Unmanned aerial vehicle target micro-motion feature extraction method and system based on adaptive variable waveform

Abstract

The application provides a UAV target micro-motion feature extraction method and system based on adaptive variable waveforms, and relates to the technical fields of radar anti-UAV detection and target micro-motion feature extraction. The application firstly constructs an identification waveform library including identification waveforms of all distance segments in the distance range of the radar target micro-motion feature extraction; then based on the identification request information of the target track of the micro-motion feature target to be extracted, the identification waveform library is adaptively called to adapt to the identification request information of the target track; finally, the micro-motion feature extraction of the target track of the micro-motion feature to be extracted is completed based on the adapted identification waveform. The application extracts the target micro-motion feature based on the adaptive variable waveform, thereby providing a solution for the practical engineering application of the UAV recognition by using the micro-motion feature.

Description

Technical Field

[0001] This invention relates to the field of radar counter-UAV detection and target micro-motion feature extraction technology, specifically to a method and system for UAV target micro-motion feature extraction based on adaptive variable waveform. Background Technology

[0002] Micromotion features refer to the additional Doppler frequency modulation phenomenon caused by micromotions such as oscillation, rotation, and vibration in the target's echo signal, in addition to the target's main motion. Aircraft engine blades, UAV rotors, and bird wings all produce unique micromotion features. Micromotion features caused by rotor rotation can serve as an important characteristic for identifying slow-moving, small UAVs and have received considerable attention from researchers. The foundation for using micromotion features for UAV identification lies in how to effectively extract these micromotion features from the target.

[0003] Currently, there is very little research on target micro-motion feature extraction, and publicly reported experimental results demonstrating the ability to extract micro-motion features from micro-sized UAVs over a distance exceeding 1 km are extremely rare. The main reason is that the RCS (radar cross section) of micro-sized UAVs is inherently very small, and the rotor's RCS is even lower by tens of dB (and may be 30-40 dB lower depending on the material, UAV motion, and flight attitude of different UAVs). Therefore, the extremely weak rotor echo makes it very difficult to detect. Furthermore, conventional anti-UAV radars often lack the energy required for rotor micro-motion feature extraction during routine detection, making it difficult to extract the micro-motion features of UAV targets at long distances. Summary of the Invention

[0004] (a) Technical problems to be solved

[0005] To address the shortcomings of existing technologies, this invention provides a method and system for extracting micro-motion features of UAV targets based on adaptive variable waveforms, which solves the problem that existing technologies struggle to extract micro-motion features of UAV targets over long distances using a single detection waveform.

[0006] (II) Technical Solution

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] Firstly, this invention proposes a method for extracting micro-motion features of UAV targets based on adaptive variable waveforms, the method comprising:

[0009] Construct a recognition waveform library required for extracting micro-motion features; the recognition waveform library includes recognition waveforms for all range segments within the range in which radar extracts target micro-motion features;

[0010] The system receives a target trajectory recognition request from a target whose micro-motion features are to be extracted, and adaptively retrieves a recognition waveform that matches the target trajectory recognition request from the recognition waveform library based on the target trajectory recognition request; the target trajectory recognition request includes distance information of the target trajectory.

[0011] Based on the matching recognition waveform, the micro-motion features of the target trajectory to be extracted are completed.

[0012] Preferably, the construction of the identification waveform library includes:

[0013] S11. Divide the range R of the target micro-motion features extracted by the radar into N range segments, where the coverage area of ​​each range segment is R. i_min ~R i_max (i = 1, 2, ..., N); where R i_min R represents the minimum distance value of the i-th segment; i_max This represents the maximum distance value of the i-th segment;

[0014] S12. Calculate the waveform parameters of the identification waveform required for extracting micro-motion features in each range segment according to the radar equation and the requirements for extracting micro-motion features. Construct an identification waveform library based on the identification waveforms corresponding to all calculated waveform parameters. The waveform parameters include radar pulse repetition frequency, pulse width, number of pulses, and bandwidth.

[0015] Preferably, S12 includes:

[0016] S121. Initialize the farthest distance segment R N_min ~R N_max The waveform parameters of the identified waveform; among them, the radar pulse repetition frequency parameter of the identified waveform at the farthest range needs to meet the following conditions:

[0017]

[0018] Where c is the speed of light, f D Ω is the maximum speed at the tip of the UAV rotor, L is the length of the UAV rotor blade, Ω is the rotational speed of the UAV rotor, θ is the angle between the radar line of sight and the plane of rotation, and λ is the radar wavelength.

[0019] The pulse width τ of the waveform identified at the furthest distance N Pulse count M N and bandwidth B N The calculation formula is:

[0020]

[0021] The maximum distance R of the farthest segment N_max Conditions to be met:

[0022]

[0023] Among them, M max The maximum number of pulses accumulated by the radar; B max σ is the maximum signal bandwidth of the radar; P is the target cross-sectional area; t τ is the transmit power; G is the transmit pulse width; t G r These represent the antenna transmit gain and receive gain, respectively; λ is the wavelength; k is the Boltzmann constant; T0 is the temperature; F n denoted as receiver noise figure (SNR). min The signal-to-noise ratio for detecting micro-motion characteristics; L represents the radar system loss;

[0024] Minimum distance R of the farthest segment N_min Conditions to be met:

[0025]

[0026] S122, Order R i-1_max =R i_min Adjust the duty cycle D of the i-th distance segment. i Pulse count M i Signal bandwidth B i Following the iterative calculation method described in S121 above, R is calculated... i_max R i_min satisfy:

[0027]

[0028] Among them, R i-1_max R represents the maximum distance of the (i-1)th distance segment. i_min This represents the minimum distance of the i-th distance segment;

[0029] S123. Repeat step S122 to perform distance segmentation iteration until the minimum distance R of the first segment is reached. 1_min When the range is less than the radar's blind zone, all range segment identification waveforms are acquired, and an identification waveform library is formed based on all acquired range segment identification waveforms.

[0030] Preferably, the step of adaptively retrieving a recognition waveform from the recognition waveform library that matches the target track recognition request information based on the target track recognition request information includes:

[0031] Extract the distance information of the target track from the target track identification request information, and adaptively select a matching identification waveform from the identification waveform library based on the distance information of the target track, so that the distance of the target track falls exactly within the distance range of the matching identification waveform.

[0032] Preferably, the extraction of micro-motion features from the target trajectory based on the phase-adapted recognition waveform includes:

[0033] The direct digital frequency synthesizer generates a corresponding transmission signal based on the phase-matched identification waveform. The signal processor performs pulse compression, multi-pulse MTD, and constant false alarm rate detection through a digital signal processor (DSP) to achieve coherent accumulation of the received identification waveform and complete the extraction of micro-motion features from the range-Doppler two-dimensional spectrum.

[0034] Secondly, this invention also proposes a UAV target micro-motion feature extraction system based on adaptive variable waveform, the system including a task scheduling module, the task scheduling module including:

[0035] The waveform library acquisition submodule is used to construct the waveform library required for extracting micro-motion features; the waveform library includes waveforms for all range segments within the range for radar to extract target micro-motion features;

[0036] The waveform adaptive retrieval submodule is used to receive target trajectory recognition request information from the target to which micro-motion features are to be extracted, and adaptively retrieve a recognition waveform that matches the target trajectory recognition request information from the recognition waveform library based on the target trajectory recognition request information; the target trajectory recognition request information includes the distance information of the target trajectory.

[0037] The micro-motion feature extraction submodule is used to extract the micro-motion features of the target track based on the matching recognition waveform.

[0038] Preferably, the waveform recognition library acquisition submodule constructs the waveform recognition library by including:

[0039] S11. Divide the range R of the target micro-motion features extracted by the radar into N range segments, where the coverage area of ​​each range segment is R. i_min ~R i_max (i = 1, 2, ..., N); where R i_min R represents the minimum distance value of the i-th segment; i_max This represents the maximum distance value of the i-th segment;

[0040] S12. Calculate the waveform parameters of the identification waveform required for extracting micro-motion features in each range segment according to the radar equation and the requirements for extracting micro-motion features. Construct an identification waveform library based on the identification waveforms corresponding to all calculated waveform parameters. The waveform parameters include radar pulse repetition frequency, pulse width, number of pulses, and bandwidth.

[0041] Preferably, S12 includes:

[0042] S121. Initialize the farthest distance segment R N_min ~R N_max The waveform parameters of the identified waveform; among them, the radar pulse repetition frequency parameter of the identified waveform at the farthest range needs to meet the following conditions:

[0043]

[0044] Where c is the speed of light, f D Ω is the maximum speed at the tip of the UAV rotor, L is the length of the UAV rotor blade, Ω is the rotational speed of the UAV rotor, θ is the angle between the radar line of sight and the plane of rotation, and λ is the radar wavelength.

[0045] The pulse width τ of the waveform identified at the furthest distance N Pulse count M N and bandwidth B N The calculation formula is:

[0046]

[0047] The maximum distance R of the farthest segment N_max Conditions to be met:

[0048]

[0049] Among them, M max The maximum number of pulses accumulated by the radar; B max σ is the maximum signal bandwidth of the radar; P is the target cross-sectional area; t τ is the transmit power; G is the transmit pulse width; t G r These represent the antenna transmit gain and receive gain, respectively; λ is the wavelength; k is the Boltzmann constant; T0 is the temperature; F n denoted as receiver noise figure (SNR). min The signal-to-noise ratio for detecting micro-motion characteristics; L represents the radar system loss;

[0050] Minimum distance R of the farthest segment N_min Conditions to be met:

[0051]

[0052] S122, Order R i-1_max =R i_min Adjust the duty cycle D of the i-th distance segment. i Pulse count M i Signal bandwidth B i Following the iterative calculation method described in S121 above, R is calculated... i_max R i_min satisfy:

[0053]

[0054] Among them, R i-1_max R represents the maximum distance of the (i-1)th distance segment. i_min This represents the minimum distance of the i-th distance segment;

[0055] S123. Repeat step S122 to perform distance segmentation iteration until the minimum distance R of the first segment is reached. 1_min When the range is less than the radar's blind zone, all range segment identification waveforms are acquired, and an identification waveform library is formed based on all acquired range segment identification waveforms.

[0056] Preferably, the adaptive waveform retrieval submodule adaptively retrieves a waveform from the waveform library that matches the target track identification request information based on the target track identification request information, including:

[0057] Extract the distance information of the target track from the target track identification request information, and adaptively select a matching identification waveform from the identification waveform library based on the distance information of the target track, so that the distance of the target track falls exactly within the distance range of the matching identification waveform.

[0058] Preferably, the micro-motion feature extraction submodule completes the micro-motion feature extraction of the target track based on the adapted recognition waveform, including:

[0059] The direct digital frequency synthesizer generates a corresponding transmission signal based on the phase-matched identification waveform. The signal processor performs pulse compression, multi-pulse MTD, and constant false alarm rate detection through a digital signal processor (DSP) to achieve coherent accumulation of the received identification waveform and complete the extraction of micro-motion features from the range-Doppler two-dimensional spectrum.

[0060] (III) Beneficial Effects

[0061] This invention provides a method and system for extracting micro-motion features of UAV targets based on adaptive variable waveforms. Compared with existing technologies, it has the following advantages:

[0062] This invention proposes an adaptive variable waveform-based UAV target micro-motion feature extraction technology. First, it constructs a recognition waveform library containing recognition waveforms for all distance segments within the range from which radar can extract target micro-motion features. Then, based on the recognition request information of the target trajectory of the target to be extracted, it adaptively retrieves a recognition waveform from the aforementioned recognition waveform library that matches the target trajectory's recognition request information. Finally, it completes the micro-motion feature extraction of the target trajectory based on the matched recognition waveform. Compared to existing technologies, this invention achieves micro-motion feature extraction for extremely weak micro-motion targets. Furthermore, it can extract target micro-motion features based on adaptive variable waveforms according to requests issued by targets at different distance segments, thus providing a solution for practical engineering applications of UAV identification using micro-motion features. Attached Figure Description

[0063] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0064] Figure 1 This is a flowchart of a method for extracting micro-motion features of UAV targets based on adaptive variable waveforms according to the present invention;

[0065] Figure 2 This is an embodiment of an adaptive variable waveform-based method for extracting micro-motion features of UAV targets, as described in this invention.

[0066] Figure 3 The RD spectrum of the UAV target from which the micro-motion features are to be extracted in this embodiment of the invention is measured at a distance of 5.9 km.

[0067] Figure 4 This is the Doppler spectrum of the 35th distance unit in this embodiment of the invention;

[0068] Figure 5 This is the signal envelope spectrum after autocorrelation of the modulation spectrum in an embodiment of the present invention. Detailed Implementation

[0069] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention are described clearly and completely. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0070] This application provides a method and system for extracting micro-motion features of UAV targets based on adaptive variable waveforms, which solves the problem that existing technologies cannot extract micro-motion features of UAV targets from a distance based on adaptive variable waveforms, thereby achieving the practical engineering application of using micro-motion features for UAV identification.

[0071] The technical solution in this application is to solve the above-mentioned technical problems, and the general idea is as follows:

[0072] To extract the micro-motion features of targets at different distances—that is, to extract the micro-motion features of targets based on adaptive variable waveforms—and thus achieve the goal of using micro-motion features for UAV identification, this technical solution first constructs a recognition waveform library containing recognition waveforms for all distances within the range from which radar can extract the micro-motion features of targets. Then, based on the recognition request information of the target trajectory from which the micro-motion features of the target to be extracted, a recognition waveform that matches the recognition request information of the target trajectory is adaptively retrieved from the aforementioned recognition waveform library. Finally, based on the matched recognition waveform, the micro-motion features of the target trajectory from which the micro-motion features of the target to be extracted are extracted. This technical solution not only solves the problem that existing technologies struggle to extract micro-motion features from extremely weak micro-motion targets, but also further enhances the flexibility of radar operation by extracting the micro-motion features of targets based on adaptive variable waveforms, avoiding the waste of radar time and hardware resources.

[0073] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0074] Example 1:

[0075] Phased array radars offer flexible and controllable beam pointing, dwell time, spatial radiation power, and radar resources. Track and Search (TAS) is a typical operating mode for phased array radars. Based on the time-separation principle, phased array radars insert tracking tasks within search tasks, with the two operating states alternating independently. This allows for simultaneous searching of a designated area and precise tracking of multiple targets. This mode fully leverages the flexibility of phased array radars. Similar to inserting tracking tasks into TAS, this invention addresses the problem that general anti-UAV radars struggle to extract micro-motion characteristics of rotor targets from single detection waveforms. It inserts an identification task and then extracts the micro-motion characteristics of rotor targets using specialized identification waveforms.

[0076] Firstly, this invention proposes a method for extracting micro-motion features of UAV targets based on adaptive variable waveforms, see [link to relevant documentation]. Figure 1 The method includes:

[0077] S1. Construct a recognition waveform library required for extracting micro-motion features; the recognition waveform library includes recognition waveforms for all range segments within the range for radar to extract target micro-motion features;

[0078] S2. Receive target trajectory recognition request information from the target whose micro-motion features are to be extracted, and adaptively retrieve a recognition waveform that matches the target trajectory recognition request information from the recognition waveform library based on the target trajectory recognition request information; the target trajectory recognition request information includes the distance information of the target trajectory;

[0079] S3. Based on the matching identification waveform, the micro-motion feature extraction of the target trajectory to be extracted is completed.

[0080] As can be seen, the UAV target micro-motion feature extraction technology based on adaptive variable waveform proposed in this embodiment first constructs a recognition waveform library including recognition waveforms for all distance segments within the range of radar-extracted target micro-motion features; then, based on the recognition request information of the target trajectory of the target to be extracted, an recognition waveform that matches the recognition request information of the target trajectory is adaptively retrieved from the above recognition waveform library; finally, the micro-motion feature extraction of the target trajectory to be extracted is completed based on the matched recognition waveform. Compared with the prior art, this embodiment realizes the extraction of micro-motion features from extremely weak micro-motion targets. Furthermore, it can also extract target micro-motion features based on adaptive variable waveforms according to the requests issued by targets at different distance segments, thus providing a solution for practical engineering applications of UAV identification using micro-motion features.

[0081] The following is in conjunction with the appendix Figure 1-2 The following details the implementation process of an embodiment of the present invention, including explanations of the specific steps S1-S3.

[0082] The RCS (radar cross section) of micro-drones is very small; the RCS of their rotors is often tens of dB lower than that of the fuselage. In this embodiment, the DJI Phantom 4 is used as a typical target for micro-motion feature extraction. The DJI Phantom 4's fuselage RCS is only 0.01 square meters, and its rotor RCS is even smaller. Furthermore, there are significant differences in drone materials, and the rotor's RCS can be 30-40 dB lower than the fuselage's due to the drone's motion and flight attitude.

[0083] Because the RCS of a UAV rotor is much weaker than that of the fuselage, and is significantly affected by factors such as material composition and target motion attitude, no radar system or data currently publicly defines the extraction range for rotor micro-motion features. Therefore, to meet the need for extracting micro-motion features at different distances, phased array radars pre-store a recognition waveform library in their mission scheduling module. Once the target has been stably detected by the phased array radar, the mission scheduling module determines in real-time which recognition waveform to transmit to the target based on the target's trajectory request information (speed, distance, altitude). To ensure the extraction of micro-motion features from long-range targets, the recognition waveform uses a high duty cycle, large signal bandwidth, and high pulse count to enhance the echo intensity of the UAV rotor. When the target is closer, the echo signal-to-noise ratio is relatively high, so the dwell time and bandwidth of the recognition waveform can be appropriately shortened to reduce the dwell time of the recognition beam and the computational load of the phased array radar, achieving optimal radar system performance. Based on this,

[0084] This embodiment proposes a method for extracting micro-motion features of UAV targets based on adaptive variable waveforms. The method specifically includes the following steps:

[0085] S1. Construct a recognition waveform library required for extracting micro-motion features; the recognition waveform library includes recognition waveforms for all range segments within the range of radar extraction of target micro-motion features.

[0086] S11. Divide the range R of the target micro-motion features extracted by the radar into N range segments, where the coverage area of ​​each range segment is R. i_min ~R i_max (i = 1, 2, ..., N); where R i_min R represents the minimum distance value of the i-th segment; i_max This represents the maximum distance value of the i-th segment.

[0087] First, a waveform recognition library is established in the task scheduling module. The range R required by the anti-UAV phased array radar to extract the target micro-motion characteristics is divided into N segments, with each segment covering a range of R. i_min ~R i_max (i = 1, 2, ..., N). Where R i_min R represents the minimum distance value of the i-th segment; i_max This represents the maximum distance value of the i-th segment.

[0088] S12. Calculate the waveform parameters of the identification waveform required for extracting micro-motion features in each range segment according to the radar equation and the energy requirements for extracting micro-motion features. Then, construct an identification waveform library based on the identification waveforms corresponding to all calculated waveform parameters. The waveform parameters include radar pulse repetition frequency, pulse width, number of pulses, and bandwidth.

[0089] Based on the radar equations and the requirements for extracting micro-motion features of the UAV rotor, the radar pulse repetition frequency Freq required for extracting micro-motion features at each range segment is calculated. i Parameters, pulse width τ i Pulse count M i Bandwidth B i Waveform parameters, etc. Specifically, step S12 includes the following sub-steps in its implementation:

[0090] S121. Initialize the farthest distance segment (R) N_min ~R N_max The waveform parameters corresponding to the identified waveform.

[0091] Calculate the farthest distance segment (R) required for extracting micro-motion features. N_min ~R N_max The radar pulse repetition frequency parameter Freq corresponding to the identified waveform. N Pulse width parameter τ N Pulse number parameter M N Bandwidth parameter B N Waveform parameters, etc.

[0092] To meet the requirements for extracting rotor micro-motion characteristics and the maximum distance, the maximum distance segment (R) N_min ~R N_max The pulse repetition frequency Freq of the identified waveform N The following conditions must be met:

[0093]

[0094] Where c is the speed of light, f D Ω is the maximum speed at the tip of the UAV rotor, L is the length of the UAV rotor blade, Ω is the rotational speed of the UAV rotor, θ is the angle between the radar line of sight and the plane of rotation, and λ is the radar wavelength.

[0095] In addition, the pulse width τ of the waveform identified in the farthest distance segment mentioned above N M N B N The duty cycle, pulse number (satisfying a power of 2), and maximum bandwidth supported by the radar hardware system are determined as follows:

[0096]

[0097] Among them, D max Indicates the maximum duty cycle supported by the system hardware; M max Indicates the maximum duty cycle supported by the system's processing capacity; B max This indicates the maximum bandwidth supported by the system hardware.

[0098] To guide the practical engineering implementation of rotor micro-motion feature extraction for small UAVs or to propose corresponding indicators, given the rotor RCS size (typically RCSσ = 0.0001) and the rotor extraction signal-to-noise ratio (compared to the conventional detection signal-to-noise ratio, the detection signal-to-noise ratio can be appropriately reduced, typically SNR), this approach aims to improve the efficiency of rotor extraction. min In the case of 10dB, according to the radar equations, the maximum range R of the farthest range segment is... N_max The following conditions must be met:

[0099]

[0100] Among them, M max The maximum number of pulses accumulated by the radar; B max σ is the maximum signal bandwidth of the radar; P is the target cross-section (RCS); t τ is the transmit power; G is the transmit pulse width; t G r These represent the antenna transmit gain and receive gain, respectively; λ is the wavelength; k is the Boltzmann constant; T0 is the temperature; F n denoted as receiver noise figure (SNR). min The signal-to-noise ratio (in dB) is used to detect micro-motion characteristics; L is the radar system loss.

[0101] The minimum distance of the farthest segment must satisfy:

[0102]

[0103] S122, Order R i-1_max =R i_min Adjust the duty cycle D of the i-th distance segment. i Pulse count M i Signal bandwidth B i Following the iterative calculation method described in S121 above, R is calculated... i_max R i_min Each of the following conditions must be met:

[0104]

[0105] Among them, R i-1_max R represents the maximum distance of the (i-1)th distance segment. i_min This represents the minimum distance of the i-th distance segment.

[0106] To simplify calculations, in a preferred implementation of this embodiment, M is... i B i Decreasing the value by multiples of 2 reduces the dwell time of the identification beam and the computational load of signal processing for phased array radar, thus avoiding unnecessary waste of radar resources.

[0107] S123. Repeat step S122 to perform distance segmentation iteration until the minimum distance R of the first segment is reached. 1_min When the range is less than the radar's blind zone, all range segment identification waveforms are acquired, and an identification waveform library is formed based on all acquired range segment identification waveforms.

[0108] Repeat step S122 above to perform distance segmentation iteration until the minimum distance R of the first segment is reached. 1_min When the range is less than the radar's blind zone, the waveform parameters of the identification waveforms for all range segments are calculated. At this point, the identification waveforms for all range segments are obtained, and an identification waveform library is built based on these identification waveforms.

[0109] S2. Receive target trajectory recognition request information from the target whose micro-motion features are to be extracted, and adaptively retrieve a recognition waveform that matches the target trajectory recognition request information from the recognition waveform library based on the target trajectory recognition request information; the target trajectory recognition request information includes the distance information of the target trajectory.

[0110] Conventional anti-drone radar, operating on a single waveform, struggles to simultaneously meet the requirements of detection and identification. Phased array radar, by inserting identification waveforms that satisfy the extraction of target micro-motion features, achieves the extraction of micro-motion features of the drone rotor. Specifically,

[0111] Once a target from which micro-motion features are to be extracted has been searched and batched by a phased array radar via TWS and tracked with high precision and high data using TAS, the user can initiate an identification request on the display interface. At this time, the radar, while performing routine search and tracking tasks, inserts an identification task and transmits an identification waveform to extract micro-motion features. After receiving the identification request information from the target track that has already established a track, the task scheduling module adaptively determines which identification waveform to retrieve from the identification waveform library for subsequent micro-feature extraction based on the target track request information. The target track request information generally includes speed, distance, and altitude. During the decision-making process, the distance information of the target track in the identification request information is extracted, and an appropriate identification waveform is adaptively selected from the identification waveform library based on the target track's distance information. This ensures that the target track's distance falls within the range of the appropriate identification waveform, thus matching the most suitable identification waveform for the feature extraction of the target.

[0112] S3. Based on the matching identification waveform, the micro-motion feature extraction of the target trajectory to be extracted is completed.

[0113] The direct digital frequency synthesizer (DDS) generates a corresponding transmission signal based on the matched identification waveform. The signal processor performs pulse compression, multi-pulse MTD, and constant false alarm rate detection through the digital signal processor (DSP) to achieve coherent accumulation of the received identification waveform and to extract the micro-motion features of the target track from the range-Doppler (RD) two-dimensional spectrum.

[0114] For other target applications whose micro-motion features need to be extracted, the task scheduling module automatically retrieves and modifies the recognition waveform from the recognition waveform library in real time based on the feedback distance, so as to ensure that the waveform transformation is timely and accurate, thereby quickly and accurately extracting micro-motion features.

[0115] This completes the entire process of the UAV target micro-motion feature extraction method based on adaptive variable waveform in this embodiment.

[0116] To verify the effectiveness of the UAV target micro-motion feature extraction method based on adaptive variable waveform proposed in this invention, we will demonstrate it through experiments below.

[0117] Using a certain S-band anti-drone radar as an example, the test subject was a DJI Phantom 4 drone. The drone's four rotors are primarily made of carbon fiber and plastic. Each rotor is 0.24m long, and the blade length from the rotor center to the tip is L = 0.12m. The key technical parameters of the tested anti-drone radar are shown in Table 1. Under the conventional search waveforms shown in Table 2, although the radar can detect quadcopter drones at a range of 9km, it lacks the ability to extract the micro-motion characteristics of small rotorcraft.

[0118] The anti-drone radar technical parameters listed in Table 1

[0119] Transmit pulse width (µs)_τ 3.5 Transmit gain (dB)_Gt 19.32 Receive gain (dB)_Gr 23.09 Accumulated pulse count (number) _Gd 256.00 Transmit frequency (MHz)_f 2700

[0120] To improve the radar's target recognition capability, a system was established based on the radar's hardware capabilities.

[0121] The identification waveform library in Table 2 is shown in Table 2. The calculation is based on the following formula:

[0122]

[0123] The PRF of the selected identification waveform is 5000Hz. In order to maximize the radar's micro-motion feature extraction capability, waveform 2 uses the maximum bandwidth, pulse number, and duty cycle supported by the radar's computing hardware. This waveform also represents the farthest micro-motion feature extraction distance that the radar can achieve. When the target falls to other distances, the task scheduling module calls the corresponding identification waveform from the identification waveform library according to the target's identification request distance.

[0124] The waveform parameters used by the anti-drone radar described in Table 2

[0125]

[0126] The effectiveness of the technical solution in this embodiment is demonstrated below through actual test results. During the test, the DJI quadcopter flew along the radial direction of the radar at a typical cruising speed (10m / s ± 3m / s). After the target formed a stable track at 5.9km, a request for identification of the drone target was initiated, and the following results were obtained: Figure 3 The RD two-dimensional spectrum is shown. The echo of the UAV fuselage is marked with a circle. Frequency modulation phenomenon can be clearly seen on both sides of the fuselage (marked with rectangles). The UAV is located on the 35th distance cell after the signal processing window is opened.

[0127] Figure 4 For the Doppler spectrum at distance unit 35, see [link / reference]. Figure 4 The modulated Doppler spectrum of the UAV, composed of a series of line spectra in the frequency domain, can be seen more clearly. Calculations show that the target body spectral line corresponds to a target velocity of -8.91 m / s, which is close to the UAV's cruising speed of 10 m / s ± 3 m / s. Furthermore, based on the Hilbert transform, the modulated spectrum after autocorrelation can be further enveloped, allowing observation of the Doppler interval from the envelope spectrum, and thus estimation of detailed information such as the rotational speed and blade length of the rotating blades. Figure 5 (in, Figure 5 'a' represents the envelope spectrum of the signal from 0 to 1500 Hz. Figure 5 (b indicates that the spectrum after 300Hz is magnified and displayed). The solid black line in the figure represents the main frequency shift of the fuselage. The black and gray dashed lines mark two sets of impact peaks, representing the rotation of the two sets of propeller blades, respectively. From the figure, the Doppler frequencies corresponding to the signals of the black dashed lines are 358.9Hz, 539.6Hz, 712.9Hz, 893.6Hz, 1072Hz, and 1250Hz. It can be further calculated that the rotational speed of this set of propeller blades is 89r / s, and the blade length is 10.42cm, which is close to the true value of the blade. It can be seen that the method of this embodiment can extract effective micro-motion features for target classification and recognition.

[0128] In addition, from Figure 4It can be seen that the rotor echo in the actual detection was more than 30 decibels weaker than the fuselage echo. The main reasons are as follows: (1) The DJI Phantom 4 used in the test is made of carbon fiber and plastic. Compared with the metal material used in some drones, the RCS of the rotor is smaller; (2) Unlike many tests in which the drone is in a hovering state, this test was conducted during the movement of the drone. The attitude change of the drone can be more drastic, and the energy of the rotor echo may be weaker. Since the rotor echo is several orders of magnitude smaller than the fuselage echo, it is difficult for general anti-drone radar to extract the micro-motion characteristics of micro-small drone targets in conventional processing. By inserting the segmented identification waveforms shown in Table 2 in the work, this embodiment solves the problem that it is difficult for ordinary anti-drone radar to simultaneously meet the requirements of detection and identification tasks under the single waveform. Based on the method of extracting target micro-motion characteristics by increasing duty cycle, signal bandwidth, increasing pulse number, and appropriately reducing detection signal-to-noise ratio proposed in this embodiment, it points the way for the engineering long-distance extraction of target micro-motion characteristics.

[0129] Example 2:

[0130] Secondly, the present invention also provides a UAV target micro-motion feature extraction system based on adaptive variable waveform, the system comprising:

[0131] The waveform library acquisition submodule is used to construct the waveform library required for extracting micro-motion features; the waveform library includes waveforms for all range segments within the range for radar to extract target micro-motion features;

[0132] The waveform adaptive retrieval submodule is used to receive target trajectory recognition request information from the target to which micro-motion features are to be extracted, and adaptively retrieve a recognition waveform that matches the target trajectory recognition request information from the recognition waveform library based on the target trajectory recognition request information; the target trajectory recognition request information includes the distance information of the target trajectory.

[0133] The micro-motion feature extraction submodule is used to extract the micro-motion features of the target track based on the matching recognition waveform.

[0134] Optionally, the waveform recognition library acquisition submodule constructs the waveform recognition library by including:

[0135] S11. Divide the range R of the target micro-motion features extracted by the radar into N range segments, where the coverage area of ​​each range segment is R. i_min ~R i_max (i = 1, 2, ..., N); where R i_min R represents the minimum distance value of the i-th segment; i_max This represents the maximum distance value of the i-th segment;

[0136] S12. Calculate the waveform parameters of the identification waveform required for extracting micro-motion features in each range segment according to the radar equation and the requirements for extracting micro-motion features. Construct an identification waveform library based on the identification waveforms corresponding to all calculated waveform parameters. The waveform parameters include radar pulse repetition frequency, pulse width, number of pulses, and bandwidth.

[0137] Optionally, S12 includes:

[0138] S121. Initialize the farthest distance segment R N_min ~R N_max The waveform parameters of the identified waveform; among them, the radar pulse repetition frequency parameter of the identified waveform at the farthest range needs to meet the following conditions:

[0139]

[0140] Where c is the speed of light, f D Ω is the maximum speed at the tip of the UAV rotor, L is the length of the UAV rotor blade, Ω is the rotational speed of the UAV rotor, θ is the angle between the radar line of sight and the plane of rotation, and λ is the radar wavelength.

[0141] The pulse width τ of the waveform identified at the furthest distance N Pulse count M N and bandwidth B N The calculation formula is:

[0142]

[0143] The maximum distance R of the farthest segment N_max Conditions to be met:

[0144]

[0145] Among them, M max The maximum number of pulses accumulated by the radar; B max σ is the maximum signal bandwidth of the radar; P is the target cross-sectional area; t τ is the transmit power; G is the transmit pulse width; t G r These represent the antenna transmit gain and receive gain, respectively; λ is the wavelength; k is the Boltzmann constant; T0 is the temperature; F n denoted as receiver noise figure (SNR). min The signal-to-noise ratio for detecting micro-motion characteristics; L represents the radar system loss;

[0146] Minimum distance R of the farthest segment N_min Conditions to be met:

[0147]

[0148] S122, Order R i-1_max =R i_min Adjust the duty cycle D of the i-th distance segment. i Pulse count M i Signal bandwidth B i Following the iterative calculation method described in S121 above, R is calculated... i_max R i_min satisfy:

[0149]

[0150] Among them, R i-1_max R represents the maximum distance of the (i-1)th distance segment. i_min This represents the minimum distance of the i-th distance segment;

[0151] S123. Repeat step S122 to perform distance segmentation iteration until the minimum distance R of the first segment is reached. 1_min When the range is less than the radar's blind zone, all range segment identification waveforms are acquired, and an identification waveform library is formed based on all acquired range segment identification waveforms.

[0152] Optionally, the adaptive waveform retrieval submodule adaptively retrieves a waveform from the waveform library that matches the target track identification request information based on the target track identification request information, including:

[0153] Extract the distance information of the target track from the target track identification request information, and adaptively select a matching identification waveform from the identification waveform library based on the distance information of the target track, so that the distance of the target track falls exactly within the distance range of the matching identification waveform.

[0154] Optionally, the micro-motion feature extraction submodule, based on the adapted recognition waveform, completes the micro-motion feature extraction of the target trajectory from which the micro-motion features to be extracted, including:

[0155] The direct digital frequency synthesizer generates a corresponding transmission signal based on the phase-matched identification waveform. The signal processor performs pulse compression, multi-pulse MTD, and constant false alarm rate detection through a digital signal processor (DSP) to achieve coherent accumulation of the received identification waveform and complete the extraction of micro-motion features from the range-Doppler two-dimensional spectrum.

[0156] It is understood that the UAV target micro-motion feature extraction system based on adaptive variable waveform provided in this embodiment of the invention corresponds to the above-mentioned UAV target micro-motion feature extraction method based on adaptive variable waveform. The explanation, examples, and beneficial effects of the relevant content can be referred to the corresponding content in the UAV target micro-motion feature extraction method based on adaptive variable waveform, and will not be repeated here.

[0157] In summary, compared with existing technologies, it has the following beneficial effects:

[0158] 1. This invention proposes an adaptive variable waveform-based UAV target micro-motion feature extraction technology. First, it constructs a recognition waveform library containing recognition waveforms for all distance segments within the range from which radar can extract target micro-motion features. Then, based on the recognition request information of the target trajectory of the target to be extracted, it adaptively retrieves a recognition waveform from the aforementioned recognition waveform library that matches the recognition request information of the target trajectory. Finally, it completes the micro-motion feature extraction of the target trajectory based on the matched recognition waveform. Compared to existing technologies, this invention achieves micro-motion feature extraction for extremely weak micro-motion targets. Furthermore, it can extract target micro-motion features based on adaptive variable waveforms according to requests issued by targets at different distance segments, thus providing a solution for practical engineering applications of UAV identification using micro-motion features.

[0159] 2. This invention divides the range of radar for extracting target micro-motion features into several range segments, then acquires the identification waveform for each range segment, and builds all the identification waveforms into an identification waveform library. This allows the invention to automatically call different identification waveforms from the identification waveform library for different range segments, thereby achieving the purpose of UAV target micro-motion feature extraction based on adaptive variable waveforms. This improves the flexibility of radar operation and avoids wasting radar time and hardware resources.

[0160] 3. This invention solves the problem that ordinary anti-UAV radar has difficulty simultaneously meeting the requirements of detection and identification tasks under the single waveform. The phased array achieves the extraction of UAV rotor micro-motion characteristics by inserting identification waveforms that meet the target micro-motion characteristics extraction.

[0161] 4. This invention proposes methods such as increasing the duty cycle, signal bandwidth, increasing the number of pulses, and appropriately reducing the detection signal-to-noise ratio to extract the micro-motion features of the target, pointing the way for the engineering of long-distance extraction of the micro-motion features of the target.

[0162] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0163] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for extracting micro-motion features of UAV targets based on adaptive variable waveform, characterized in that, The method includes: Construct a recognition waveform library required for extracting micro-motion features; the recognition waveform library includes recognition waveforms for all range segments within the range in which radar extracts target micro-motion features; In response to a target trajectory recognition request from a user on the display interface, an identification waveform that matches the target trajectory recognition request is adaptively retrieved from the identification waveform library based on the target trajectory recognition request; the target trajectory recognition request includes distance information of the target trajectory. Based on the phase-adapted recognition waveform, the micro-motion features of the target trajectory to be extracted are completed; The recognition waveform library required for constructing and extracting micro-motion features includes: S11. Range at which radar extracts the target's micro-motion characteristics Divided into There are several distance segments, and the coverage area of ​​each distance segment is... ~ , ;in, Indicates the first i The minimum distance value of a segment; Indicates the first i The maximum distance value of a segment; S12. Calculate the waveform parameters of the identification waveform required for extracting micro-motion features in each range segment according to the radar equation and the requirements for extracting micro-motion features. Construct an identification waveform library based on the identification waveforms corresponding to all calculated waveform parameters. The waveform parameters include radar pulse repetition frequency, pulse width, number of pulses, and bandwidth. S12 includes: S121. Initialize the farthest distance segment ~ The waveform parameters of the identified waveform; among them, the radar pulse repetition frequency parameter of the identified waveform at the farthest range needs to meet the following conditions: in, For the speed of light, The maximum speed at the tip of the drone's rotor. For the length of the drone rotor blades, For the rotational speed of the drone rotor, The angle between the radar line of sight and the plane of rotation, The radar wavelength; Pulse width of waveform identified at the furthest distance Pulse count and bandwidth The calculation formula is: In the formula, Indicates the maximum duty cycle supported by the system hardware; Maximum distance of the farthest segment Conditions to be met: in, The maximum number of pulses accumulated for the radar; This represents the maximum signal bandwidth of the radar. The target cross-sectional area; This refers to the transmission power. The transmit pulse width; , These are the antenna transmit gain and receive gain, respectively. Wavelength; Boltzmann's constant; For temperature; The receiver noise figure; To detect the signal-to-noise ratio of micro-motion features; For radar system losses; Minimum distance of the farthest segment Conditions to be met: S122, Order Adjustment of the first i Duty cycle of each distance segment Pulse count Signal bandwidth Iterative calculations are performed using the method described in S121 above, so that... , satisfy: in, Indicates the first i The maximum distance of the -1 distance segment Indicates the first i The minimum distance of the distance segment; S123. Repeat step S122 to perform distance segmentation iteration until the minimum distance of the first segment is reached. When the range is less than the radar's blind zone, all range segment identification waveforms are acquired, and an identification waveform library is formed based on all acquired range segment identification waveforms.

2. The method as described in claim 1, characterized in that, The adaptive retrieval of identification waveforms from the identification waveform library that are compatible with the target track identification request information based on the target track identification request information includes: Extract the distance information of the target track from the target track identification request information, and adaptively select a matching identification waveform from the identification waveform library based on the distance information of the target track, so that the distance of the target track falls exactly within the distance range of the matching identification waveform.

3. The method as described in claim 1, characterized in that, The extraction of micro-motion features from the target trajectory based on the phase-adapted identification waveform includes: The direct digital frequency synthesizer generates a corresponding transmission signal based on the phase-matched identification waveform. The signal processor performs pulse compression, multi-pulse MTD, and constant false alarm rate detection through a digital signal processor (DSP) to achieve coherent accumulation of the received identification waveform and complete the extraction of micro-motion features from the range-Doppler two-dimensional spectrum.

4. A UAV target micro-motion feature extraction system based on adaptive variable waveform, characterized in that, The system includes a task scheduling module, which includes: The waveform library acquisition submodule is used to construct the waveform library required for extracting micro-motion features; the waveform library includes waveforms for all range segments within the range for radar to extract target micro-motion features; The waveform adaptive retrieval submodule is used to respond to the target trajectory recognition request information issued by the user on the display interface, and adaptively retrieve the recognition waveform that matches the target trajectory recognition request information from the recognition waveform library based on the target trajectory recognition request information; the target trajectory recognition request information includes the distance information of the target trajectory; The micro-motion feature extraction submodule is used to extract the micro-motion features of the target track based on the matching recognition waveform; The waveform library required for constructing the waveform library acquisition submodule to extract micro-motion features includes: S11. Range at which radar extracts the target's micro-motion characteristics Divided into There are several distance segments, and the coverage area of ​​each distance segment is... ~ , ;in, This represents the minimum distance value of the i-th segment; Indicates the first i The maximum distance value of a segment; S12. Calculate the waveform parameters of the identification waveform required for extracting micro-motion features in each range segment according to the radar equation and the requirements for extracting micro-motion features. Construct an identification waveform library based on the identification waveforms corresponding to all calculated waveform parameters. The waveform parameters include radar pulse repetition frequency, pulse width, number of pulses, and bandwidth. S12 includes: S121. Initialize the farthest distance segment ~ The waveform parameters of the identified waveform; among them, the radar pulse repetition frequency parameter of the identified waveform at the farthest range needs to meet the following conditions: in, For the speed of light, The maximum speed at the tip of the drone's rotor. For the length of the drone rotor blades, For the rotational speed of the drone rotor, The angle between the radar line of sight and the plane of rotation, The radar wavelength; Pulse width of waveform identified at the furthest distance Pulse count and bandwidth The calculation formula is: In the formula, Indicates the maximum duty cycle supported by the system hardware; Maximum distance of the farthest segment Conditions to be met: in, The maximum number of pulses accumulated for the radar; This represents the maximum signal bandwidth of the radar. The target cross-sectional area; This refers to the transmission power. The transmit pulse width; , These are the antenna transmit gain and receive gain, respectively. Wavelength; Boltzmann's constant; For temperature; The receiver noise figure; To detect the signal-to-noise ratio of micro-motion features; For radar system losses; Minimum distance of the farthest segment Conditions to be met: S122, Order Adjustment of the first i Duty cycle of each distance segment Pulse count Signal bandwidth Iterative calculations are performed using the method described in S121 above, so that... , satisfy: in, Indicates the first i The maximum distance of the -1 distance segment Indicates the first i The minimum distance of the distance segment; S123. Repeat step S122 to perform distance segmentation iteration until the minimum distance of the first segment is reached. When the range is less than the radar's blind zone, all range segment identification waveforms are acquired, and an identification waveform library is formed based on all acquired range segment identification waveforms.

5. The system as described in claim 4, characterized in that, The adaptive waveform retrieval submodule adaptively retrieves identification waveforms from the identification waveform library that are compatible with the target track identification request information based on the target track identification request information, including: Extract the distance information of the target track from the target track identification request information, and adaptively select a matching identification waveform from the identification waveform library based on the distance information of the target track, so that the distance of the target track falls exactly within the distance range of the matching identification waveform.

6. The system as described in claim 4, characterized in that, The micro-motion feature extraction submodule extracts the micro-motion features of the target track based on the matched recognition waveform, including: The direct digital frequency synthesizer generates a corresponding transmission signal based on the phase-matched identification waveform. The signal processor performs pulse compression, multi-pulse MTD, and constant false alarm rate detection through a digital signal processor (DSP) to achieve coherent accumulation of the received identification waveform and complete the extraction of micro-motion features from the range-Doppler two-dimensional spectrum.

Images