Hovering unmanned aerial vehicle target detection method based on micro-Doppler effect

Abstract

The invention discloses a hovering unmanned aerial vehicle target detection method based on a micro-Doppler effect. The method comprises the following steps: firstly, receiving radar echo data, carrying out distance dimension pulse compression and moving target detection, then enhancing and extracting micro-Doppler features, judging whether the micro-Doppler features exceed a set threshold value or not, and obtaining a detection result whether an unmanned aerial vehicle target exists or not. According to the scheme, the micro-Doppler effect is generated by utilizing the fact that components such as a propeller of the hovering unmanned aerial vehicle rotate, the micro-Doppler characteristic of a target reflects the electromagnetic characteristic, the geometric structure and the motion characteristic of the target, firstly, a reference unit is selected, the micro-Doppler characteristic is extracted by applying a variance analysis method, and the micro-Doppler characteristic of the target is extracted; the target detection is carried out on the feature data, the calculation is simple, the robustness is high, and the hovering unmanned aerial vehicle target can be stably detected in a clutter environment.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of unmanned aerial vehicle detection, and particularly relates to a hovering unmanned aerial vehicle target detection method based on micro-Doppler effect. BACKGROUND

[0002] Low-altitude economy is a new comprehensive economic form, which takes low-altitude flight activities as the core. Therefore, unmanned flight activities are increasing, and taking photos and recording videos by carrying unmanned aerial vehicles during travel and play have become a new trend and fashion of public consumption and entertainment. However, the unmanned aerial vehicle illegal aerial photography and survey activities are increasing, especially the "black flight" and "random flight" activities ignoring the flight regulations, which interfere with the normal flight of civil aviation aircraft, reveal national defense secrets through aerial photography, and lead to safety accidents caused by out-of-control crashes, thus bringing risks to national security and social security. The safety problems caused by unmanned aerial vehicle "black flight" have attracted much attention.

[0003] In view of the problems of "black flight" and "random flight", anti-unmanned aerial vehicle technology and equipment are urgently needed, and anti-unmanned aerial vehicle system technology mainly includes detection, tracking, identification, interference or destruction and other links. To deal with unmanned aerial vehicles, it is necessary to know where the unmanned aerial vehicles are, which requires unmanned aerial vehicle detection technology.

[0004] The current mainstream detection technology is radar detection. The radar is sensitive to moving targets and not sensitive to static targets. When the unmanned aerial vehicle normally flies in the air, it is relatively easy to detect the unmanned aerial vehicle for the radar. The current difficulty is that it is not easy to detect the unmanned aerial vehicle when the unmanned aerial vehicle hovers in the air.

[0005] The current mainstream detection method is static clutter map detection. In the environment with clutter, the detection effect is greatly reduced, and even the hovering unmanned aerial vehicle target cannot be detected. However, when the unmanned aerial vehicle hovers, the propeller and other components of the target are rotating, and these rotating movements will produce micro-Doppler effect. SUMMARY

[0006] In view of the above problems, the purpose of the present application is to provide a hovering unmanned aerial vehicle target detection method based on micro-Doppler effect, which can stably detect the hovering unmanned aerial vehicle target in the clutter environment.

[0007] The specific technical scheme for realizing the purpose of the present application is as follows:

[0008] A hovering unmanned aerial vehicle target detection method based on micro-Doppler effect, comprising the following steps:

[0009] Step 1, receiving radar echo data;

[0010] Step 2, performing range dimension pulse compression on the echo data;

[0011] Step 3, performing moving target detection;

[0012] Step 4: Enhance and extract micro-Doppler features;

[0013] Step 5: Perform CFAR detection based on the extracted micro-Doppler features, and determine whether the detection results exceed the set threshold to obtain the detection results of whether there is a drone target.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0015] The present invention utilizes the fact that the propellers and other components of the hovering UAV are rotating, thereby generating the micro-Doppler effect. The micro-Doppler characteristics of the target reflect the electromagnetic properties, geometric structure and motion characteristics of the target. First, a reference cell is selected, and the micro-Doppler characteristics are extracted by the variance analysis method. Target detection is performed on the feature data. The calculation is simple, the robustness is high, and the hovering UAV target can be detected stably even in cluttered environments.

[0016] The present invention will be further described below with reference to specific embodiments. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of a hovering UAV target detection method based on the micro-Doppler effect according to the present invention.

[0018] Figure 2 This is a schematic diagram of the micro-Doppler time-frequency characteristics of the rotating blades of a drone in an embodiment of the present invention.

[0019] Figure 3 This is a schematic diagram of the distance-dimensional pulse compression result in an embodiment of the present invention.

[0020] Figure 4 This is a schematic diagram of moving target detection in an embodiment of the present invention.

[0021] Figure 5 This is a schematic diagram of the moving target detection results in an embodiment of the present invention.

[0022] Figure 6 This is a schematic diagram of channel 0 data in the RD graph of the moving target detection results in an embodiment of the present invention.

[0023] Figure 7 This is a schematic diagram of CFAR detection results based on channel 0 data in the RD diagram of the moving target detection results in the prior art of the present invention.

[0024] Figure 8 This is a schematic diagram showing the selection of reference units in an embodiment of the present invention.

[0025] Figure 9 This is a schematic representation of the F-distribution in an embodiment of the present invention.

[0026] Figure 10 This is the target Doppler spectrum diagram in an embodiment of the present invention.

[0027] Figure 11 This is a schematic diagram of the left reference cell and variance analysis results in an embodiment of the present invention.

[0028] Figure 12 This is a schematic diagram of the right-side reference cell and the variance analysis results in an embodiment of the present invention.

[0029] Figure 13 This is a schematic diagram of the micro-Doppler features extracted by the analysis of variance method in an embodiment of the present invention. Detailed Implementation

[0030] Example

[0031] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. The described embodiments are only some embodiments of the present invention, and 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.

[0032] As indicated in this application and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of explicitly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0033] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of this application. It should also be understood that, for ease of description, the dimensions of the various parts shown in the drawings are not drawn to actual scale. Techniques, methods, and devices known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and devices should be considered part of the specification. In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values. It should be noted that similar reference numerals and letters in the following drawings denote similar items; therefore, once an item is defined in one drawing, it need not be further discussed in subsequent drawings.

[0034] Combination Figure 1A target detection method for hovering unmanned aerial vehicles based on the micro-Doppler effect includes the following steps:

[0035] First, a mathematical model of the rotor blades is established. The UAV rotor has multiple scattering centers. For a rotor consisting of M blades, the M blades have M different initial rotation angles:

[0036] (1)

[0037] The radar echo signals of all blades are

[0038]

[0039] (2)

[0040] In the formula, R is the distance from the center of the rotor blade to the radar, and H is the height of the rotor blade. Let t be the angular velocity of the rotor blades rotating around the center of the fuselage, and t be time. Where λ is the wavelength of the electromagnetic wave, and β is the elevation angle of the target observed by the radar. It is the Sink function: when x=0, ;when hour, Phase function for:

[0041] (3)

[0042] In the formula, L is the length from the rotor center to the blade tip;

[0043] The temporal characteristic amplitude of a rotor blade is defined as follows:

[0044]

[0045] (4)

[0046] Based on the length and rotation speed of the leaf tip, the velocity of the leaf tip is obtained as follows: Therefore, the maximum Doppler frequency shift is ;

[0047] The Doppler characteristics caused by the modulation applied to the radar echo signal by the rotating blades are identified using the short-time Fourier transform (STFT). Since the time derivative of the signal phase function is the instantaneous frequency of the signal, the phase function in equation (3) is used... Differentiating with time as the variable, we can obtain the instantaneous Doppler frequency caused by the k-th rotating blade as:

[0048] (5)

[0049] Combination Figure 2 As can be seen from Equation 5, the Doppler frequency is modulated by the rotation rate using two sinusoidal functions. The Doppler modulation caused by the rotating rotor blades is considered a unique feature of hovering UAVs.

[0050] Step 1: Receive radar echo data;

[0051] A radar transmitter emits electromagnetic waves into the air through its antenna. These waves are reflected back after hitting a target in the air. The radar receiver receives the echo, performs down-conversion, and samples it in the fast time domain to obtain the radar echo signal. It can be described as:

[0052]

[0053] In the formula, For the target echo signal, For clutter, For noise;

[0054] The reflected echo from a target at a distance R can be described as follows:

[0055]

[0056] In the formula, A represents the amplitude of the radar target echo signal. Delay of the target echo signal ( (representing the speed of light) The Doppler frequency shift caused by the target's motion relative to the radar. For the target speed of motion, Indicates the carrier frequency. Let be the envelope function of the transmitted signal pulse. Generally, radars often use linear frequency modulated (LFM) signals as the transmitted signal. Then:

[0057]

[0058] In the formula, Represents a rectangle function. For frequency modulation slope, For signal bandwidth, This represents the pulse width.

[0059] Step 2: Perform range-dimensional pulse compression on the echo data:

[0060] Based on the principle of pulse compression, pulse compression is performed on the down-converted signal in the distance dimension within the fast time domain to obtain a pulse-compressed signal. ;

[0061]

[0062]

[0063]

[0064] in, Pulse compression signal for distance dimension Let R be the target echo reflected from point R, and A be the amplitude of the radar target echo signal. Delay of the target echo signal, At the speed of light, The Doppler frequency shift caused by the target's motion relative to the radar. For the target speed of motion, Indicates the carrier frequency. Let be the envelope function of the transmitted signal pulse. Represents a rectangle function. For frequency modulation slope, For signal bandwidth, The pulse width. To indicate complex conjugate, This indicates a convolution operation.

[0065] A schematic diagram of the pulse compression result of the target echo signal is shown below. Figure 3 As shown.

[0066] Step 3: Perform moving target detection, which involves performing an FFT transform on the slow-time domain signal of each range cell to obtain the range Doppler spectrum, such as... Figure 4 As shown; the RD (Range-Doppler) spectrum is obtained, as follows. Figure 5 As shown, the micro-Doppler feature extraction in this scheme is... Figure 5 Based on this, the maximum value is determined for each row of data, thus obtaining the final spectrum after variance analysis.

[0067] Channel 0 is in the middle of the RD spectrum. Hovering drones and stationary clutter are both on channel 0. Extracting data from channel 0 separately (e.g.) Figure 6 (As shown).

[0068] Step 4: Enhance and extract micro-Doppler features:

[0069] For hovering drone targets, the traditional detection method involves performing CFAR detection calculations on channel 0. The detection results are as follows: Figure 7 As shown, at a greater distance, the clutter is weaker, and the target at a distance of 139 has a stronger amplitude and is detected; at a closer distance, the clutter is very strong, and the target at a distance of 21 is masked by the strong clutter nearby and is not detected.

[0070] from Figure 7It is known that traditional methods detect hovering targets by performing CFAR detection on channel 0, which compares the reflection strength of the target and clutter. In areas with weak clutter, the target can be detected normally; in areas with strong clutter, the target is obscured by the clutter and is difficult to detect, resulting in unstable detection performance.

[0071] Therefore, in this application, the signal in the slow time domain of each distance unit is extracted, and the data of channel 0 is used as the current detection unit; P guard units are taken on both sides of the current detection unit, and then N reference units are taken on the outer side of each guard unit, as follows. Figure 8 As shown;

[0072] The selection of P protection units is to prevent target energy leakage to the reference unit and avoid overestimation of noise level. It is determined by the physical size of the target itself and the Doppler resolution of the radar.

[0073] The selection of N reference elements is for estimating the average power of the background noise around the detection element. The number needs to be balanced between estimation accuracy and environmental uniformity. Generally, 6 to 32 elements are selected on one side, but an appropriate N can be selected according to the clutter environment.

[0074] The selected reference cell data were processed using the following analysis of variance method to extract micro-Doppler features;

[0075] The Doppler frequency is modulated by the rotation rate with a sine function. Therefore, a spectral peak should appear at the frequency corresponding to the rotation rate on the Doppler spectrum. The period of the spectrum is searched by performing spectral window estimation on the Doppler spectrum amplitude sequence. The peak that appears periodically on the spectrum is the micro-Doppler frequency of the hovering UAV.

[0076] In spectral window estimation, the choice of cutoff point and window shape significantly affects the size and position of spectral peaks. Furthermore, the periodicity of the sequence and fluctuations in the periodogram caused by random sampling can lead to spurious peaks at corresponding frequencies. To avoid these problems, a method is proposed to identify and extract integer periodic values ​​from Doppler amplitude sequences using analysis of variance (ANOVA). The basic principle of ANOVA is to arrange time series signals at certain time intervals and use the comparison of within-group and between-group differences, i.e., applying the F-test of variance ratio, to determine whether a certain arrangement has significant periodicity. The specific process of this method includes:

[0077] (1) Assume the micro-Doppler spectrum amplitude sequence is Implicitly contains a length of The periodic pattern The slow-time domain signal data of the selected reference cell's distance cell is decomposed as follows:

[0078]

[0079] In the formula, , , i and j are temporary variables;

[0080] (2) Let , not exceeding The largest integer, calculate the data. mean square deviation and and the sum of squares :

[0081]

[0082]

[0083]

[0084] (3) Calculate the sum of squared errors between groups Sum of squared errors within groups

[0085]

[0086]

[0087] In the formula, , ;

[0088] (4) Calculate the mean square error and perform an F-test.

[0089]

[0090]

[0091]

[0092] set up For significance level, if If the test is significant, then it is acceptable. One full cycle of bit data, and vice versa. If the test is not significant, it is unacceptable. Assuming the data spans an entire period, then change The value is then returned to step (1), where The value can be obtained by looking up a table (F distribution table), or by using functions in Excel or MATLAB. Figure 9 for This refers to the F-distribution table with a 95% confidence level.

[0093] In this embodiment, the target distance from the measured data is used to calibrate the data of all Doppler channels, such as... Figure 10 As shown;

[0094] The results of the analysis of variance are as follows Figure 11 and 12 As shown; the variance analysis results of the left and right reference cells for all distance scales in channel 0 are maximized as follows: Figure 13 As shown, the target dimensionality features extracted based on the analysis of variance method are much stronger than the original spectrum, and hovering drones can be detected based on the enhanced results.

[0095] Step 5: Perform CFAR detection based on the extracted micro-Doppler features, and determine whether the detection result exceeds the set threshold to obtain the detection result of whether there is a drone target.

[0096] Based on the determined data period To obtain the spectrum after variance analysis, such as Figure 13 As shown, CFAR detection is performed based on the spectrum after variance analysis. According to the detection results, it is determined whether the set threshold is exceeded. If so, it is determined that there is a hovering drone target; otherwise, it is not.

[0097] The present invention utilizes the fact that the propellers and other components of the hovering UAV are rotating, thereby generating the micro-Doppler effect. The micro-Doppler characteristics of the target reflect the electromagnetic properties, geometric structure and motion characteristics of the target. First, a reference cell is selected, and the micro-Doppler characteristics are extracted by the variance analysis method. Target detection is performed on the feature data. The calculation is simple, the robustness is high, and the hovering UAV target can be detected stably even in cluttered environments.

[0098] This solution also provides a hovering UAV target detection module based on the micro-Doppler effect, including the following modules:

[0099] Echo data module: used to receive radar echo data;

[0100] Data processing module: used to perform range-dimensional pulse compression on echo data for moving target detection;

[0101] Micro-Doppler Feature Module: Used to enhance and extract micro-Doppler features;

[0102] Target detection module: Used to determine whether the micro-Doppler features exceed the set threshold value, and to obtain the detection result of whether there is a drone target.

[0103] This solution also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to perform the following steps:

[0104] Step 1: Receive radar echo data;

[0105] Step 2: Perform range-dimensional pulse compression on the echo data;

[0106] Step 3: Perform moving target detection;

[0107] Step 4: Enhance and extract micro-Doppler features;

[0108] Step 5: Determine whether the micro-Doppler feature exceeds the set threshold value to obtain the detection result of whether there is a drone target.

[0109] This solution also provides a computer-readable storage medium on which a computer program is stored, wherein the computer program, when executed by a processor, performs the following steps:

[0110] Step 1: Receive radar echo data;

[0111] Step 2: Perform range-dimensional pulse compression on the echo data;

[0112] Step 3: Perform moving target detection;

[0113] Step 4: Enhance and extract micro-Doppler features;

[0114] Step 5: Determine whether the micro-Doppler feature exceeds the set threshold value to obtain the detection result of whether there is a drone target.

[0115] The embodiments described above are merely one implementation method of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A hovering unmanned aerial vehicle target detection method based on micro-Doppler effect, characterized in that, The method comprises the following steps: Step 1, receiving radar echo data; Step 2, performing range dimension pulse compression on the echo data; Step 3, performing moving target detection; Step 4, enhancing and extracting micro-Doppler features; Step 5, performing CFAR detection based on the extracted micro-Doppler features, judging whether a set threshold is exceeded according to a detection result, and obtaining a result of whether a UAV target is detected.

2. The micro-Doppler effect based hovering UAV target detection method according to claim 1, wherein, The range dimension pulse compression on the echo data in the step 2 is specifically: ； ； ； where, is the range bin of the pulse compressed signal, is the target echo at range R, A is the radar target echo signal amplitude, is the target echo signal delay, is the speed of light, is the Doppler shift due to target relative radar motion, is the target velocity, denotes the carrier frequency, is the envelope function of the transmitted signal pulse, denotes the rectangular function, is the frequency modulation slope, is the signal bandwidth, is the pulse width, denotes the complex conjugate, denotes the convolution operation. 3.The micro-Doppler effect based hovering UAV target detection method of claim 1, wherein, The moving target detection in the step 3 is FFT transformation on a slow time domain signal of each range cell to obtain a range Doppler spectrum diagram.

4. The micro-Doppler effect based hovering UAV target detection method of claim 3, wherein, The enhancing and extracting micro-Doppler features in the step 4 are specifically: Extracting a signal of a slow time domain of each range cell, taking data of a 0 channel as a current detection cell, taking P protection cells on both sides of the current detection cell, and then taking N reference cells outside the protection cells; Extracting micro-Doppler features according to the reference cell data.

5. The micro-Doppler effect based hovering UAV target detection method of claim 4, wherein, The process of extracting the micro-Doppler features is: (1) Assume that the micro-Doppler spectrum amplitude sequence is , which implies a periodic rule with a length of , and , the distance unit slow-time domain signal data of the selected reference unit is decomposed as follows: ； In the formulae, , , ; (2) Set , the maximum integer not exceeding , calculate the mean square error of data and and the sum of squares : ； ； ； (3) Calculate the sum of squares of inter-group errors and the sum of squares of intra-group errors ; ； ； In the formulae, , ; (4) calculating an error mean square value and performing F test; ； ； ； Set For the significance level, if The test is significant, then accept One complete cycle of the bit data, otherwise The test is not significant, then reject The hypothesis that the data is one complete cycle, then change The value and return to step (1).

6. The micro-Doppler effect based hovering UAV target detection method of claim 5, wherein, The judging whether the micro-Doppler features exceed a set threshold in the step 5 is specifically: According to the determined data cycle , obtaining the frequency spectrum after variance analysis, performing CFAR detection based on the frequency spectrum, judging whether the threshold is exceeded according to the detection result, if yes, judging that the unmanned aerial vehicle target exists, otherwise, judging that the unmanned aerial vehicle target does not exist.

7. A hovering drone target detection module based on micro-Doppler effect, characterized in that, The method comprises the following modules: An echo data module for receiving radar echo data; A data processing module for performing range dimension pulse compression on the echo data and performing moving target detection; A micro-Doppler feature module for enhancing and extracting micro-Doppler features; A target detection module for performing CFAR detection based on the extracted micro-Doppler features, judging whether a set threshold is exceeded according to a detection result, and obtaining a result of whether a UAV target is detected.

8. A computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the computer program to realize the steps of the method in any one of claims 1-6.

9. A computer storable medium having stored thereon a computer program, characterized in that The computer program is executed by the processor to realize the steps of the method in any one of claims 1-6.

Images