optimizing the formation of a target beam confined between two constraint lines.

[0005] However, this optimization method cannot be used when there are faulty channels on the array, and it cannot meet the meteorological requirements for maintaining a stable beamwidth and relatively high sidelobe suppression when some channels on the array are faulty. Summary of the Invention

[0006] To address the problems existing in the prior art, this invention provides a method for adaptive beam-keeping algorithm that incorporates faulty channels into the beamforming algorithm to form a localized faulty channel adaptive beam-keeping algorithm. This method adaptively optimizes the optimal amplitude and phase of the remaining normal channels by using faulty channel information fed back from the radar monitoring link, thereby removing the influence of faulty channels and maintaining the stability of indicators such as beamwidth and sidelobe suppression, thus meeting the meteorological field's requirement for stable beam shape.

[0007] The objective of this invention is achieved through the following technical solutions.

[0008] A channel-adaptive beam-holding method for de-localized faulting includes the following steps:

[0009] Step 1: Obtain information on radar array fault channels through internal and external radar monitoring methods;

[0010] Step 2: Establish evolutionary objectives, which specify the maximum and minimum values ​​of the radar beam.

[0011] Step 3: Substitute the fault channel information to establish the first-generation population, and set the gene value corresponding to the fault channel of all individuals in the first-generation population to zero; the first-generation population is represented as... ,in This represents the gene value corresponding to the amplitude and phase information of each channel. The gene value corresponding to the faulty channel is set to zero, i.e., the corresponding... The population size is , It is the first Each individual has a number of genes. ;

[0012] Step 4: Establish a fitness function based on array features ,in and These represent the upper and lower limits of the evolutionary goal. The results of array radiation beams calculated for individuals in the population;

[0013] Step 5: Evaluate the fitness of the first generation population, score and rank each individual;

[0014] Step 6: Set up a selection zone, randomly select the father and mother individuals from the selection zone, and randomly select gene breakpoints. Exchange the gene segments of the father and mother individuals to generate a pair of offspring individuals.

[0015] Step 7: Set up gene mutation factors Maximum range of gene mutations Range of gene variations ,in, This is the evolution time; during this process, the number and location of randomly mutated genes are set to zero, that is, the gene corresponding to the faulty channel is locked so that it does not mutate during evolution;

[0016] Step 8: Generate the corresponding number of offspring individuals according to the number of pairings, and score and sort the fitness of the offspring population and the parent population together. Select and retain only the number of individuals specified for the population size, which is the population size in this case. ;

[0017] Step 9: Repeat the evolution process until the specified number of evolutions is reached or the beam result corresponding to the best individual fully meets the evolution goal;

[0018] Step 10: Verify whether the channel amplitude and phase obtained by the algorithm meet the beam-keeping requirements.

[0019] The above completes the adaptive beam-keeping algorithm for the localized fault-free channel. This algorithm effectively maintains the beam generated by the array without replacing array components, within a certain failure rate range. In the application of weather radar, it ensures high reliability, high stability, and high effectiveness of the phased array radar in a cost-effective manner.

[0020] The most important information about the fault channels in step one is the location of the fault channels. All fault channels do not radiate energy during use.

[0021] In step two, the evolution target can constrain indicators such as beamwidth and sidelobes. The evolution target is based on the indicator requirements. The main purpose of the evolution is to enable the beam to achieve high sidelobe suppression while keeping the beamwidth constant, or to enable the beam to achieve the highest sidelobe suppression and provide feedback on the beamwidth.

[0022] In step three, the fault channel information is substituted to establish the first-generation population. , The information is obtained from the fault channel information acquired in step one, and the number of genes is... The population size is determined by the number of channels in the array. The settings are determined by the required precision and speed of the algorithm.

[0023] In step four, a fitness function is established to optimize the target upper bound. and lower limit and radiation beam results The number of calculation points is the same, and they are mapped one-to-one with the spatial angles.

[0024] In step six, a selection zone is set up. The selection zone is the range of parents to be selected. The optimal range is defined according to the fitness score, and the number of gene breakpoints is adjusted. That is, the position and number of gene fragment exchanges between parents are carried out at multiple random positions.

[0025] In step seven, gene mutation factors Maximum range of gene variation These are all constants, which determine the degree of mutation. The range of gene change is a quantity that decreases over time, and the evolution time corresponds to the number of evolutions in the algorithm.

[0026] In step ten, it is necessary to verify whether the channel amplitude and phase obtained by the algorithm meet the beam-keeping requirements. The channel amplitude and phase results obtained by the algorithm must meet the amplitude and phase quantization requirements of each radar channel. The results are then applied to the actual radar array, and the beam results after amplitude and phase quantization of each channel are verified by electromagnetic simulation software.

[0027] Compared with the prior art, the advantages of the present invention are: 1. The array beam holding function is completed by the local fault-free channel adaptive algorithm within the array failure rate range, which ensures the high reliability, high stability and high effectiveness of the array in a low-cost manner.

[0028] 2. The array beam can be maintained and optimized simply by using the fault channel information fed back by the internal and external monitoring functions. It can be applied in various fault rate ranges and has a wide range of applicable faults.

[0029] 3. During beam holding, the beamwidth of the array beam remains unchanged and the sidelobes are kept highly suppressed, fully meeting the application requirements of weather radar. Attached Figure Description

[0030] Figure 1 This is a flowchart of the algorithm of the present invention.

[0031] Figure 2 This describes the optimization process and results for the 3% channel random fault in an embodiment of the present invention.

[0032] Figure 3 This is a beam comparison of 1% channel random fault and no fault in the embodiment of the present invention.

[0033] Figure 4 This is a beam comparison between 3% channel random faults and no faults in an embodiment of the present invention.

[0034] Figure 5 This is a comparison of the 5% channel random fault case 1 and the fault-free beam in Embodiment 1 of the present invention.

[0035] Figure 6 This is a comparison of the 5% channel random fault case 2 and the fault-free beam in Embodiment 2 of the present invention.

[0036] Figure 7 This is a beam comparison of channel random faults and repair faults in embodiment 1 of the present invention.

[0037] Figure 8 The amplitude and phase distribution results of each channel after the random fault repair of 1% channel in the embodiment of the present invention are shown.

[0038] Figure 9 This is a beam comparison of random faults and repair faults in the 3% channel of the embodiment of the present invention.

[0039] Figure 10 The amplitude and phase distribution results of each channel after the random fault repair of 3% channel in the embodiment of the present invention are shown.

[0040] Figure 11 This is a beam comparison of the 5% channel random fault case 1 and the fault repair case in Embodiment 1 of the present invention.

[0041] Figure 12 The amplitude and phase distribution results of each channel after repair in Example 1 of the present invention, which shows the 5% channel random fault situation.

[0042] Figure 13 This is a beam comparison of scenario 2 with a 5% channel random fault and a fault repair in embodiment 2 of the present invention.

[0043] Figure 14 The amplitude and phase distribution results of each channel after repair in Example 2 of the present invention, which shows the 5% channel random fault situation.

[0044] Figure 15 This is a comparison of beams used to repair faults under all fault conditions in the embodiments of the present invention. Detailed Implementation

[0045] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0046] A channel-adaptive beam-holding method for de-localizing faults includes the following design steps:

[0047] Step 1: Obtain information on radar array fault channels through internal and external radar monitoring methods.

[0048] Step 2: Establish evolutionary objectives, which specify the maximum and minimum values ​​of the radar beam.

[0049] Step 3: Substitute the faulty channel information to establish the first-generation population, and set the gene value corresponding to the faulty channel of all individuals in the first-generation population to zero. The first-generation population is represented as follows: ,in This represents the gene value corresponding to the amplitude and phase information of each channel. The gene value corresponding to the faulty channel is set to zero, i.e., the corresponding... The population size is , It is the first Each individual has a number of genes. .

[0050] Step 4: Establish a fitness function based on array features ,in and These represent the upper and lower limits of the evolutionary goal. The results of array radiation beams calculated for individuals in the population.

[0051] Step 5: Evaluate the fitness of the first generation population, score and rank each individual.

[0052] Step 6: Set up a selection zone, randomly select the father and mother individuals from the selection zone, and randomly select gene breakpoints. Exchange the gene segments of the father and mother individuals to generate a pair of offspring individuals.

[0053] Step 7: Set up gene mutation factors Maximum range of gene mutations Range of gene variations ,in, This refers to the evolutionary time. It also involves the number and location of randomly mutated genes. During this process, the mutation amount at the gene location corresponding to the faulty channel is set to zero. In other words, the gene corresponding to the faulty channel is locked, preventing it from mutating during evolution.

[0054] Step 8: Generate the corresponding number of offspring individuals according to the number of pairings, and score and sort the fitness of the offspring population and the parent population together. Select and retain only the number of individuals specified for the population size, which is the population size in this case. .

[0055] Step 9: Repeat the evolution process until the specified number of evolutions is reached or the beam result corresponding to the best individual fully meets the evolution goal.

[0056] Step 10: Verify whether the channel amplitude and phase obtained by the algorithm meet the beam-keeping requirements.

[0057] In step one, radar array fault channel information is obtained through internal and external radar monitoring. The most important information about the fault channel is its location. All fault channels do not radiate energy during use.

[0058] In step two, the evolution objective can constrain indicators such as beamwidth and sidelobes. The evolution objective is based on the indicator requirements. The main goal of the evolution can be to achieve high sidelobe suppression while keeping the beamwidth constant, or to ensure the beam has the highest sidelobe suppression and provide feedback on the beamwidth (the degree of sidelobe suppression has a limit, which changes according to the proportion of faulty channels).

[0059] In step three, the fault channel information is substituted to establish the first-generation population. , The information is obtained from the fault channel information acquired in step one, and the number of genes is... The population size is determined by the number of channels in the array. The settings are determined by the required precision and speed of the algorithm.

[0060] In step four, a fitness function is established to optimize the target upper bound. and lower limit and radiation beam results The number of calculation points is the same, and they are mapped one-to-one with the spatial angles.

[0061] In step six, the selection zone is set up. The selection zone is the range of parents to be selected. The preferred range is defined according to the fitness score. This range can be adjusted.

[0062] In step six, gene breakpoints are randomly selected: the number of gene breakpoints can be adjusted, meaning that the location and number of gene fragment exchanges between parents can be performed at multiple random locations.

[0063] Step seven involves setting gene mutation factors. Maximum range of gene mutations Range of gene variations ,in, It's about evolutionary time. Genetic mutation factors. Maximum range of gene variation These are all constants and can be adjusted, determining the degree of mutation. The range of gene variation is a quantity that decreases over time, with the evolution time corresponding to the number of evolutions in the algorithm.

[0064] In step seven, the mutation value of the gene corresponding to the faulty channel is set to zero. By setting the mutation value of the corresponding gene to zero using the faulty channel information obtained in step one, it is ensured that the gene value corresponding to the faulty channel is always zero during the evolutionary process and does not participate in the evolutionary process.

[0065] Step nine continues until the specified number of evolutions is reached or the beam result corresponding to the best individual fully meets the evolution goal. The evolution result is not unique, and this step can be performed multiple times to obtain multiple different evolution results.

[0066] Step 10 verifies whether the channel amplitude and phase obtained by the algorithm meet the beam-keeping requirements. This requires ensuring that the channel amplitude and phase results obtained by the algorithm meet the amplitude and phase quantization requirements of each radar channel, making it applicable to actual radar arrays. The beam results after amplitude and phase quantization of each channel can be verified using electromagnetic simulation software. Therefore, steps 3 to 9 can be performed multiple times to verify and select the optimal result from the evolution results in step 9.

[0067] All steps can be applied to all random channel failure scenarios within the normal operating range of the array.

[0068] All these steps can be applied not only to 1D linear arrays but also to 2D area arrays. In practice, it is only necessary to perform all the steps described in claim 1 in each of the two dimensions.

[0069] Example 1

[0070] like Figure 1 As shown, an adaptive beam-holding method for de-localized channel failures includes the following steps. This embodiment uses a linear array composed of 160 elements as an example. The requirements are to maintain a beamwidth of 1 degree under a maximum of 5% random channel failure, and to have the highest possible sidelobe suppression. The embodiments are set to 1% random channel failure, 3% random channel failure, and 5% random channel failure in both cases:

[0071] Step 1: Obtain radar array fault channel information through internal and external monitoring methods; in this embodiment, 1% random fault channels are channels 18 and 71; 3% random fault channels are channels 37, 59, 64, 74 and 122; 5% random fault channel case 1 is channels 15, 65, 90, 94, 117, 118, 128 and 156; 5% random fault channel case 2 is channels 5, 30, 32, 53, 83, 86, 108 and 158.

[0072] Step 2: Establish an evolution target, which specifies the maximum and minimum values ​​of the radar beam. In this embodiment, the maximum beam value within ±0.5° is defined as 0 dB, and the minimum beam value is defined as -3 dB.

[0073] Step 3: Substitute the faulty channel information to establish the first-generation population, and set the gene value corresponding to the faulty channel of all individuals in the first-generation population to zero. The first-generation population is represented as follows: ,in This represents the gene value corresponding to the amplitude and phase information of each channel. The gene value corresponding to the faulty channel is set to zero, i.e., the corresponding... In this embodiment, the gene value at the corresponding fault channel location in step one is set to zero. The population size is... , It is the first Each individual has a number of genes. .

[0074] Step 4: Establish evaluation functions for 160 units ,in and These represent the upper and lower limits of the evolutionary goal. The results of array radiation beams calculated for individuals in the population.

[0075] Step 5: Evaluate the fitness of the first generation population, score and rank each individual.

[0076] Step 6: Set the selection zone to 10, randomly select the father and mother individuals from the top 10 individuals, randomly set gene breakpoints, and exchange gene segments between the father and mother individuals to generate a pair of offspring individuals.

[0077] Step 7: Set up gene mutation factors Maximum range of gene mutations Range of gene variations ,in, This refers to the evolutionary time. It also involves the number and location of randomly mutated genes. During this process, the mutation amount at the gene location corresponding to the faulty channel is set to zero. In other words, the gene corresponding to the faulty channel is locked, preventing it from mutating during evolution.

[0078] Step 8: Generate the corresponding number of offspring individuals based on the number of pairings, set to 25. Perform fitness scoring and sorting on both the offspring and parent populations. Select and retain only the number of individuals specified for the population size (i.e., the total population size). .

[0079] Step Nine: Repeat the evolution process until the specified number of evolutions is reached or the beamforming result corresponding to the best individual fully meets the evolutionary objective. The optimization process and results for the 3% channel random failure are as follows: Figure 2 As shown.

[0080] Step 10: Verify whether the channel amplitude and phase obtained by the algorithm meet the beam-keeping requirements.

[0081] Figure 3 The beamwidth changes from 1° to 0.99° and the sidelobe suppression deteriorates from 40 dB to 32.5 dB when there is 1% random channel fault and no fault. Figure 4 The beamwidth changes from 1° to 0.98° and the sidelobe suppression deteriorates from 40 dB to 26.9 dB when there is a 3% channel random fault and no fault. Figure 5 The comparison between the beamwidth in case 1 (5% channel random failure) and the beamwidth in case 2 (no failure) is shown. In this case, the beamwidth changes from 1° to 0.99° and the sidelobe suppression deteriorates from 40 dB to 26.0 dB. Figure 6 The diagram shows a comparison between a 5% channel random failure scenario (2) and a fault-free beam. In this scenario, the beamwidth changes from 1° to 0.98°, and the sidelobe suppression deteriorates from 40 dB to 25.16 dB. It is evident that as the number of channel failures increases, the sidelobe suppression significantly worsens.

[0082] Figure 7 The image shows a beamwidth comparison between a 1% random channel fault and the fault-corrected beam. After correction, the beamwidth is 1°, and the sidelobe suppression is 35 dB. The optimized amplitude and phase distribution results for each channel are shown below. Figure 8 As shown. Figure 9 The image shows a beamwidth comparison after a 3% random channel fault and after fault repair. After repair, the beamwidth is 1°, and the sidelobe suppression is 30 dB. The optimized amplitude and phase distribution results for each channel are shown below. Figure 10 As shown. Figure 11 The image shows a beamwidth comparison between case 1 (5% random channel failure) and the case after failure repair. After repair, the beamwidth is 1°, and the sidelobe suppression is 30 dB. The optimized amplitude and phase distribution results for each channel are shown below. Figure 12 As shown. Figure 13 The image shows a beamwidth comparison between case 2 (5% random channel failure) and the repaired beamwidth. After repair, the beamwidth is 1°, and the sidelobe suppression is 30 dB. The optimized amplitude and phase distribution results for each channel are shown below. Figure 14 As shown.

[0083] Beams repaired under various fault conditions, such as Figure 15 As shown, after each fault condition was repaired, the beamwidth remained highly consistent, and the sidelobes were also well suppressed while maintaining a highly consistent beamwidth.

Claims

1. A channel adaptive beamholding method for de-localized faulting, characterized in that... Includes the following steps: Step 1: Obtain information on radar array fault channels through internal and external radar monitoring methods; Step 2: Establish evolutionary objectives, which specify the maximum and minimum values ​​of the radar beam. Step 3: Substitute the fault channel information to establish the first-generation population, and set the gene value corresponding to the fault channel of all individuals in the first-generation population to zero; the first-generation population is represented as... ,in This represents the gene value corresponding to the amplitude and phase information of each channel. The gene value corresponding to the faulty channel is set to zero, i.e., the corresponding... The population size is , It is the first Each individual has a number of genes. ; Step 4: Establish a fitness function based on array features ,in and These represent the upper and lower limits of the evolutionary goal. The results of array radiation beams calculated for individuals in the population; Step 5: Evaluate the fitness of the first generation population, score and rank each individual; Step 6: Set up a selection zone, randomly select the father and mother individuals from the selection zone, and randomly select gene breakpoints. Exchange the gene segments of the father and mother individuals to generate a pair of offspring individuals. Step 7: Set up gene mutation factors Maximum range of gene mutations Range of gene variations ,in, This is the evolution time; during this process, the number and location of randomly mutated genes are set to zero, that is, the gene corresponding to the faulty channel is locked so that it does not mutate during evolution; Step 8: Generate the corresponding number of offspring individuals according to the number of pairings, and score and sort the fitness of the offspring population and the parent population together. Select and retain only the number of individuals specified for the population size, which is the population size in this case. ; Step 9: Repeat the evolution process until the specified number of evolutions is reached or the beam result corresponding to the best individual fully meets the evolution goal; Step 10: Verify whether the channel amplitude and phase obtained by the algorithm meet the beam-keeping requirements.

2. The adaptive beamholding method for de-localized channels according to claim 1, characterized in that... The most important information about the fault channels in step one is the location of the fault channels. All fault channels do not radiate energy during use.

3. The adaptive beam-holding method for de-localized channels according to claim 1, characterized in that... In step two, the evolution target can constrain indicators such as beamwidth and sidelobes. The evolution target is based on the indicator requirements. The main purpose of the evolution is to enable the beam to achieve high sidelobe suppression while keeping the beamwidth constant, or to enable the beam to achieve the highest sidelobe suppression and provide feedback on the beamwidth.

4. The adaptive beamholding method for de-localized channels according to claim 1, characterized in that... In step three, the fault channel information is used to establish the first-generation population. , The information is obtained from the fault channel information acquired in step one, and the number of genes is... The population size is determined by the number of channels in the array. The settings are determined by the required precision and speed of the algorithm.

5. The adaptive beamholding method for de-localized channels according to claim 1, characterized in that... In step four, a fitness function is established to optimize the target upper bound. and lower limit and radiation beam results The number of calculation points is the same, and they are mapped one-to-one with the spatial angles.

6. The adaptive beamholding method for de-localized channels according to claim 1, characterized in that... In step six, a selection zone is set up. The selection zone is the range of parents to be selected. The optimal range is defined according to the fitness score, and the number of gene breakpoints is adjusted. That is, the position and number of gene fragment exchanges between parents are carried out at multiple random positions.

7. The adaptive beamholding method for de-localized channels according to claim 1, characterized in that... In step seven, gene mutation factors Maximum range of gene variation These are all constants, which determine the degree of mutation. The range of gene change is a quantity that decreases over time, and the evolution time corresponds to the number of evolutions in the algorithm.

8. The adaptive beamholding method for de-localized channels according to claim 1, characterized in that... In step ten, it is necessary to verify whether the channel amplitude and phase obtained by the algorithm meet the beam-keeping requirements. The channel amplitude and phase results obtained by the algorithm must meet the amplitude and phase quantization requirements of each radar channel. The results are then applied to the actual radar array, and the beam results after amplitude and phase quantization of each channel are verified by electromagnetic simulation software.

Images