System and method for identifying unmanned aerial vehicle based on radar data

Abstract

The invention discloses a system and method for unmanned aerial vehicle identification based on radar data, and relates to the technical field of unmanned aerial vehicle identification, and the system comprises a radar detection module, a radar frequency modulation module, a detection data processing module, a target positioning module, and a detection data identification module. According to the system and method for identifying the unmanned aerial vehicle based on the radar data, whether the target is the unmanned aerial vehicle or not is identified through the returned electromagnetic wave signal, and the number of the unmanned aerial vehicles is identified; when the number of the unmanned aerial vehicle is one, the second reflection wave signal and the reflection wave signal received later in the reflection frequency are filtered within the duration corresponding to one set period, and when the number of the unmanned aerial vehicle is larger than one, the reflection wave signal is filtered; the reflection wave signals received by the radar detection module are directly reflected after meeting the unmanned aerial vehicle in the reflection wave signals of each reflection frequency, the number of the reflection wave signals needing to be processed is greatly reduced, and the positioning calculation efficiency is improved while the positioning precision of the unmanned aerial vehicle is guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of drone identification, and in particular to a system and method for drone identification based on radar data. Background Art

[0002] The process of drone identification is usually accompanied by the positioning of the drone. For drones working in complex environments, when using radar alignment positioning, there are usually obstacles near the drone, which will cause multipath effects, causing the radar positioning of the drone to drift and affect the positioning accuracy. That is, the radar signal may encounter obstacle reflections during the propagation process, resulting in the reception of multiple signals, thus causing positioning errors. To reduce the multipath effect error, it is generally necessary to improve the antenna design, use directional antennas and array antennas, which increases the hardware cost of the antenna, or directly rely on algorithm resolution, which increases the data processing volume of the algorithm. Summary of the invention

[0003] The purpose of the present invention is to provide a system and method for drone identification based on radar data to solve the above-mentioned deficiencies in the prior art.

[0004] In order to achieve the above-mentioned object, the present invention provides the following technical solutions: a system for identifying drones based on radar data, comprising a radar detection module, a radar frequency modulation module, a detection data processing module, a target positioning module, and a detection data identification module;

[0005] The radar detection module is used to transmit electromagnetic wave signals outward and receive reflected wave signals. The reflected wave signals are electromagnetic wave signals reflected back after encountering the target.

[0006] The radar detection module is also used to edit a time mark for the emitted electromagnetic wave signal, so that the emitted electromagnetic wave signal carries the time mark and records the emission time of the electromagnetic wave. The time mark is used to indicate the emission time of the electromagnetic wave signal. When the radar detection module receives the electromagnetic wave signal and encounters the reflected wave signal returned by the target, the reflected wave signal can be matched with the emitted electromagnetic wave signal through the time mark, and the emission time of the electromagnetic wave signal corresponding to the reflected wave signal can be obtained;

[0007] The radar frequency modulation module is used to periodically adjust the electromagnetic wave signal emitted by the radar detection module. The periodic adjustment is to cyclically adjust the frequency of the emitted electromagnetic wave signal in the set frequency range in the order of frequency size. The duration of each cycle is the duration of a set cycle, wherein the time taken for the emitted electromagnetic wave signal to be adjusted from the minimum value of the set frequency range to the maximum value each time is the duration of a cycle; in one embodiment, the frequency range is set to 300MHz-30GHz, and the radar frequency modulation module uniformly adjusts the frequency of the emitted electromagnetic wave signal from 300MHz to 30GHz within a set cycle, and then returns the emitted electromagnetic wave signal to 300MHz to start adjustment in the next set cycle;

[0008] The detection data processing module is used to determine the frequency of the reflected wave signal after the radar detection module receives the reflected wave signal, obtain the reflected frequency, and filter the second and subsequent reflected wave signals received in the reflected frequency within a time length corresponding to a set period;

[0009] The target positioning module is used to obtain the target distance based on the time difference between the electromagnetic wave signal emitted by the radar detection module and the corresponding reflected wave signal received. The target distance is the distance between the target and the radar detection module when the electromagnetic wave signal contacts the target;

[0010] The target positioning module is also used to obtain the target angle based on the scanning angle of the radar detection module when the electromagnetic wave signal corresponding to the target distance is emitted, that is, the angle of the target contacted by the electromagnetic wave signal corresponding to the target distance relative to the radar detection module, wherein the target angle may be an azimuth angle, and further, if the radar detection module can measure the vertical angle of the target relative to the horizontal plane, the target angle may also include a pitch angle;

[0011] The target positioning module is also used to obtain the target coordinates based on the target distance and the target angle, where the coordinates are expressed in angles and distances, and can also be based on the target angle. The target distance and trigonometric function formula will calculate the point coordinates of the target (x, y), where x is the horizontal coordinate of the target and y is the vertical coordinate of the target;

[0012] The detection data identification module is used to set multiple first judgment labels, associate the historical reflected wave signals with the corresponding first judgment labels, wherein the first judgment label is used to mark whether the reflected wave signal is a drone, and train the first machine learning model based on the historical reflected wave signals and the corresponding first judgment labels to obtain a drone identification model for outputting the predicted first judgment label based on the input reflected wave signal. The present invention can freely select a machine learning model for training according to specific needs, such as: support vector machine (SVM), convolutional neural network (CNN), recurrent neural network (RNN), long short-term memory network (LSTM), etc. Specifically, each The input of the training sample is set as the reflected wave signal, and the output of each training sample is set as the first judgment label; based on the input and output of each sample, a sample set is obtained; (4) based on the sample set, a first machine learning model is trained. Specifically, the sample set can be divided into a training set, a validation set and a test set in a ratio of 8:1:1; based on the training set, the first machine learning model is trained to obtain model parameters, and based on the validation set, the hyperparameters of the first machine learning model are adjusted to prevent the model from overfitting; based on the test set, the accuracy, recall rate, F1-score and other indicators of the first machine learning model are evaluated. If the accuracy, recall rate, F1-score and other indicators meet the setting requirements.

[0013] Furthermore, the detection data processing module is used to determine the frequency of the reflected wave signal after the radar detection module receives the reflected wave signal, obtain the reflected frequency, and filter the second and subsequent reflected wave signals received in the reflected frequency within a time length corresponding to a set period, including the following steps:

[0014] The detection data processing module obtains the reflected wave signal received by the radar detection module and the return time of the received reflected wave signal;

[0015] Match the reflected wave signal to the electromagnetic wave signal one by one. Specifically, each transmitted electromagnetic wave signal has a unique time mark or "pulse code". When the radar transmits a pulse (i.e., electromagnetic wave signal), it will record the transmission time of this pulse and determine the target distance by the delay of the received signal. By accurately controlling the transmission time of each pulse, the radar system can calculate the corresponding target distance based on the reception time of the echo signal and match it with a specific transmission pulse;

[0016] Determine the frequency of each reflected wave signal to obtain the reflection frequency;

[0017] The corresponding electromagnetic wave signals within the same set period are retained, and the reflected wave signals of each reflection frequency with the earliest return time are retained, and other reflected wave signals of each reflection frequency are filtered. In the process of radar detection, when there are obstacles near the target, it may cause multipath effects, causing the radar to drift in positioning the target and affect the positioning accuracy. That is, the radar signal may encounter obstacle reflections during propagation, resulting in the reception of multiple signals, thereby causing positioning errors. When the electromagnetic wave signal contacts the target, the direct return distance is the shortest and the time is also the shortest, so the earliest returned electromagnetic wave signal is the electromagnetic wave that has not been reflected by other obstacles.

[0018] Furthermore, the target positioning module is used to obtain the target distance based on the time difference between the electromagnetic wave signal emitted by the radar detection module and the corresponding reflected wave signal received, including the following steps:

[0019] Determine the return time of the target reflected wave signal to obtain the target return time;

[0020] Based on the emission time of the electromagnetic wave signal corresponding to the target reflected wave signal and the target return time, the target distance can be calculated using the following formula:

[0021] The speed of electromagnetic waves can be the speed of light.

[0022] Furthermore, the detection data processing module is also used to determine the number of detected drones based on the reflected wave signal received by the radar detection module, including the following steps:

[0023] Identify the received reflected wave signal through the drone identification model;

[0024] Filter out the reflected wave signals belonging to the drone, obtain the reflected wave signals of the drone, and obtain the signal strength of the reflected wave signals of each drone;

[0025] Based on the return time of the reflected wave signal of each drone and the emission time of the corresponding electromagnetic wave signal, the target distance of the drone corresponding to the reflected wave signal of each drone is calculated to obtain the drone distance;

[0026] Set multiple second judgment tags, and associate the drone distance and signal strength corresponding to each drone reflection wave signal in history with the corresponding second judgment tag, wherein the second judgment tag is used to mark the drone distance and signal strength corresponding to the drone reflection wave signal, whether it meets the signal strength directly reflected back after the electromagnetic wave signal contacts the drone;

[0027] Based on the drone distance and signal strength of the drone reflected wave signal and the second judgment label, a second machine learning model is trained to obtain a direct reflection signal recognition model for outputting a predicted second judgment label based on the input drone distance and signal strength;

[0028] Based on the direct reflection signal recognition model, the drone reflection wave signal that is directly reflected by the drone and received by the radar detection module is identified in the drone reflection wave signal to obtain the real position reflection wave signal;

[0029] The number of reflected wave signals at the real position of each reflection frequency of the corresponding electromagnetic wave signal within the same setting cycle is counted to obtain the number of drones.

[0030] Further, training the second learning model includes the following steps:

[0031] A regression analysis was performed on the drone distance and the corresponding signal strength, and the signal strength prediction formula based on the drone distance prediction signal strength was obtained:

[0032] y=ax k +b,

[0033] Among them, y is the predicted signal strength, x is the distance of the drone, and a, b, and k are fixed parameters;

[0034] Get the actual signal strength Set the judgment formula:

[0035]

[0036] The results of P≥0 and P<0 are respectively associated with the corresponding second judgment labels, for example, P≥0 is associated with "yes" and P<0 is associated with no, so that "yes" is output when P≥0 and "no" is output when P<0.

[0037] Furthermore, the detection data processing module is also used to retain the real position reflection wave signal in the reflection wave signal of each reflection frequency when the number of drones is greater than one, that is, the real position reflection wave signal in the reflection wave signal of each reflection frequency is not filtered.

[0038] A method for drone identification based on radar data comprises the following steps:

[0039] S1. Periodically adjusting the frequency of electromagnetic wave signals emitted by the radar detection module;

[0040] S2, obtaining historical information of the radar detection module transmitting electromagnetic wave signals, receiving reflected wave signals, and detecting whether the target is a drone;

[0041] S3, training the first machine learning model through the detection data recognition module to obtain a drone recognition model;

[0042] S4. Training a second machine learning model through the detection data processing module to obtain a direct reflection signal recognition model;

[0043] S5, judging whether the target corresponding to the reflected wave signal is a drone based on the drone recognition model;

[0044] S6. If yes, then the number of drones is obtained based on the direct reflection signal recognition model;

[0045] S7. If the number of drones is one, filter the second and subsequent reflected wave signals received in the reflected frequency within a time period corresponding to a set period;

[0046] S8. If the number of drones is greater than one, when filtering the second and subsequent reflected wave signals received in the reflected frequency within a time length corresponding to a set period, the real position reflected wave signal in the reflected wave signal of each reflected frequency is retained.

[0047] 1. Compared with the prior art, the system and method for identifying drones based on radar data provided by the present invention can identify whether a target is a drone and the number of drones through the returned electromagnetic wave signal by setting a radar detection module, a radar frequency modulation module, a detection data processing module, a target positioning module, and a detection data identification module;

[0048] 2. Compared with the prior art, the system and method for drone identification based on radar data provided by the present invention also filters the second and subsequent reflected wave signals received in the reflection frequency within a time length corresponding to a set period when the number of drones is one, and retains the reflected wave signals of each reflection frequency that are directly reflected and received by the radar detection module after encountering the drone when the number of drones is greater than one, thereby greatly reducing the number of reflected wave signals that need to be processed, ensuring the drone positioning accuracy while improving the efficiency of positioning calculation. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0050] Figure 1 A system structure block diagram provided for an embodiment of the present invention;

[0051] Figure 2A diagram of the method steps provided for an embodiment of the present invention. DETAILED DESCRIPTION

[0052] In order to enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings.

[0053] In the description of the present invention, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined. In addition, the terms "installed", "connected" and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0054] Example embodiments will be described more fully below with reference to the accompanying drawings, but the example embodiments may be embodied in different forms and should not be construed as limited to the embodiments set forth herein. On the contrary, the purpose of providing these embodiments is to make the present disclosure thorough and complete and to enable those skilled in the art to fully understand the scope of the present disclosure.

[0055] In the absence of conflict, the various embodiments of the present disclosure and the various features therein may be combined with each other.

[0056] As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0057] The terms used herein are only used to describe specific embodiments and are not intended to limit the present disclosure. As used herein, the singular forms "a", "an" and "the" are also intended to include the plural forms, unless the context clearly indicates otherwise. It will also be understood that when the terms "comprising" and / or "made of" are used in this specification, the presence of the features, wholes, steps, operations, elements and / or components is specified, but the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or groups thereof is not excluded.

[0058] See also Figure 1-Figure 2 , a system for identifying UAVs based on radar data, including a radar detection module, a radar frequency modulation module, a detection data processing module, a target positioning module, and a detection data identification module;

[0059] The radar detection module is used to transmit electromagnetic wave signals outward and receive reflected wave signals. The reflected wave signals are electromagnetic wave signals reflected back after encountering the target.

[0060] The radar detection module is also used to edit a time mark for the emitted electromagnetic wave signal, so that the emitted electromagnetic wave signal carries the time mark and records the emission time of the electromagnetic wave. The time mark is used to indicate the emission time of the electromagnetic wave signal. When the radar detection module receives the electromagnetic wave signal and encounters the reflected wave signal returned by the target, the reflected wave signal can be matched with the emitted electromagnetic wave signal through the time mark, and the emission time of the electromagnetic wave signal corresponding to the reflected wave signal can be obtained;

[0061] The radar frequency modulation module is used to periodically adjust the electromagnetic wave signal emitted by the radar detection module. The periodic adjustment is to cyclically adjust the frequency of the emitted electromagnetic wave signal in the set frequency range in the order of frequency size. The duration of each cycle is the duration of a set cycle, wherein the time taken for the emitted electromagnetic wave signal to be adjusted from the minimum value of the set frequency range to the maximum value each time is the duration of a cycle; in one embodiment, the frequency range is set to 300MHz-30GHz, and the radar frequency modulation module uniformly adjusts the frequency of the emitted electromagnetic wave signal from 300MHz to 30GHz within a set cycle, and then returns the emitted electromagnetic wave signal to 300MHz to start adjustment in the next set cycle;

[0062] The detection data processing module is used to determine the frequency of the reflected wave signal after the radar detection module receives the reflected wave signal, obtain the reflected frequency, and filter the second and subsequent reflected wave signals received in the reflected frequency within a time length corresponding to a set period, including the following steps:

[0063] (1) The detection data processing module obtains the reflected wave signal received by the radar detection module and the return time of the received reflected wave signal;

[0064] (2) Matching reflected wave signals with electromagnetic wave signals. Specifically, each transmitted electromagnetic wave signal has a unique time stamp or "pulse code". When the radar transmits a pulse (i.e., electromagnetic wave signal), it records the transmission time of this pulse and determines the target distance by the time delay of the received signal. By precisely controlling the transmission time of each pulse, the radar system can calculate the corresponding target distance based on the reception time of the echo signal and match it with a specific transmitted pulse.

[0065] (3) Determine the frequency of each reflected wave signal to obtain the reflection frequency;

[0066] (4) The corresponding electromagnetic wave signals are retained within the same set period, and the reflected wave signals of each reflection frequency with the earliest return time are retained, and other reflected wave signals of each reflection frequency are filtered. In the process of radar detection, when there are obstacles near the target, it may cause multipath effect, resulting in the drift of radar positioning of the target and affecting the positioning accuracy. That is, the radar signal may encounter obstacle reflection during the propagation process, resulting in the reception of multiple signals, thereby causing positioning errors. When the electromagnetic wave signal contacts the target, the direct return distance is the shortest and the time is also the shortest, so the earliest returned electromagnetic wave signal is the electromagnetic wave that has not been reflected by other obstacles.

[0067] The detection data processing module is also used to retain the real position reflected wave signal in the reflected wave signal of each reflection frequency when the number of drones is greater than one, that is, the real position reflected wave signal in the reflected wave signal of each reflection frequency is not filtered.

[0068] The target positioning module is used to obtain the target distance based on the time difference between the electromagnetic wave signal emitted by the radar detection module and the corresponding reflected wave signal received. The target distance is the distance between the target and the radar detection module when the electromagnetic wave signal contacts the target, and includes the following steps:

[0069] (1) Determine the return time of the target reflected wave signal to obtain the target return time;

[0070] (2) Based on the emission time of the electromagnetic wave signal corresponding to the target reflected wave signal and the target return time, the target distance is calculated using the following formula:

[0071]

[0072] The speed of electromagnetic waves can be the speed of light.

[0073] The target positioning module is also used to obtain the target angle based on the scanning angle of the radar detection module when the electromagnetic wave signal corresponding to the target distance is emitted, that is, the angle of the target contacted by the electromagnetic wave signal corresponding to the target distance relative to the radar detection module, wherein the target angle may be an azimuth angle, and further, if the radar detection module can measure the vertical angle of the target relative to the horizontal plane, the target angle may also include a pitch angle;

[0074] The target positioning module is also used to obtain the target coordinates based on the target distance and the target angle, where the coordinates are expressed in angles and distances, and can also be based on the target angle. The target distance and trigonometric function formula will calculate the point coordinates of the target (x, y), where x is the horizontal coordinate of the target and y is the vertical coordinate of the target;

[0075] The detection data identification module is used to set multiple first judgment labels, associate the historical reflected wave signals with the corresponding first judgment labels, wherein the first judgment label is used to mark whether the reflected wave signal is a drone, and train the first machine learning model based on the historical reflected wave signals and the corresponding first judgment labels to obtain a drone identification model for outputting the predicted first judgment label based on the input reflected wave signal. The present invention can freely select a machine learning model for training according to specific needs, such as: support vector machine (SVM), convolutional neural network (CNN), recurrent neural network (RNN), long short-term memory network (LSTM), etc. Specifically, each The input of the training sample is set as the reflected wave signal, and the output of each training sample is set as the first judgment label; based on the input and output of each sample, a sample set is obtained; (4) based on the sample set, a first machine learning model is trained. Specifically, the sample set can be divided into a training set, a validation set and a test set in a ratio of 8:1:1; based on the training set, the first machine learning model is trained to obtain model parameters, and based on the validation set, the hyperparameters of the first machine learning model are adjusted to prevent the model from overfitting; based on the test set, the accuracy, recall rate, F1-score and other indicators of the first machine learning model are evaluated. If the accuracy, recall rate, F1-score and other indicators meet the setting requirements.

[0076] The detection data processing module is also used to determine the number of detected drones based on the reflected wave signal received by the radar detection module, including the following steps:

[0077] (1) Identify the received reflected wave signal through the UAV identification model;

[0078] (2) filtering out the reflected wave signals belonging to the UAVs, obtaining the reflected wave signals of the UAVs, and obtaining the signal strength of the reflected wave signals of each UAV;

[0079] (3) Based on the return time of the reflected wave signal of each drone and the emission time of the corresponding electromagnetic wave signal, the target distance of the drone corresponding to the reflected wave signal of each drone is calculated to obtain the drone distance;

[0080] (4) setting a plurality of second judgment tags, and associating the drone distance and signal strength corresponding to each drone reflected wave signal in history with the corresponding second judgment tag, wherein the second judgment tag is used to mark the drone distance and signal strength corresponding to the drone reflected wave signal, and whether it satisfies the signal strength directly reflected back after the electromagnetic wave signal contacts the drone;

[0081] (5) Based on the drone distance and signal strength of the drone reflected wave signal and the second judgment label, a second machine learning model is trained to obtain a direct reflection signal recognition model for outputting a predicted second judgment label based on the input drone distance and signal strength;

[0082] In one embodiment, training the second learning model includes the following steps:

[0083] a. Perform regression analysis on the drone distance and the corresponding signal strength to obtain the signal strength prediction formula based on the drone distance prediction signal strength:

[0084] y=ax k +b,

[0085] Among them, y is the predicted signal strength, x is the distance of the drone, and a, b, and k are fixed parameters;

[0086] b. Get the actual signal strength Set the judgment formula:

[0087]

[0088] The results of P≥0 and P<0 are respectively associated with the corresponding second judgment labels, for example, P≥0 is associated with "yes" and P<0 is associated with no, so that "yes" is output when P≥0 and "no" is output when P<0.

[0089] (6) Based on the direct reflection signal recognition model, the drone reflection wave signal that is directly reflected by the drone and received by the radar detection module is identified in the drone reflection wave signal to obtain the real position reflection wave signal;

[0090] (7) Count the number of reflected wave signals at the real position of each reflection frequency of the corresponding electromagnetic wave signal within the same setting cycle to obtain the number of drones.

[0091] A method for drone identification based on radar data comprises the following steps:

[0092] S1. Periodically adjusting the frequency of electromagnetic wave signals emitted by the radar detection module;

[0093] S2, obtaining historical information of the radar detection module transmitting electromagnetic wave signals, receiving reflected wave signals, and detecting whether the target is a drone;

[0094] S3, training the first machine learning model through the detection data recognition module to obtain a drone recognition model;

[0095] S4. Training a second machine learning model through the detection data processing module to obtain a direct reflection signal recognition model;

[0096] S5, judging whether the target corresponding to the reflected wave signal is a drone based on the drone recognition model;

[0097] S6. If yes, then the number of drones is obtained based on the direct reflection signal recognition model;

[0098] S7. If the number of drones is one, filter the second and subsequent reflected wave signals received in the reflected frequency within a time period corresponding to a set period;

[0099] S8. If the number of drones is greater than one, when filtering the second and subsequent reflected wave signals received in the reflected frequency within a time length corresponding to a set period, the real position reflected wave signal in the reflected wave signal of each reflected frequency is retained.

[0100] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A system for identifying drones based on radar data, characterized in that: It includes radar detection module, radar frequency modulation module, detection data processing module, target positioning module and detection data identification module; The radar detection module is used to transmit electromagnetic wave signals outwardly and receive reflected wave signals, wherein the reflected wave signals are electromagnetic wave signals reflected back after the electromagnetic wave signals encounter the target; The radar detection module is also used to edit a time mark for the emitted electromagnetic wave signal so that the emitted electromagnetic wave signal carries the time mark and records the emission time of the electromagnetic wave, wherein the time mark is used to indicate the emission time of the electromagnetic wave signal; The radar frequency modulation module is used to periodically adjust the electromagnetic wave signal emitted by the radar detection module. The periodic adjustment is to cyclically adjust the frequency of the emitted electromagnetic wave signal in the set frequency range in the order of frequency magnitude, and the duration of each cycle is the duration of a set period; The detection data processing module is used to determine the frequency of the reflected wave signal after the radar detection module receives the reflected wave signal, obtain the reflected frequency, and filter the second and subsequent reflected wave signals received in the reflected frequency within a time length corresponding to a set period; The target positioning module is used to obtain the target distance based on the time difference between the electromagnetic wave signal emitted by the radar detection module and the corresponding reflected wave signal received, and the target distance is the distance between the target and the radar detection module when the electromagnetic wave signal contacts the target; The target positioning module is also used to obtain the target angle based on the scanning angle of the radar detection module when the electromagnetic wave signal corresponding to the target distance is emitted; The target positioning module is also used to obtain target coordinates based on the target distance and target angle; The detection data identification module is used to set multiple first judgment labels, associate historical reflection wave signals with corresponding first judgment labels, and train a first machine learning model based on historical reflection wave signals and corresponding first judgment labels to obtain a drone identification model that outputs a predicted first judgment label based on the input reflection wave signal.

2. The system for identifying drones based on radar data according to claim 1, characterized in that: The detection data processing module is used to determine the frequency of the reflected wave signal after the radar detection module receives the reflected wave signal, obtain the reflected frequency, and filter the second and subsequent reflected wave signals received in the reflected frequency within a time length corresponding to a set period, including the following steps: The detection data processing module obtains the reflected wave signal received by the radar detection module and the return time of receiving the reflected wave signal; Matching the reflected wave signal with the electromagnetic wave signal one by one; Determine the frequency of each reflected wave signal to obtain the reflection frequency; The corresponding electromagnetic wave signals are retained within the same setting cycle, and the reflected wave signals of each reflected frequency with the earliest return time are retained, and other reflected wave signals of each reflected frequency are filtered.

3. The system for identifying drones based on radar data according to claim 2, characterized in that: The target positioning module is used to obtain the target distance based on the time difference between the electromagnetic wave signal emitted by the radar detection module and the corresponding reflected wave signal received, and includes the following steps: Determine the return time of the target reflected wave signal to obtain the target return time; The target distance is calculated based on the emission time of the electromagnetic wave signal corresponding to the target reflected wave signal and the target return time.

4. The system for identifying drones based on radar data according to claim 3, characterized in that: The detection data processing module is also used to determine the number of detected drones based on the reflected wave signal received by the radar detection module, including the following steps: Identify the received reflected wave signal through the drone identification model; Filter out the reflected wave signals belonging to the drone, obtain the reflected wave signals of the drone, and obtain the signal strength of the reflected wave signals of each drone; Based on the return time of the reflected wave signal of each drone and the emission time of the corresponding electromagnetic wave signal, the target distance of the drone corresponding to the reflected wave signal of each drone is calculated to obtain the drone distance; Set a plurality of second judgment tags, and associate the drone distance and signal strength corresponding to each drone reflection wave signal in history with the corresponding second judgment tag; Based on the drone distance and signal strength of the drone reflected wave signal and the second judgment label, a second machine learning model is trained to obtain a direct reflection signal recognition model for outputting a predicted second judgment label based on the input drone distance and signal strength; Based on the direct reflection signal recognition model, the drone reflection wave signal that is directly reflected by the drone and received by the radar detection module is identified in the drone reflection wave signal to obtain the real position reflection wave signal; The number of reflected wave signals at the real position of each reflection frequency of the corresponding electromagnetic wave signal within the same setting cycle is counted to obtain the number of drones.

5. The system for identifying drones based on radar data according to claim 4, characterized in that: The training of the second learning model comprises the following steps: A regression analysis is performed on the drone distance and the corresponding signal strength to obtain a signal strength prediction formula based on the drone distance prediction signal strength: y=ax k +b, Among them, y is the predicted signal strength, x is the distance of the drone, and a, b, and k are fixed parameters; Get the actual signal strength Set the judgment formula: The results of P≥0 and P<0 are associated with the corresponding second judgment labels respectively.

6. The system for identifying drones based on radar data according to claim 5, characterized in that: The detection data processing module is also used to retain the real position reflection wave signal in the reflection wave signal of each reflection frequency when the number of drones is greater than one.

7. A method for identifying drones based on radar data, applicable to a system for identifying drones based on radar data as claimed in any one of claims 1 to 6, characterized in that: The following steps are involved: S1. performing the periodic adjustment on the frequency of the electromagnetic wave signal emitted by the radar detection module; S2, obtaining historical information of the radar detection module transmitting electromagnetic wave signals, receiving reflected wave signals, and detecting whether the target is a drone; S3, training the first machine learning model through the detection data recognition module to obtain a drone recognition model; S4. Training a second machine learning model through the detection data processing module to obtain a direct reflection signal recognition model; S5. Determine whether the target corresponding to the reflected wave signal is a drone based on the drone identification model; S6. If yes, then the number of drones is obtained based on the direct reflection signal recognition model; S7. If the number of drones is one, filtering the second and subsequent reflected wave signals received in the reflected frequency within a time period corresponding to a set period; S8. If the number of drones is greater than one, when filtering the second and subsequent reflected wave signals received in the reflected frequency within the time length corresponding to a set period, the real position reflected wave signal in the reflected wave signal of each reflected frequency is retained.

Images