A distributed multi-node joint waveform design method to resist range Doppler shielding

Abstract

The invention discloses a distributed multi-node joint anti-distance Doppler shielding waveform design method, comprising the following steps: determining a clarity ratio standard, including a clutter clear area and a distance clear area; determining a distance detection interval #imgabs0#, a Doppler detection interval #imgabs1# and an optional PRF range #imgabs2#; extracting individual genes from the optional PRF range, and creating an initial population #imgabs3# from the individual genes; wherein a population #imgabs4# includes M individuals #imgabs5#, an individual #imgabs 6# includes N waveforms of different PRFs #imgabs7# as individual genes; iteratively optimize the initial population #imgabs8#, generate a new generation population #imgabs9#, set the number of iterations #imgabs10# as the termination condition of the genetic algorithm, and if the evolution is terminated, output the new population #imgabs11# as the final population #imgabs12#; output all #imgabs15# PRFs contained in the individual #imgabs14# with the highest fitness function value in the final population #imgabs13# to form a waveform design result. According to the above technical solution, the computational complexity can be reduced while ensuring a good anti-occlusion effect.

Description

Technical Field

[0001] The present invention relates to the field of distributed radar waveform design, and in particular to a distributed multi-node joint anti-range Doppler shielding waveform design method. Background Art

[0002] Pulse Repetition Frequency (PRF) is the number of pulses emitted by a radar or sonar system in one second. It is an important parameter of pulse radar signals, which affects the maximum detection range, resolution, and ability to distinguish multiple targets of the radar system. When a distributed multi-node radar system operates at a low pulse repetition frequency (LPRF), it is susceptible to a large amount of interference from ground and sea clutter, and the radar cannot effectively detect moving targets. When the radar operates at a high PRF, multiple range ambiguities are introduced. In order to achieve detection effects in the range and Doppler dimensions, most airborne phased array pulse Doppler radars use a medium pulse repetition frequency (MPRF). However, the use of MPRF will cause a certain degree of ambiguity in both range and Doppler, so the range-Doppler occlusion area will also expand, making it impossible for targets in the occlusion area to be correctly and effectively detected by the radar system.

[0003] In order to overcome the influence of the range-Doppler obstruction area on the radar detection range-Doppler range, the distributed multi-node radar system usually adopts the working mode of transmitting multiple groups of PRF (pulse repetition frequency). Although each PRF inevitably has a certain obstruction area, since the obstruction area of ​​each PRF is different, these PRFs can be reasonably selected to make their common obstruction area as small as possible.

[0004] If the public occlusion area is to be as small as possible, the double PRF method is often used in waveform design. With the double PRF, all possible combinations can be traversed and calculated to directly find the PRF combination that maximizes the clear area. In the design of the double PRF, the number of combinations will not be too many. However, in order to meet a certain clear area ratio, usually 3 or more PRFs are required to meet the detection requirements. At this time, the number of PRF combinations is large, and the calculation amount of traversing all PRF combinations is too large to be realized; therefore, a multi-pulse repetition frequency waveform design scheme with low calculation amount and meeting the clear area ratio requirements is needed to resist distance-Doppler occlusion. Summary of the invention

[0005] To achieve the above objectives, the present application provides a distributed multi-node joint anti-range Doppler shielding waveform design method, comprising the following steps:

[0006] Determine a clarity ratio standard and define a clear area, wherein the clear area includes a clutter clear area and a distance clear area;

[0007] Determine the distance detection interval , Doppler detection interval and the selectable PRF range , and discretize the PRF range evenly; wherein, the distance detection interval refers to the detection range determined according to the minimum value and the maximum value for the target distance R corresponding to a PRI; the Doppler detection interval refers to the detection range determined according to the minimum value and the maximum value for the Doppler detection range ; the PRF range refers to the frequency range determined according to the minimum value and the maximum value for the PRF;

[0008] Extract individual genes from the selectable PRF range and create an initial population from the individual genes ; wherein, a population includes M individuals , and an individual includes N waveforms of different PRFs as individual genes, where n is the waveform serial number, and: 0 < n < N; m is the individual serial number, and: 0 < m < N;

[0009] Iteratively optimize the initial population according to the clutter-free zone , and output a new population as the final population ;

[0010] The individuals with the highest fitness function in the final population contain all the PRFs that are the PRF combination with the largest clutter-free zone, and the PRF combination is the waveform design result. wherein, the clutter-free zone is a frequency region, expressed as:

[0011] , where , is the Doppler ambiguity multiplicity, is the clutter main lobe width; the method for determining the clutter main lobe width is:

[0012] , where the magnitude of the platform speed is , the radar wavelength is , the azimuth angle of the platform speed direction relative to the antenna beam center is , the pitch angle is , and the beam width is .

[0013] The distance clear zone is expressed as: ,in, is the pulse width, To protect time, is the pulse repetition interval, is the target distance corresponding to one PRI.

[0014] Furthermore, the distance clear area is converted to the distance blocked area, which is expressed as:

[0015] ,in, is the duty cycle, The speed of light.

[0016] Furthermore, the population is expressed as: ,

[0017] Individuals are represented by: ,

[0018] The waveforms of different PRFs are expressed as: .

[0019] Further, the initial population is iteratively optimized , output the new population The final population include:

[0020] Extract individuals from the initial population ;

[0021] Calculate individual The fitness function value of

[0022] According to the fitness function value, the population is iteratively evolved to generate a new generation of population ;

[0023] Set the number of iterations As the termination condition of the genetic algorithm, if the evolution is terminated, the new population is output The final population .

[0024] Among them, calculating the fitness function value of an individual includes:

[0025] Traversing individuals Detection range, including: distance detection range and Doppler detection range , the detection range is expressed as: ,in, , ;

[0026] Judging individuals Distance detection range Is it blocked? If not blocked, then In the clear zone;

[0027] Judging individuals Doppler detection range Is it blocked? If not blocked, then In the clear zone;

[0028] Get individual middle The number of combinations in the clear zone and the ratio of the number of combinations in the clear zone to the number of all combinations are defined as individual The fitness function value of .

[0029] Among them, the evolutionary processing of the population includes:

[0030] Select the individuals that will be inherited into the next generation of new population to form the new population of this generation;

[0031] The optimal preservation method is used to preserve the individuals with the highest fitness function value in the previous generation and the new population of this generation, and the individuals are used to replace the individuals with the smallest fitness function value in the new population of this generation;

[0032] Traverse all individuals in the new population of this generation, perform genetic crossover operations, and generate new individuals;

[0033] Perform genetic variation processing from new individuals to achieve evolution and generate new populations ;

[0034] Determine the evolution termination condition, if evolution is terminated, output the new population For the new generation of population ; Otherwise, calculate the new population The fitness function value of the individuals in the new population Perform evolutionary processing on the population.

[0035] Furthermore, the conditions for determining the termination of evolution include: setting the number of iterations As the termination condition of genetic algorithm.

[0036] According to the present invention, multiple groups of repetition frequency waveforms can be reasonably selected and transmitted from distributed multi-radar nodes, and their common occlusion area can be made as small as possible, thereby realizing the range-Doppler anti-occlusion function. The waveform designed by the present invention has more common clear areas than a single repetition frequency waveform, which can further improve the detection performance of the distributed radar system; in the implementation process of the present invention, the use of genetic algorithms has advantages such as small computational complexity, strong parallel processing, and strong stability compared to traditional traversal algorithms. The more repetition frequency scenes are designed, the more obvious their advantages are; that is, the waveform designed by the present invention: while ensuring good anti-occlusion effect, it reduces the computational complexity. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a step diagram of a waveform design method for distributed multi-node joint resistance to distance Doppler shielding provided by an embodiment of the present invention;

[0038] Figure 2 It is a flow chart of a waveform design method provided according to an embodiment of the present invention. DETAILED DESCRIPTION

[0039] The present invention combines an algorithm for the clear area ratio corresponding to a PRF with a genetic algorithm: the clear area ratio corresponding to a group of PRFs is used as the fitness function of the genetic algorithm to evaluate the fitness of the group of PRFs with the design requirements, and each selectable PRF is encoded as an individual gene to form an initial population. The population is subjected to genetic operations of selection, crossover, and compilation to generate a new population. After the fitness of the individuals in the new population is evaluated, a new round of genetic operations is performed. After multiple generations of cycles, the population is gradually evolved to a relatively optimal state, so as to design a PRF combination that meets the anti-occlusion requirements.

[0040] The specific implementation of the present invention is described in detail below with reference to the accompanying drawings.

[0041] The waveform design method for distributed multi-node joint anti-distance Doppler shielding proposed by the present invention is as follows: Figure 1 As shown, the following steps are included:

[0042] Step S100: determining a clarity ratio standard and defining a clear area, wherein the clear area includes a clutter clear area and a distance shielding area;

[0043] In practical applications, Doppler occlusion mainly comes from low-speed clutter. Targets in this area will be regarded as clutter and cannot be effectively detected. At the same time, due to the blurring of MPRF Doppler, the occlusion area will be extended.

[0044] Although ground clutter is mainly composed of main lobe clutter, altitude line clutter, side lobe clutter and clutter caused by radome refraction, the area that cannot be used for target detection is the main lobe clutter area, so this area is called the detection blind area.

[0045] On this basis, the present invention defines a frequency region: a clutter clear region, which is expressed as:

[0046] ,in, is the Doppler ambiguity multiplicity, is the clutter main lobe width;

[0047] Among them, the method for determining the clutter main lobe width is:

[0048] , where the platform velocity is , the radar wavelength is , the azimuth angle of the platform velocity direction relative to the center of the antenna beam is , the pitch angle is , the beam width is .

[0049] For example, at platform speed ,wavelength , the azimuth of the antenna relative to the beam center , pitch angle , beam width , the calculated clutter main lobe width .

[0050] On the other hand, mainly because the pulse signal has a certain pulse width, and the radar usually shares the same antenna for transmission and reception, it is impossible to receive when the pulse is transmitted, which creates a distance shielding area.

[0051] Only the echo delay is The target echo signal is complete, that is, not blocked, among which, is the pulse width, To protect time, is the pulse repetition interval, is the fuzzy multiplicity .

[0052] Since the target distance corresponding to a PRI is: , which can convert the delay into distance and define the distance clear area after blurring ( ): ;

[0053] The main factor affecting the fuzzy area is the duty cycle. ,The larger the duty cycle, the smaller the distance clear area is relatively;

[0054] The distance occlusion area can be expressed as: ,in, is the duty cycle, The speed of light.

[0055] In summary, the clutter clear area, distance clear area and distance blocked area can be used to determine whether a single PRF is clear or blocked: The standard for clarity is: judging the value after blurring Whether it is in the distance occlusion area, where: Representation calculation The remainder of

[0056] For this PRF, determine a Doppler The standard for clarity is: observe the value after blurring Is it in a clear clutter area?

[0057] When only two PRFs need to be designed, there are generally not too many combinations, and we can traverse all possible combinations to directly find the PRF combination that maximizes the clear area. However, usually, we need three or more PRFs to meet a certain clear area ratio. At this time, there are too many combinations and it is impossible to traverse all situations.

[0058] Therefore, in the subsequent steps of the present invention, a genetic algorithm is used to implement waveform design: a clear area ratio corresponding to a group of PRFs is used as the fitness function of the genetic algorithm.

[0059] Step S110: Determine the distance detection interval , Doppler detection range and selectable PRF range , discretize the PRF range uniformly;

[0060] Among them, the distance detection interval Refers to the minimum value of the target distance R corresponding to a PRI and maximum value Determined detection range; Doppler detection range Refers to the Doppler detection range Limited by minimum value and maximum value Determined detection range; PRF range Refers to the minimum value of PRF and maximum value Determine the frequency range.

[0061] The Doppler detection interval can be considered in half due to its positive and negative symmetry.

[0062] Step S120: Select from the selectable PRF range Extract individual genes from the population and create an initial population from the individual genes ;

[0063] In this step, each selectable PRF is used as a gene to establish a gene pool. Each individual randomly selects a PRF as the gene included in the individual from the gene pool using the binary coding method, that is, "0" corresponds to the individual not selecting this gene, and "1" corresponds to the individual selecting this gene. After all individuals are encoded, a diverse population is formed.

[0064] Specifically, a population includes M individuals , an individual includes N waveforms of different PRFs as the individual genes, where n is the waveform serial number, and: 0 < n < N; m is the individual serial number, and: 0 < m < N;

[0065] The population is represented as: ,

[0066] The individual is represented as: ,

[0067] The waveforms of different PRFs are represented as: .

[0068] Step S130: Optimize the initial population , generate the final population ;

[0069] In this step, the proportion of the common clear area corresponding to all PRFs in each individual is used as the fitness function of the genetic algorithm to optimize the initial population , including the following steps:

[0070] First, extract the individuals in the initial population , and then perform subsequent optimization processing:

[0071] Step S131: Calculate the fitness function values of the individuals in the population;

[0072] In this step, first obtain the population content, and then calculate the fitness function values of the individuals in the population; the supported populations can be the initial population generated in step S120 or the populations evolved in subsequent calculations.

[0073] The flowchart for calculating the fitness function values of the individuals in the population is as shown in Figure 2 part S210 of

[0074] Step S211: Extract the radar model parameters, including: extract the individual , traverse the detection range of the individual, including: the distance detection range and the Doppler detection range , the detection range is expressed as: ,in, , ;

[0075] Step S212: Calculate individual Distance Doppler clear zone of a single gene (PRF):

[0076] Specifically, judge the individual Doppler detection range of a single gene (PRF) Is it blocked? If not blocked, then In the clear zone;

[0077] Judging individuals Distance detection range of a single gene (PRF) Is it blocked? If not blocked, then In the clear zone;

[0078] In specific implementation, by judging Is it in the distance blocking area? , to judge Whether it is blocked, is the speed of light, is the duty cycle, To protect time, For the genes in this individual, is the pulse repetition interval corresponding to the individual gene, is the maximum detection range corresponding to the individual gene;

[0079] judge Is it in , to judge Whether it is blocked, is the clutter main lobe width, The genes in this individual.

[0080] Step S213: Obtain individual middle The number of combinations in the clear zone and the ratio of the number of combinations in the clear zone to the number of all combinations are defined as individual The fitness function value of .

[0081] When it is implemented, it will be in a clear area. The number of total The proportion of this individual The fitness function value of .

[0082] In the present invention, individuals are evaluated according to the fitness function value. A function value of 0 indicates that the entire detection area is in the distance blockage area or the Doppler blockage area; a function value of 1 indicates that the entire detection area is in the distance clear area and the clutter clear area, that is, the closer the function value is to 1, the better the individual is.

[0083] Step S132: performing evolution processing on the population according to the fitness function values ​​of the individuals in the population;

[0084] The process of evolving a population is as follows Figure 2 The S220 section is shown, including:

[0085] Step S221: Select individuals that will be inherited to the next generation of new populations to form the new population of this generation;

[0086] In this step, the roulette wheel selection method is first used to select individuals that will be inherited to the next generation of the new population, and the next generation of the new population is updated to the current generation of the new population.

[0087] Step S222: Use the optimal preservation method to preserve the individuals with the highest fitness function value (i.e., higher clear area ratio and individual fitness function value closer to 1) in the previous generation and the current generation of new population, and use them to replace the individuals with the smallest fitness function value (i.e., individual fitness function value closest to 0) in the current generation of new population.

[0088] Step S223: traverse all individuals in the new population of this generation, perform genetic crossover operations, and generate new individuals;

[0089] In this step, a single-point crossover method is used and the crossover probability is set to , It can be a fixed value or an adaptive value according to the situation of each generation (generally in the range of 0.4~0.9). All individuals in the population are paired, and a gene point is randomly set as the crossover point for the paired individuals. The gene segments on the same side of the crossover point of the paired individuals are swapped to produce two new individuals. This cycle is repeated until all individuals in the population have completed the crossover operation.

[0090] Step S224: Perform genetic variation processing on new individuals to achieve evolution and generate a new population ;

[0091] In this step, we first set the mutation probability , usually around 0.01~0.1, according to the mutation probability Determine whether the individual mutates. For individuals that need to mutate, randomly select a gene that is not contained in this individual in the population to replace any gene on the individual.

[0092] Step S225: When the selection, crossover, and mutation operations are completed, it means that the evolution has completed one generation and a new population after this evolution is generated. ;

[0093] In this step, the evolution termination condition is also determined. If the evolution is terminated, the new population Determined as the final population Otherwise, re-enter step S131 to calculate the population The fitness function value of the individual in the new population is obtained, and then step S132 is performed to Perform evolutionary processing.

[0094] The termination of evolution can be determined based on the number of iterations: Set the genetic generation The condition for determining whether the optimization iteration is terminated is that after the algorithm has gone through G cycles of iterations, it can stop and output the individual with the fitness value closest to 1 in the last generation. All PRFs contained in this individual are the final waveform group design results. If the number of iterations is less than the number of generations, the cycle continues.

[0095] Step S133: After the evolution is terminated, the final population is output .

[0096] Step S140: Through the final population Output waveform design results;

[0097] The final population generated in step S130 The individual with the highest fitness function All PRFs included are the largest PRF combination in the clear zone. This PRF combination is the waveform design result in which the clear zone meets the requirements within the radar detection distance-Doppler range.

[0098] The present invention also provides simulation cases to reflect the waveform design results under different circumstances. The specific simulation parameters are as follows:

[0099] Distance detection range: 1-60km, discrete by 20m;

[0100] Doppler detection range: 1.5-75kHz, discrete by 100Hz. Due to left-right symmetry, only the positive direction is considered (assuming that the platform speed has been compensated, so the center of the clutter area is 0Hz);

[0101] Main lobe clutter width: Since the main lobe clutter is related to various radar parameters, a conservative choice is 3kHz;

[0102] Selectable PRF range: 30-60kHz, discretized in 50Hz intervals;

[0103] Protection time: 190ns;

[0104] Duty cycle: 0.2.

[0105] According to the genetic algorithm provided in this embodiment, the anti-blocking 3PRF is designed to be: 44.9kHz, 47.1kHz, 51.65kHz, and the clear area ratio is 0.9550. It can be seen from the embodiment that the 3PRF waveform designed and confirmed by the waveform design method provided by the present invention has the largest distance-Doppler clear area in the common area (the smallest blockage area), which can make the effective detection range of the radar system larger and the detection effect better.

[0106] In the present invention, a set of waveforms are designed within the range of repetition frequency selectability so that their common distance-Doppler clear area is the largest. The use of a genetic algorithm can solve the problem that the large number of repetition frequency combinations leads to excessive calculation and inability to traverse all possible problems. At the same time, processing the structure of multiple individuals in the population is also conducive to parallelization, thereby significantly improving the efficiency of waveform design. In addition, the genetic algorithm starts to find the optimal solution from a set of multiple initial values, rather than iteratively solving from a single initial value. Therefore, compared with traditional optimization algorithms, the genetic algorithm is not easy to fall into local optimality, has better globality, and will not affect the multi-frequency anti-blocking effect.

[0107] The above disclosures are only several specific embodiments of the present invention; however, the present invention is not limited thereto, and any changes that can be conceived by those skilled in the art should fall within the protection scope of the present invention.

Claims

1. A distributed multi-node joint anti-range Doppler shielding waveform design method, characterized in that: The following steps are involved: Determine a clarity ratio standard and define a clear area, wherein the clear area includes a clutter clear area and a distance clear area; Determine the distance detection range , Doppler detection range and selectable PRF range , the PRF range is uniformly discretized; wherein the distance detection interval Refers to the minimum value of the target distance R corresponding to a PRI and maximum value Determined detection range; the Doppler detection interval Refers to the Doppler detection range Limited by minimum value and maximum value Determine the detection range; the PRF range Refers to the minimum value of PRF and maximum value Determine the frequency range; Extract individual genes from the selectable PRF range and create an initial population from the individual genes ; wherein, a population comprises M individuals , and an individual comprises N waveforms of different PRFs as individual genes, where n is the waveform serial number and: 0 < n < N; m is the individual serial number and: 0 < m < N; According to the clear area, iteratively optimize the initial population , output the new population The final population ; The iterative optimization of the initial population Includes: Extracting individuals from the initial population Calculate the individual The fitness function value of the population is obtained; according to the fitness function value, the population is iteratively evolved; and the new generation population with the highest fitness function value is output. ; wherein the calculation of the fitness function value of the individual includes: traversing the individual Detection range, including: distance detection range and Doppler detection range , the detection range is expressed as: ,in, , ; Determine the individual Distance detection range Is it blocked? If not blocked, then In the clear zone; judge the individual Doppler detection range Is it blocked? If not blocked, then In the clear zone; get individual middle The number of combinations in the clear zone, the ratio of the number of combinations in the clear zone to the number of all combinations is defined as individual The fitness function value of The final population The individual with the highest fitness function All PRFs included are the PRF combination with the largest clear area, and the PRF combination is the waveform design result; The clutter clear area is a frequency region, which is expressed as: ,in, is the Doppler ambiguity multiplicity, is the clutter main lobe width; The distance clearing area is expressed as: ,in, is the pulse width, To protect time, is the pulse repetition interval, is the target distance corresponding to one PRI.

2. The waveform design method according to claim 1, characterized in that: The method for determining the clutter main lobe width is: , where the platform velocity is , the radar wavelength is , the azimuth angle of the platform velocity direction relative to the center of the antenna beam is , the pitch angle is , the beam width is .

3. The waveform design method according to claim 1, characterized in that: The population is represented by: , The individuals are represented by: , The waveforms of the different PRFs are expressed as: .

4. The waveform design method according to claim 1, characterized in that: The evolutionary processing of the population comprises: Select the individuals that will be inherited into the next generation of new population to form the new population of this generation; The optimal preservation method is used to preserve the individuals with the highest fitness function values ​​in the previous generation and the new population of this generation, and the individuals with the lowest fitness function value in the new population of this generation are replaced by the individuals; Traversing all individuals in the new population of this generation, performing genetic crossover operations, and generating new individuals; Perform genetic variation processing on the new individuals to achieve evolution and generate a new population ; Determine the evolution termination condition, if evolution is terminated, output the new population The final population ; Otherwise, calculate the new population The fitness function value of the individuals in the new population Perform evolutionary processing.

5. The waveform design method according to claim 4, characterized in that: The determination of the evolution termination condition includes: setting the number of iterations As the termination condition of genetic algorithm.

6. The waveform design method according to claim 1, characterized in that: The distance clear area is converted into the distance blocked area, which is expressed as: ,in, is the duty cycle, The speed of light.

Images