A method and system for managing a multi-station radar with adaptive selection of radiation frequency

Abstract

The application discloses a kind of multi-station radar management methods and systems of radiation frequency adaptive selection, the method includes using angle domain division algorithm, the azimuth of target when different frequency radiation is divided into useless angle domain set and angle domain set to be used;Using heuristic algorithm, the correlation of angle domain set to be used is carried out with maximum constraint of fluctuation factor;Using overlap, vacancy angle decision algorithm, the decision and filling of overlapping angle and vacancy angle after heuristic correlation are carried out, and angle domain set after planning is obtained;By using interactive multiple model algorithm, the azimuth of target at each time is predicted, and the radiation frequency selection of each radar node is carried out according to the angle domain of azimuth, and the radar and frequency with maximum fluctuation factor are taken as the radiation source and radiation frequency at next time, to realize multi-station radar management.The application solves the problem that target loss is caused by weak target echo characteristics under certain angle, and improves detection performance.

Description

Technical Field

[0001] This invention belongs to the field of radar signal processing, and specifically relates to a multi-station radar management method and system with adaptive selection of radiation frequency. Background Technology

[0002] The Radar Cross Section (RCS) reflects a target's ability to reflect radar electromagnetic waves, and its size and shape have a significant impact on radar detection, target identification, and anti-jamming capabilities. In maneuvering target detection, the target's azimuth, elevation angle, and radar radiation frequency all affect the RCS, causing fluctuations. In research on maneuvering target detection performance, radar waveform design and target energy accumulation are two commonly used methods. Regarding radar waveform design, Feng Xiang et al. proposed a low-sidelobe composite waveform design method based on particle sampling projection to suppress the impact of range sidelobes on detection performance; Jiu Bo et al. proposed a fast constant modulus multi-input multi-output radar waveform design method to suppress noise interference while minimizing the radiated power in the interference and target areas. Target energy accumulation is mainly divided into single-frame coherent accumulation and multi-frame non-coherent accumulation. Regarding single-frame coherent accumulation, Wang Hui proposed a coherent accumulation algorithm for variable-acceleration moving targets based on Keystone and conjugate time-reversal transform, reducing the impact of range migration and Doppler frequency migration on detection performance. Regarding multi-frame non-coherent accumulation, Wu Zhihong et al. proposed a multi-frame accumulation detection DP-TBD algorithm based on adaptive state transition sets to solve the problem of low detection accuracy caused by motion model mismatch during maneuvering. However, none of the above algorithms consider the impact of target RCS fluctuation characteristics on detection performance. In practical applications, the fluctuation characteristics of maneuvering targets are difficult to predict, and the radiation frequency often mismatches with the target characteristics during maneuvering target detection, resulting in poor detection performance. Summary of the Invention

[0003] The technical problem to be solved by this invention is to propose a multi-station radar management method and system with adaptive selection of radiation frequency. Based on heuristic frequency planning, the radar node and radiation frequency are adaptively selected. By designing a frequency set that matches the target characteristics through heuristic correlation between different frequencies in the angular domain, the frequency selection that maximizes the detection probability at each moment is completed in this frequency set according to the azimuth angle prediction value. The selection of radar node and radiation frequency at each moment is completed with the maximum fluctuation factor as a constraint. This solves the problem of low detection probability caused by the mismatch between radiation frequency and target characteristics when detecting moving targets, and effectively improves the detection performance of moving targets.

[0004] To solve the above technical problems, the present invention adopts the following technical solution:

[0005] The present invention proposes a multi-station radar management method with adaptive radiation frequency selection, comprising the following steps:

[0006] S1. Using the angle domain partitioning algorithm, the azimuth angle of the target under different frequency radiation is divided into a set of unused angle domains and a set of unused angle domains.

[0007] S2. Using a heuristic algorithm, the set of corner domains to be used is associated with the maximum fluctuation factor as a constraint to obtain the associated corner domain set.

[0008] S3. Using the overlapping and missing angle decision algorithm, decide on and fill in the overlapping and missing angles in the associated angle domain set to obtain the planned angle domain set.

[0009] S4. By using an interactive multi-model algorithm, the azimuth angle θ(k) of the target at each moment is predicted, and the radiation frequency of each radar node is selected according to the angular domain where θ(k) is located. The radar and frequency with the largest fluctuation factor are taken as the radiation source and radiation frequency at the next moment, so as to realize multi-station radar management.

[0010] Furthermore, assume that the radar radiation frequency of the c-th sampling interval is f. c The j-th azimuth angle of the i-th angular domain is The i-th corner domain has a total of b i There are several azimuth angles. Since the target maneuvers within a certain range, there is an upper limit to the number of angles included in each azimuth range. and lower limit To maximize the fluctuation factor, the RCS fluctuation factor must be greater than a threshold value within a specific angular domain. In the known RCS database, f c The angle contained in the i-th angular region during radiation is The azimuth range of the region of interest corresponding to n frequencies is then divided into multiple angular domains, forming a set of planned angular domains.

[0011]

[0012] Furthermore, in step S1, considering the large range of target maneuverability, it is necessary to find the available angular domain within the azimuth angle when radiating at different frequencies, i.e., the RCS database. Under certain constraints, it is divided into a set of unused corner domains. Specifically, it includes the following sub-steps:

[0013] S101, Set the initial frequency set initial frequency f c =F1, starting angle The radiation frequency interval is f in The angular interval is a in The upper limit of the angle range is The lower limit of the angle range is

[0014] S102. Select the target RCS database Initial frequency f c The starting angle within the corresponding search angle range is Termination angle is One of the sections,

[0015] S103, if the fluctuation factor of the angular domain satisfies and but Execute S102; if Then It is divided into a subset of the unused angular domain.

[0016] S104, Order like Execute S102.

[0017] S105, Order f c =f c +f in .

[0018] S106, Repeat S101 to S105 until... Get the set of corner domains to be used

[0019] Furthermore, in step S2, heuristic algorithms are often used to solve complex, imprecise problems. They are empirical rules or methods used to guide algorithm selection. In addition, heuristic algorithms do not need to associate all subproblems; they only seek locally optimal solutions, thus requiring fewer resources and being easier to implement. This invention uses a heuristic algorithm for frequency planning. The heuristic algorithm consists of two parts: angle domain association and overlap, and missing angle decision. The angle domain association algorithm aims to... Heuristic association is performed on the qualified corner domains, including the following sub-steps:

[0020] S201, Set of Triangular Domains to be Used Compare all angle domains with an initial angle of 1, and take the angle domain with the largest fluctuation factor as the starting angle domain.

[0021] S202, in the set of unused corner domains Find all angle regions that have an angle count of 1 that coincides with the previous angle region; if no angle region that meets the requirements is found, increase the number of coincident angles by 1 and search again.

[0022] S203. If a matching corner domain is found, proceed to S205; otherwise, proceed to S204.

[0023] S204. Set the interval angle to 0 degrees, that is, the interval between the ending angle of the previous angle domain and the starting angle of the currently searched angle domain is 0 degrees, in the set of angle domains to be used. Search for a suitable angle region within the range; if no suitable angle region is found, add 'a' to the interval angle. in Search again; until the terminating angle of the previous angle region plus the interval angle equals a. max Execute S206.

[0024] S205. Find the corner domain with the largest fluctuation factor in the corner domain and set it as the connecting corner domain to the previous corner domain. Set the currently found corner domain as the previous corner domain and continue to execute S202.

[0025] S206. End the search and obtain the set of associated corner domains.

[0026] Furthermore, in step S3, after the angle domain association algorithm is completed, there are cases where some frequencies are associated with multiple angle domains simultaneously, and there are also cases where some angles are not associated with any frequency. Therefore, it is necessary to... The decision-making process for overlapping and missing angles ensures that each angle has exactly one associated angle domain. The decision-making process for overlapping and missing angles includes the following sub-steps:

[0027] S301, Regarding the set of associated corner domains Analyze all angular domains to find the set of overlapping angles. and the set of missing angles

[0028] S302, to The fluctuation factors of several overlapping angular domains are compared and judged, and the overlapping part is assigned to the angular domain with the highest fluctuation factor.

[0029] S303, The overlapping parts of other corner domains are deleted to form new corner domains.

[0030] S304. Repeat S302 and S303 until all overlapping parts have been decided and a set is obtained. in

[0031] S305, Judgment Check if the condition is met. If it is met, the algorithm ends; otherwise, execute S306.

[0032] S306. Determine if the angle range of the leftmost empty space is less than... If the value is less than 0, the missing angle is assigned to the neighboring angle domain with the largest fluctuation factor; otherwise, the missing angle is assigned to a new range, and the frequency with the largest fluctuation factor is selected from all frequencies.

[0033] S307. Repeat S305 to S306 until the loop ends, and obtain the set of planned corner domains.

[0034] Furthermore, in step S4, multi-station radar management includes the following:

[0035] Let the model set M, which comprises all the motion models of the target, be:

[0036] M = {m 1 ,m 2 ,...,m i ,...,m r}

[0037] Where, m i This refers to the motion model of the i-th target, where i = 1, 2, ..., r, and r is a positive integer.

[0038] The model at the (k-1)th sampling time is m i The model is from arrive The transition probability is π ij The observation state Z(k-1) in the radar c1 coordinate system at the (k-1)th sampling time is:

[0039]

[0040] in, This represents the x-coordinate of radar c1. This represents the y-coordinate of radar c1. This represents the z-coordinate of radar c1.

[0041] The mixed input state value at the (k-1)th sampling time is The target mixed error covariance matrix is ​​P 0j (k-1|k-1), when the model is j, the target state transition probability matrix is ​​φ. j Then the model is the predicted state of the target at sampling time k-1 at time j to that at sampling time k. for:

[0042]

[0043] Considering that the radiation frequency set designed in this invention is within a certain angular range, the target state at time k is predicted using the model probability at the (k-1)th sampling time. The model probability matrix of the target at the (k-1)th sampling time is:

[0044]

[0045] Among them, Λ j(k-1) represents the likelihood function at the (k-1)th sampling time when the model is j, π ij φ represents the transition probability from target model i to model j. i Let represent the target state transition probability matrix when model i is , then the target prediction value at the k-th sampling time is . Represented as:

[0046]

[0047] in, It includes the coordinates of the target's trajectory in the radar coordinate system.

[0048] The estimated values ​​are obtained by transforming the radar c1 coordinate system to other coordinate systems. And transform it to the target coordinate system to obtain Calculate the relative azimuth angle θ between the target and each radar. c (k); will θ c (k) Substitute into the planning angle domain set The optimal matching frequency for each radar station is obtained as follows: Will Substitute into RCS database The corresponding fluctuation factor is obtained from this. At this point, the frequency of the fluctuation factor at its maximum is:

[0049]

[0050] Where F(k) is the radiation frequency at time k+1.

[0051] The radar c corresponding to F(k) is the radiation source at time k+1.

[0052] Furthermore, this invention proposes a multi-station radar management system with adaptive radiation frequency selection, including...

[0053] The set partitioning module is used to divide the azimuth angle of a target radiating at different frequencies into a set of usable angular domains and a set of unused angular domains using an angular domain partitioning algorithm.

[0054] The set association module is used to use heuristic algorithms to associate the set of corner domains to be used with the maximum fluctuation factor as a constraint, and obtain the associated corner domain set.

[0055] The ensemble planning module uses overlapping and missing angle decision algorithms to determine and fill overlapping and missing angles in the associated angle domain set, thereby obtaining the planned angle domain set.

[0056] The radar management module is used to predict the azimuth angle θ(k) of the target at each moment by using an interactive multi-model algorithm, and select the radiation frequency of each radar node according to the angular domain where θ(k) is located. The radar and frequency with the largest fluctuation factor are selected as the radiation source and radiation frequency for the next moment, thereby realizing multi-station radar management.

[0057] Furthermore, the present invention proposes an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the multi-station radar management method for adaptive selection of radiation frequency described above.

[0058] Furthermore, the present invention proposes a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the aforementioned multi-station radar management method with adaptive radiation frequency selection.

[0059] The present invention adopts the above technical solution, and its significant technical effects compared with the prior art are as follows:

[0060] First, a heuristic frequency planning algorithm is used to design a frequency set that matches the target's fluctuation characteristics, solving the problem of poor detection performance caused by the mismatch between the radiation frequency and target characteristics under uncertain target echo conditions. Based on this, multi-station radar is introduced into the heuristic frequency planning algorithm. Leveraging the wider field of view of multi-station radar, the frequency optimization angular domain is expanded, solving the problem of target loss due to weak target echo characteristics at certain angles, further improving detection performance. Attached Figure Description

[0061] Figure 1 This is a flowchart illustrating the overall implementation of the present invention.

[0062] Figure 2 This is a flowchart of the heuristic frequency planning algorithm of the present invention.

[0063] Figure 3 This is a diagram showing the relative positions of the mobile target and the radar station in an embodiment of the present invention.

[0064] Figure 4 This is a comparison chart of the fluctuation factors of four maneuvering target tracking algorithms in the embodiments of the present invention.

[0065] Figure 5 This is a comparison chart of the detection probabilities of four maneuvering target tracking algorithms in this embodiment of the invention.

[0066] Figure 6 This is a radiation resource allocation diagram for four maneuvering target tracking algorithms in this embodiment of the invention. Detailed Implementation

[0067] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention.

[0068] Different wavelengths and scattering mechanisms can affect the performance of radar cross-section (RCS). For the same target, different radar radiation frequencies will yield different RCS. Therefore, by designing a radar radiation frequency that matches the target characteristics, the fluctuation factor of the RCS can be improved, thereby increasing the probability of target detection.

[0069] Assume the radar radiation frequency of the c-th sampling interval is f. c The j-th azimuth angle of the i-th angular domain is The i-th corner domain has a total of b i There are several azimuth angles. Since the target maneuvers within a certain range, there is an upper limit to the number of angles included in each azimuth range. and lower limit To maximize the fluctuation factor, the RCS fluctuation factor must be greater than a threshold value k within a specific angular domain. th In known RCS databases, f c The angle contained in the i-th angular region during radiation is The azimuth range of the region of interest corresponding to n frequencies is then divided into multiple angular domains, forming a set of planned angular domains.

[0070]

[0071] This invention proposes a multi-station radar management method that adaptively selects radiation frequencies by designing a frequency set that matches target characteristics through heuristic correlation between different frequencies in the angular domain. Based on the azimuth angle prediction value, the frequency selection at each time moment maximizes the detection probability within this frequency set. This method incorporates multi-station radar based on heuristic frequency planning. Figure 1 , Figure 2 As shown, it includes:

[0072] S1. Using an angle domain partitioning algorithm, the azimuth angle of the target radiating at different frequencies is divided into a set of usable angle domains and a set of useless angle domains. The specific details are as follows:

[0073] S101, Set the initial frequency set initial frequency f c =F1, starting angle The radiation frequency interval is f in The angular interval is a in The upper limit of the angle range is The lower limit of the angle range is

[0074] S102. Select the target RCS database Initial frequency f c The starting angle within the corresponding search angle range is Termination angle is One of the sections,

[0075] S103, if the fluctuation factor of the angular domain satisfies and but Execute S102; if Then It is divided into a subset of the unused angular domain.

[0076] S104, Order like Execute S102.

[0077] S105, Order f c =f c +f in .

[0078] S106, Repeat S101 to S105 until... Get the set of corner domains to be used

[0079] S2. Using a heuristic algorithm, the set of corner domains to be used is associated with the set of corner domains constrained by the maximum fluctuation factor, to obtain the associated set of corner domains. The specific content is as follows:

[0080] S201, Set of Triangular Domains to be Used Compare all angle domains with an initial angle of 1, and take the angle domain with the largest fluctuation factor as the starting angle domain.

[0081] S202, in the set of unused corner domains Find all angle regions that have an angle count of 1 that coincides with the previous angle region; if no angle region that meets the requirements is found, increase the number of coincident angles by 1 and search again.

[0082] S203. If a matching corner domain is found, proceed to S205; otherwise, proceed to S204.

[0083] S204. Set the interval angle to 0 degrees, that is, the interval between the ending angle of the previous angle domain and the starting angle of the currently searched angle domain is 0 degrees, in the set of angle domains to be used. Search for a suitable angle region within the range; if no suitable angle region is found, add 'a' to the interval angle. in Search again; until the terminating angle of the previous angle region plus the interval angle equals a. max Execute S206.

[0084] S205. Find the corner domain with the largest fluctuation factor in the corner domain and set it as the connecting corner domain to the previous corner domain. Set the currently found corner domain as the previous corner domain and continue to execute S202.

[0085] S206. End the search and obtain the set of associated corner domains.

[0086] S3. Using the overlapping and missing angle decision algorithm, decide on and fill in the overlapping and missing angles in the associated angle domain set to obtain the planned angle domain set. The specific content is as follows:

[0087] S301, Regarding the set of associated corner domains Analyze all angular domains to find the set of overlapping angles. and the set of missing angles

[0088] S302, to The fluctuation factors of several overlapping angular domains are compared and judged, and the overlapping part is assigned to the angular domain with the highest fluctuation factor.

[0089] S303, The overlapping parts of other corner domains are deleted to form new corner domains.

[0090] S304. Repeat S302 and S303 until all overlapping parts have been decided and a set is obtained. in

[0091] S305, Judgment Check if the condition is met. If it is met, the algorithm ends; otherwise, execute S306.

[0092] S306. Determine if the angle range of the leftmost empty space is less than... If the value is less than 0, the missing angle is assigned to the neighboring angle domain with the largest fluctuation factor; otherwise, the missing angle is assigned to a new range, and the frequency with the largest fluctuation factor is selected from all frequencies.

[0093] S307. Repeat S305 to S306 until the loop ends, and obtain the set of planned corner domains.

[0094] S4. By using an interactive multi-model algorithm, the azimuth angle θ(k) of the target at each time moment is predicted. Based on the angular domain where θ(k) is located, the radiation frequency of each radar node is selected. The radar and frequency with the largest fluctuation factor are taken as the radiation source and radiation frequency for the next time moment, thus realizing multi-station radar management. The specific content is as follows:

[0095] Let the model set M, which comprises all the motion models of the target, be:

[0096] M = {m 1 ,m 2 ,...,m i ,...,m r};

[0097] Where, m i This refers to the motion model of the i-th target, where i = 1, 2, ..., r, and r is a positive integer.

[0098] The model at the (k-1)th sampling time is m i The model is from arrive The transition probability is π ij The observation state Z(k-1) in the radar c1 coordinate system at the (k-1)th sampling time is:

[0099]

[0100] in, This represents the x-coordinate of radar c1. This represents the y-coordinate of radar c1. This represents the z-coordinate of radar c1.

[0101] The mixed input state value at the (k-1)th sampling time is The target mixed error covariance matrix is ​​P 0j (k-1|k-1), when the model is j, the target state transition probability matrix is ​​φ. j Then the model is the predicted state of the target at sampling time k-1 at time j to that at sampling time k. for:

[0102]

[0103] Considering that the radiation frequency set designed in this invention is within a certain angular range, the target state at time k is predicted using the model probability at the (k-1)th sampling time. The model probability matrix of the target at the (k-1)th sampling time is:

[0104]

[0105] Among them, Λ j (k-1) represents the likelihood function at the (k-1)th sampling time when the model is j, π ij φ represents the transition probability from target model i to model j. i Let represent the target state transition probability matrix when model i is , then the target prediction value at the k-th sampling time is . Represented as:

[0106]

[0107] in, It includes the coordinates of the target's trajectory in the radar coordinate system.

[0108] The estimated values ​​are obtained by transforming the radar c1 coordinate system to other coordinate systems. And transform it to the target coordinate system to obtain Calculate the relative azimuth angle θ between the target and each radar. c (k); will θ c (k) Substitute into the planning angle domain set The optimal matching frequency for each radar station is obtained as follows: Will Substitute into RCS database The corresponding fluctuation factor is obtained from this. At this point, the frequency of the fluctuation factor at its maximum is:

[0109]

[0110] Where F(k) is the radiation frequency at time k+1.

[0111] The radar c corresponding to F(k) is the radiation source at time k+1.

[0112] In this embodiment, the experimental parameters are set as follows:

[0113] The spatial coordinate system is a two-dimensional rectangular coordinate system; the initial position of the target is (-40km, -40km); the positions of radar stations 1 to 4 are (0km, 0km), (-20km, 0km), (-40km, 0km), and (-60km, 0km), respectively; the minimum radar measurement interval is 3s; and the number of sampling points is 70.

[0114] The initial motion model is uniform linear motion. At the 21st sampling time, it suddenly changes to cooperative turning. At the 41st sampling time, it continues to perform uniform linear motion. At the 56th sampling time, it suddenly changes to cooperative turning again, and maintains this motion model until the end of the simulation.

[0115] The IMM algorithm is used to track the trajectory of a maneuvering target, and the results are as follows: Figure 3 The diagram shows the relative positions of the moving target and the radar. The state transition matrices of the IMM algorithm are φ1, φ2, and φ3, respectively:

[0116]

[0117]

[0118]

[0119] The following algorithms were used to manage multi-station radars: a fixed-frequency method, a heuristic dynamic programming algorithm, a multi-station radar management algorithm with adaptive radiation frequency selection, and a multi-station radar management algorithm based on a fixed frequency. Figure 3 Tracking is performed on the trajectory. In single-station radar detection, radar node 1 is used; in multi-station radar detection, radar nodes 1 through 4 are used, resulting in the following... Figure 4 The diagram shows a comparison of the fluctuation factors of four maneuvering target tracking algorithms.

[0120] Substituting the fluctuation factor into the detection probability formula, we can obtain the following: Figure 5 The comparison chart of detection probabilities for the four maneuvering target tracking algorithms shown below, and as follows: Figure 6 The diagram shows the radiation resource allocation for the four maneuvering target tracking algorithms.

[0121] Depend on Figure 4 It can be seen that, using heuristic frequency planning, the fluctuation factor is greater than that of the fixed frequency in 43 out of 70 sampling times. The multi-station radar management algorithm using adaptive radiation frequency selection has a greater fluctuation factor than the fixed-frequency-based multi-station radar management algorithm in 63 out of 70 sampling times. Considering the size of the frequency set, the frequency set creation process requires that the angular range covered by the same frequency be as large as possible. Therefore, some frequency coverage circles with large fluctuation factors but small ranges are discarded, resulting in fluctuation factors in some angular domains being less than that of the fixed frequency. However, both the single-station heuristic frequency planning algorithm and the fixed-frequency-based multi-station radar management algorithm outperform the fixed-frequency algorithm overall.

[0122] Depend on Figure 6 It can be seen that the detection probability of the heuristic frequency planning method is higher than that of the fixed frequency method in 43 out of 70 sampling times, while the detection probability of the multi-station radar management algorithm with adaptive radiation frequency selection is higher than that of the fixed frequency-based multi-station radar management algorithm in 63 out of 70 sampling times, which is consistent with the fluctuation factor. In terms of overall performance, the average detection probability of the fixed frequency method is 0.6356, while the average detection probability of the multi-station radar management algorithm with adaptive radiation frequency selection is 0.8169, representing an overall improvement in detection performance of 28.5%. Figure 5 Four algorithms for selecting radar nodes and allocating radiation frequencies are presented. The darkest area indicates that the radar radiation frequency is 0, meaning that the radar node is not radiating at the current sampling time. It can be seen that the heuristic frequency planning algorithm can adaptively update the radiation frequency based on the predicted angle at different sampling times. The multi-station radar management algorithm with adaptive radiation frequency selection can also adaptively select radar nodes and radiation frequencies based on the current target's RCS echo signal, and the total number of radar radiations remains consistent with the fixed-frequency method.

[0123] For the multi-station radar management algorithm based on target frequency adaptive selection, this invention selects different upper limits of angle ranges for frequency set design, and statistically analyzes the fluctuation factor and detection probability of the target using the frequency set, as shown in Table 1 below.

[0124] Table 1. Statistics on Angle Range Thresholds

[0125]

[0126] As shown in Table 1, for the multi-station radar management algorithm based on target frequency adaptive selection, if the upper limit of the angle range is set too small, the angle domain range will be too small. When the IMM prediction error is large, the predicted angle will fall into the wrong angle range, resulting in incorrect frequency selection and a decrease in detection probability. If the upper limit is too large, it will result in too few matching frequencies. Therefore, setting the upper limit of the angle range to 40 degrees is optimal.

[0127] Furthermore, when the upper limit of the angle range is 1 degree and the upper limit is 40 degrees, this invention selects different fluctuation factor thresholds for frequency set design, and statistically analyzes the number of frequencies measured by the target using this frequency set and the average detection probability, as shown in Table 2 below.

[0128] Table 2 Statistics on Fluctuation Factor Thresholds

[0129]

[0130] As shown in Table 2, when the threshold is too small, too many angles are excluded in the corner domain partitioning algorithm step, resulting in too few frequencies available for selection in subsequent steps, thus reducing the average detection probability of the target. Conversely, if the threshold is too large, the corner domain will be too small, reducing the frequency prediction accuracy. Therefore, the optimal threshold value for the fluctuation factor is 0.8.

[0131] In summary, the multi-station radar management algorithm utilizing adaptive radiation frequency selection solves the problem of poor detection performance caused by the mismatch between radiation frequency and target characteristics in moving target detection. Verification analysis results show that, compared with traditional fixed-frequency algorithms, this scheme improves the detection performance of moving targets while addressing the radar node and radiation frequency selection issues.

[0132] This invention also proposes a multi-station radar management system for adaptive selection of radiation frequency, including a set partitioning module, a set association module, a set planning module, a radar management module, and a computer program that can run on a processor. It should be noted that each module in the above system corresponds to a specific step of the method provided in this invention embodiment, possessing the corresponding functional modules and beneficial effects for executing the method. Technical details not described in detail in this embodiment can be found in the method provided in this invention embodiment.

[0133] This invention also proposes an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. It should be noted that each module in the above system corresponds to a specific step of the method provided in this invention, possessing the corresponding functional modules and beneficial effects for executing the method. Technical details not described in detail in this embodiment can be found in the method provided in this invention.

[0134] This invention also proposes a computer-readable storage medium storing a computer program. It should be noted that each module in the above system corresponds to a specific step of the method provided in this invention, possessing the corresponding functional modules and beneficial effects for executing the method. Technical details not described in detail in this embodiment can be found in the method provided in this invention.

[0135] Although embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention. Any other corresponding changes and variations made in accordance with the technical concept of the present invention should be included within the protection scope of the claims of the present invention.

Claims

1. A multi-station radar management method with adaptive selection of radiation frequency, characterized in that, include: S1. Using the angle domain partitioning algorithm, the azimuth angle of the target under different frequency radiation is divided into a set of unused angle domains and a set of unused angle domains. S2. Using a heuristic algorithm, the set of corner domains to be used is associated with the constraint of maximizing the fluctuation factor, so as to obtain the associated corner domain set. S3. Use the overlapping and missing angle decision algorithm to make decisions and fill in the overlapping and missing angles in the associated angle domain set to obtain the planned angle domain set. S4. By using an interactive multi-model algorithm, the azimuth angle of the target at each moment is predicted, and the radiation frequency of each radar node is selected according to the angular domain where the azimuth angle is located. The radar and frequency with the largest fluctuation factor are taken as the radiation source and radiation frequency at the next moment, so as to realize multi-station radar management.

2. The multi-station radar management method with adaptive selection of radiation frequency according to claim 1, characterized in that, Step S1, dividing the area into a set of unused corner domains, includes the following sub-steps: S101, Set the initial frequency set as The initial frequency is f c =F1, starting angle is The radiation frequency interval is f in The angular interval is a in The upper limit of the angle range is The lower limit of the angle range is S102. Select the target RCS database Initial frequency f c The starting angle within the corresponding search angle range is Termination angle is One of the sections, S103, if the fluctuation factor of the angular domain satisfies and but Execute S102; if Then It is divided into a subset of the set of unused corner domains; S104, Order like Execute S102; S105, Order f c =f c +f in ; S106, Repeat S101 to S105 until... Get the set of corner domains to be used 3. The multi-station radar management method with adaptive selection of radiation frequency according to claim 2, characterized in that, In step S2, obtaining the set of associated corner domains includes the following sub-steps: S201. Compare all angle domains with an initial angle of 1 in the set of angle domains to be used, and take the angle domain with the largest fluctuation factor as the starting angle domain. S202. Find all angle regions in the set of unused angle regions that have an angle number of 1 that coincides with the previous angle region; if no angle region that meets the requirements is found, increase the number of coincident angles by 1 and search again. S203. If a suitable corner domain is found, proceed to S205; otherwise, proceed to S204. S204. Set the interval between the ending angle of the previous angle domain and the starting angle of the currently searched angle domain to 0 degrees, and search for a suitable angle domain in the set of available angle domains; if no suitable angle domain is found, add 'a' to the interval angle. in Search again; until the terminating angle of the previous angle region plus the interval angle equals a. max Execute S206; S205. Find the corner domain with the largest fluctuation factor in the corner domain and set it as the connecting corner domain to the previous corner domain. Set the currently found corner domain as the previous corner domain and continue to execute S202. S206. End the search and obtain the set of associated corner domains.

4. The multi-station radar management method with adaptive selection of radiation frequency according to claim 3, characterized in that, Step S3, obtaining the set of planned corner domains includes the following sub-steps: S301. Analyze all angle domains in the associated angle domain set to find the set of overlapping angles. and the set of missing angles S302, to The fluctuation factors of several overlapping angular domains are compared and judged, and the overlapping part is assigned to the angular domain with the highest fluctuation factor. S303, In other corner fields, the overlapping parts are deleted to create new corner fields; S304. Repeat S302 and S303 until all overlapping parts have been decided and a set is obtained. in S305, Judgment Check if the condition is met. If it is met, the algorithm ends. If it is not met, execute S306. S306. Determine if the angle range of the leftmost empty space is less than... If it is less than, the missing angle is assigned to the neighboring angle domain with the largest fluctuation factor; otherwise, the missing angle is assigned to a new range, and the frequency with the largest fluctuation factor is selected from all frequencies. S307. Repeat S305 to S306. When the algorithm ends, the set of planned angle domains is obtained.

5. The multi-station radar management method with adaptive selection of radiation frequency according to claim 4, characterized in that, In step S4, multi-station radar management includes the following: Let all the motion models of the target constitute the model set M, and the specific expression is: M={m 1 ,m 2 ,...,m i ,...,m r } Where, m i This represents the motion model of the i-th target, where i = 1, 2, ..., r; r is a positive integer. The model at the (k-1)th sampling time is m i The model is from arrive The transition probability is π ij The observation state Z(k-1) in the radar c1 coordinate system at the (k-1)th sampling time is: in, These represent the x, y, and z coordinates of radar c1, respectively. When the model is j, the predicted state of the target at sampling time k-1 is: in, φ represents the state prediction value at the (k-1)th sampling time for the kth sampling time. j This represents the target state transition probability matrix when the model is j. This represents the mixed input state value at the (k-1)th sampling time. The target state at time k is predicted using the model probability at time k-1. The model probability matrix of the target at time k-1 is: Among them, Λ j (k-1) represents the likelihood function at the (k-1)th sampling time when the model is j, π ij φ represents the transition probability from target model i to model j. i This represents the target state transition probability matrix when the model is i; The target prediction value at the k-th sampling time Represented as: The estimated values ​​are obtained by transforming the radar c1 coordinate system to other coordinate systems. And transform it to the target coordinate system to obtain Calculate the relative azimuth angle between the target and each radar; substitute the relative azimuth angle into the planned angle domain set. Obtain the optimal matching frequency for each radar station; substitute the optimal matching frequency into the RCS database. The corresponding fluctuation factor is obtained from this, and the frequency of the fluctuation factor at its maximum is: Where F(k) represents the radiation frequency at time k+1, Indicates the optimal matching frequency. θ represents the fluctuation factor corresponding to the optimal matching frequency. c (k) represents the relative azimuth angle; The radar c corresponding to F(k) is the radiation source at time k+1.

6. A multi-station radar management system with adaptive selection of radiation frequency, characterized in that, include The set partitioning module is used to divide the azimuth angle of a target radiating at different frequencies into a set of usable angular domains and a set of useless angular domains using an angular domain partitioning algorithm. The set association module is used to use heuristic algorithms to associate the set of corner domains to be used with the maximum fluctuation factor as a constraint, and obtain the associated corner domain set. The ensemble planning module is used to determine and fill overlapping and missing angles in the associated angle domain set using overlapping and missing angle decision algorithms, thereby obtaining the planned angle domain set. The radar management module is used to predict the azimuth angle θ(k) of the target at each moment by using an interactive multi-model algorithm, and select the radiation frequency of each radar node according to the angular domain where θ(k) is located. The radar and frequency with the largest fluctuation factor are selected as the radiation source and radiation frequency for the next moment, thereby realizing multi-station radar management.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that, The computer program is executed by the processor to perform the method of any one of claims 1 to 5.

Images