CN113567976A - Unmanned aerial vehicle rotor detection system based on millimeter wave radar and detection method thereof - Google Patents
Unmanned aerial vehicle rotor detection system based on millimeter wave radar and detection method thereof Download PDFInfo
- Publication number
- CN113567976A CN113567976A CN202110778809.7A CN202110778809A CN113567976A CN 113567976 A CN113567976 A CN 113567976A CN 202110778809 A CN202110778809 A CN 202110778809A CN 113567976 A CN113567976 A CN 113567976A
- Authority
- CN
- China
- Prior art keywords
- unmanned aerial
- aerial vehicle
- distance
- radar
- unit
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- General Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention provides an unmanned aerial vehicle rotor detection system based on a millimeter wave radar and a detection method thereof, and relates to a design method of the unmanned aerial vehicle rotor detection system combining LFMCW radar range Doppler processing, peak detection and range-entropy detection. And performing multi-target detection and information extraction on the scattered echo signals to obtain the distance information of the unmanned aerial vehicle and the two-dimensional plane energy distribution of the Doppler information. The method comprises the steps of obtaining a distance area with an unmanned aerial vehicle target through a peak detection method, sequentially calculating amplitude distribution entropies of all distance units, carrying out peak detection on a distance-entropy curve, and integrating detection results of an RD diagram and the distance-entropy curve to finally realize detection of a rotor wing area of the unmanned aerial vehicle. On the one hand can distinguish unmanned aerial vehicle and other targets, and on the other hand can distinguish the apart from other apart from the unit of distance that unmanned aerial vehicle belongs to the rotor in the distance range, further realizes the detection of rotor position when realizing unmanned aerial vehicle position detection.
Description
Technical Field
The invention relates to an unmanned aerial vehicle rotor detection system based on a millimeter wave radar and a detection method thereof, which are used for realizing accurate detection of an unmanned aerial vehicle rotor part by combining peak detection processing according to the regional distribution characteristics of echo signals and belong to the field of digital signal processing.
Background
The micro motion is micro motion such as vibration and rotation of the target or a component, can reflect the detailed characteristics of the target motion, and can provide a new approach for target identification. With the continuous development of radar technology, technologies such as micro-motion measurement, feature extraction and identification of radar targets based on micro-doppler effect are receiving wide attention.
In recent years, the application technology of unmanned aerial vehicles is rapidly developed, and especially various rotary wing type unmanned aerial vehicles are widely applied to the civil field. The unmanned aerial vehicle technology brings great challenges to social safety while providing convenience for life of people, and is particularly critical to whether the unmanned aerial vehicle can be detected and early-warned. Rotor rotation is a unique feature of a rotor drone. The analysis to rotor rotating characteristic can carry out effective detection to unmanned aerial vehicle's running state, for example flying speed, flight attitude, whether hang article etc. and then make the early warning.
The traditional target detection method based on the radar echo range-Doppler (RD) diagram can obtain a better detection effect when the target echo energy is distributed intensively. Peak detection often uses Constant False Alarm Rate (CFAR) detection, requiring noise intensity to be estimated by several reference cells near the suspect cell. However, the high-speed rotation of the rotor wing can cause the target echo of the unmanned aerial vehicle to generate a large-range Doppler sideband, so that the energy distribution and aggregation performance is greatly reduced, the accuracy of noise estimation is seriously affected, and the detection rate is reduced. At present, the analysis of the micro-motion characteristics of the unmanned aerial vehicle is to integrally use the unmanned aerial vehicle as an analysis object, the micro-motion signals of all rotors are mixed together in the analysis process, and the rotors are difficult to separate for independent analysis. When radar range resolution is higher than inter-rotor spacing, theoretically each rotor is separable in the range dimension. If the difference of signal in rotor distance unit and the non-rotor distance unit can be analyzed, just can realize the separation of each rotor in the dimension of distance.
The entropy can be used for describing the concentration degree of energy distribution, and the signal distribution concentration degree is strongest in an area where only a translation target exists; the signal distribution concentration is second in the area where the micro-motion target exists; the noise zone signal concentration is lowest. The distribution characteristics of the scattering signals of the unmanned aerial vehicle can be effectively described by utilizing entropy, a wider Doppler sideband can be generated by a distance unit with a rotor wing, the signal dispersity is enhanced, and the entropy value is larger; the signal distribution of the non-rotor wing distance unit is concentrated, and the entropy value is smaller. Consequently, the entropy of application can distinguish rotor unmanned aerial vehicle target and other aerial targets on the one hand, and on the other hand when radar range resolution is high enough, also can be in the unmanned aerial vehicle main part within range more accurate discernment rotor place distance unit, and then make the snap judgments to unmanned aerial vehicle's rotor figure. Meanwhile, only the distance unit with the rotor wing is subjected to micro Doppler analysis, so that the signal to noise ratio of the signal can be effectively improved, and the quality of parameter estimation and feature extraction is improved.
Aiming at the above mentioned situations and practical application requirements, the invention provides a millimeter wave radar-based unmanned aerial vehicle rotor detection technology, which can realize accurate detection of the area where the rotor is located. The method is based on Linear Frequency Modulation Continuous Wave (LFMCW) millimeter wave radar measurement to obtain a target echo signal, then an entropy value of amplitude distribution in a unit is calculated according to a distance unit to obtain an entropy curve with the distance as an independent variable, and an area where a rotor wing is located is identified after CFAR detection.
Disclosure of Invention
The invention aims to provide an unmanned aerial vehicle rotor detection system based on a millimeter wave radar and a detection method thereof, and relates to a design method of the unmanned aerial vehicle rotor detection system combining LFMCW radar range Doppler processing, peak detection and range-entropy detection. And performing multi-target detection and information extraction on the scattered echo signals to obtain the distance information of the unmanned aerial vehicle and the two-dimensional plane energy distribution of the Doppler information. The method comprises the steps of obtaining a distance area with an unmanned aerial vehicle target through a peak detection method, sequentially calculating amplitude distribution entropies of all distance units, carrying out peak detection on a distance-entropy curve, and integrating detection results of an RD diagram and the distance-entropy curve to finally realize detection of a rotor wing area of the unmanned aerial vehicle. On the one hand can distinguish unmanned aerial vehicle and other targets, and on the other hand can distinguish the apart from other apart from the unit of distance that unmanned aerial vehicle belongs to the rotor in the distance range, further realizes the detection of rotor position when realizing unmanned aerial vehicle position detection.
The invention relates to an unmanned aerial vehicle rotor detection system based on a millimeter wave radar, which comprises a millimeter wave signal receiving and transmitting unit, an RD processing unit, an unmanned aerial vehicle distance range detection unit and a rotor wing area detection unit.
The millimeter wave signal receiving and sending unit is connected with the RD processing unit, is a front-end unit for RD processing, and is used for completing the radio frequency processing function of the front end. The unit controls related parameters of Linear Frequency Modulated Continuous Wave (LFMCW) Wave beams through a PC upper computer, generates radar baseband signals to be transmitted by a high-speed DA, generates high-Frequency millimeter Wave signals capable of being transmitted by a mixer, then transmits the signals to an antenna through a power amplifier, and radiates electromagnetic waves to a target space by the antenna. The transmitted millimeter wave signal is reflected by a target, received by an antenna, converted into a baseband signal through a filter and a mixer, and transmitted to a PC (personal computer) end for storage after AD (analog-to-digital) sampling, so that preparation is made for subsequent distance-Doppler processing.
The distance-Doppler RD processing unit is connected with the millimeter wave receiving and transmitting unit, the unmanned aerial vehicle distance range detection unit and the rotor wing area detection unit, inputs radar echo signals received by the millimeter wave receiving unit, and outputs AD sampling data arranged according to a time sequence. The unit rearranges data sequence according to fast time sampling point number, Chirp number and frame number defined by a radar system, wherein each dimension sequence of the data is fast time-slow time-frame number. And then Fourier transformation is carried out on the data along the fast time and the slow time respectively, extraction of RD information in the radar echo signal is completed, and an RD image of each frame of data is generated. The RD map generated will be used as input for the subsequent drone distance range detection unit and rotor area detection unit.
Unmanned aerial vehicle distance range detecting element, it is connected with RD processing unit and rotor regional detecting element for accomplish the detection of the whole distance distribution range of unmanned aerial vehicle. The unit superposes radar RD images along Doppler dimensions to obtain a one-dimensional range profile with a higher signal-to-noise ratio, and two end points of the unmanned aerial vehicle range distribution range are detected by adopting minimum of Constant False Alarm Detection (SO-CFAR) to prepare for subsequent rotor wing area Detection.
Rotor regional detecting element, it is connected with RD processing unit and unmanned aerial vehicle apart from the scope detecting element for accomplish the accurate of unmanned aerial vehicle rotor place region and draw. The unit calculates a distance-entropy curve based on a radar RD diagram, performs Ordered statistical Constant False Alarm Detection (OS-CFAR) on the distance-entropy curve to obtain a distance unit possibly having a micro-motion target, and finally determines the distance unit having a rotor wing by combining the distance range of the unmanned aerial vehicle.
The invention provides an unmanned aerial vehicle rotor detection method based on a millimeter wave radar, and the system flow is shown in figure 1 and summarized as follows:
the method comprises the following steps: transmitting millimeter wave electromagnetic signals
The millimeter wave radar equipment is installed outdoors, the wave beam of the radar is directed to the airspace where the unmanned aerial vehicle to be monitored is located, and then the millimeter wave radar is driven to transmit millimeter wave electromagnetic signals. The millimeter Wave radar adopts a Linear Frequency Modulated Continuous Wave (LFMCW) signal system, baseband signals are converted into radio Frequency analog signals after DA conversion, Frequency mixing and power amplification, and the radio Frequency analog signals are transmitted to an area to be observed by a transmitting antenna.
Step two: receiving scattered echo signals of unmanned aerial vehicle
Unmanned aerial vehicle scattering echo signal is the backscattering echo that is received by the transmission signal after unmanned aerial vehicle reflection, and the echo mainly includes unmanned aerial vehicle main part scattering signal and rotor scattering signal. In the first step, electromagnetic waves transmitted by the radar are reflected back to the radar equipment by the unmanned aerial vehicle, out-of-band noise is filtered through a filter, then intermediate-frequency signals and baseband signals are sequentially obtained through two stages of frequency mixers, and finally the intermediate-frequency signals and the baseband signals are converted into digital signals through AD sampling, namely, echo signals of the unmanned aerial vehicle.
Step three: target range-doppler analysis of each frame of data
And the echo of the unmanned aerial vehicle in the step two is packaged according to the format of the frame, and the data dimensionality is fast time, slow time and a frame sequence number. The echo contains the radar radial distance and radial velocity information of each target in the observation space, and the distance and velocity information of each target can be extracted through target distance-Doppler analysis. And Fourier transformation is carried out on each frame of signal respectively along the fast time dimension and the slow time dimension, so that a radar RD image can be obtained, and preparation is made for subsequent unmanned aerial vehicle distance range detection and rotor wing area detection.
Step four: determining a range of a distance distribution of a drone
And (3) superposing the radar RD images obtained in the third step along the Doppler dimension to obtain a one-dimensional range profile with a larger signal-to-noise ratio, and then performing one-dimensional SO-CFAR detection, wherein the geometric dimension of the unmanned aerial vehicle is larger than the radar range resolution, SO that a plurality of peak points can be detected in the area where the unmanned aerial vehicle is located. Finding the points farthest and closest to the radar in the peak points to define two end points of the drone distance distribution, namely the drone distance distribution range, and preparing for subsequent rotor area detection.
Step five: calculating a distance-entropy curve
And 8-bit uniform quantization is carried out on the radar RD image obtained in the third step, then the amplitude distribution entropy of each distance unit in the quantized RD image is calculated in sequence, and a distance-entropy curve is obtained to prepare for the subsequent detection of the rotor wing area of the unmanned aerial vehicle.
Step six: detecting rotor area
And comprehensively considering the distance-entropy curve calculated in the fifth step and the unmanned aerial vehicle distance distribution range detected in the fourth step. And detecting a plurality of peak values by applying the OS-CFAR on the distance-entropy curve, and if the peak value points appear in the unmanned aerial vehicle distance distribution range defined in the fourth step, determining that the peak value points are in the region where the rotor wing is located, otherwise, determining that the peak value points are in the region where only the unmanned aerial vehicle main body is located.
The advantages and the effects are as follows: the invention relates to an unmanned aerial vehicle rotor detection system based on a millimeter wave radar and a detection method thereof, which realize the detection of the area where the unmanned aerial vehicle rotor is located and mainly have the following advantages:
1) the millimeter wave radar is used, and the all-weather working capacity is achieved. The device still has good measurement capability in severe environments such as weak light and the like.
2) The drone RD graph is analyzed using range-entropy curves, which can further describe the detail differences in drone echo distribution in the range dimension. Can not only detect out unmanned aerial vehicle and place the region, can also further detect out the rotor and place the region.
3) Compared with the traditional unmanned aerial vehicle micro-motion characteristic analysis method, the method can accurately extract the signals of the area where the rotor wing is located, and improves the signal-to-noise ratio of the signals in the follow-up micro-motion analysis.
Drawings
FIG. 1 is a flow chart of the system of the present invention.
Fig. 2 is a block diagram of a system for transmitting and receiving millimeter wave electromagnetic signals by a radar device according to the present invention.
FIG. 3 is a flow chart of the distance-entropy curve calculation of the present invention.
Fig. 4 is a flow chart of the unmanned aerial vehicle distance distribution range and rotor wing distribution area detection according to the present invention.
The symbols in the figures are as follows:
RD range-doppler; a DAC digital-to-analog converter; an ADC analog-to-digital converter; an IF intermediate frequency filter; an LO local oscillator; a PA power amplifier; an LNA low noise amplifier; detecting the CFAR constant false alarm rate;
Detailed Description
Referring to fig. 1-4, the invention relates to a millimeter wave radar-based unmanned aerial vehicle rotor detection method, which comprises a millimeter wave signal receiving and transmitting unit, an RD processing unit, an unmanned aerial vehicle distance range detection unit and a rotor area detection unit. They are connected to each other.
The invention provides an unmanned aerial vehicle rotor detection method based on a millimeter wave radar, which comprises the following steps:
the method comprises the following steps: transmitting millimeter wave electromagnetic signals
The millimeter wave radar equipment is installed outdoors, the wave beam of the radar is directed to the airspace where the unmanned aerial vehicle to be monitored is located, and then the millimeter wave radar is driven to transmit millimeter wave electromagnetic signals. The millimeter wave radar adopts an LFMCW signal system, and the signal system can obtain the distance, speed and angle information of each target in an observation space through multi-dimensional Fourier transform. The case of transmitting an electromagnetic wave signal is shown in fig. 2.
Suppose the slope of the radar sweep frequency is k, the period is T, and the initial phase is
For any time t, the LFMCW radar transmits a signal sT(t) can be expressed as:
where A is the signal amplitude, f0For the signal initial frequency, M is 1, 2.., M denotes the mth pulse currently transmitted by the radar; j is an imaginary unit.
The method for sending the millimeter wave electromagnetic signal comprises the following steps: designing relevant parameters of radar waveforms according to scene requirements, configuring parameters of a radar system by using an upper computer of radar equipment, driving the radar to transmit electromagnetic waves, generating baseband signals by the radar system through a digital-to-analog converter, generating high-frequency signals to be transmitted through a mixer, and finally radiating the signals to a space to be observed through a power amplifier and an antenna.
Step two: receiving scattered echo signals of unmanned aerial vehicle
Unmanned aerial vehicle scattering echo signal is the backscattering echo that is received by the transmission signal after unmanned aerial vehicle reflection, and the echo mainly includes unmanned aerial vehicle main part scattering signal and rotor scattering signal. The situation of receiving the scattering echo signal of the unmanned aerial vehicle is shown in fig. 2;
the rotor unmanned aerial vehicle targets more than one scattering point, and the actual echo is the superposition of the main body of the unmanned aerial vehicle and echo signals of the scattering points of each rotor blade. If unmanned aerial vehicle has N rotors, every rotor has K blades, then to single receiving channel, echo signal S of targetR(t) can be expressed as:
wherein sigmanIs the electromagnetic scattering intensity, t, of the n-th scattering centern0Delaying the transmission of the bidirectional signal from the nth scattering center to the radar;
the method for receiving the scattered echo signals of the unmanned aerial vehicle comprises the following steps: in the first step, electromagnetic waves transmitted by the radar are reflected back to the radar equipment by the unmanned aerial vehicle, out-of-band noise of high-frequency echoes is filtered through a filter, then intermediate-frequency signals and baseband signals are sequentially obtained through two stages of frequency mixers, and finally digital sampling of the baseband signals is completed through an analog-to-digital converter, so that receiving of scattered echo signals of the unmanned aerial vehicle is completed.
Step three: target range-doppler analysis of each frame of data
The LFMCW radar echo is packaged according to a frame format, and the data dimension is generally fast time, slow time and a frame number. The fast time indicates a number of sampling points in a single pulse arranged at sampling intervals, the slow time indicates a number of pulses in a frame arranged at a pulse repetition period, and the frame indicates how many groups such a pulse sequence is transmitted in total. The radar echo contains the radar radial distance and radial velocity information of each target in the observation space, and the distance and velocity information of each target can be extracted through target distance-Doppler analysis.
In the LFMCW signal system radar, the fast time frequency and the slow time frequency of an echo baseband signal are respectively in direct proportion to the radial distance and the speed of a target. Therefore, Fourier transformation of two dimensions is carried out on the echo signals to obtain the frequencies of the two dimensions of the echo signals, and then frequency information is converted into distance and speed information according to radar parameters.
The method for performing the range-Doppler analysis on the data comprises the following steps: arranging the echo signals received in the step two into a three-dimensional echo matrix according to the fast time-slow time-frame arrangement, then performing Fourier transform on the signals along the fast time dimension to obtain a range profile matrix, then performing Fourier transform on the range profile matrix along the slow time dimension to obtain an RD matrix, and obtaining an RD image of one frame of signals after taking a module value. And finally, performing the same processing on each frame of signal to obtain an RD image sequence in one radar measurement.
Step four: determining a range of a distance distribution of a drone
In order to determine the distance area where the rotor wing of the unmanned aerial vehicle is located, the whole distance distribution range of the unmanned aerial vehicle must be judged firstly. CFAR detection is a common method for detecting targets based on radar RD maps. And when the target detection is carried out based on the radar RD image obtained in the step three, firstly, a one-dimensional range profile with a larger signal-to-noise ratio is obtained along the Doppler dimension in an overlapping mode, and then one-dimensional CFAR detection is carried out. For a certain distance unit to be detected, a plurality of units around the unit are taken as protection units, and then a plurality of units are selected as reference units outside the units for estimating the noise intensity. And comparing the value of the unit to be detected with the noise intensity, calculating a threshold value according to the false alarm rate required by the system, and if the result is greater than the threshold value, determining that the unit is an effective scattering point.
The method for determining the distance distribution range of the unmanned aerial vehicle comprises the following steps: after obtaining the RD image of the unmanned aerial vehicle, since the step mainly concerns the range of the distance distribution of the unmanned aerial vehicle, but not the range of the velocity distribution thereof, the RD image is superimposed along the doppler dimension to obtain a one-dimensional range profile with higher signal-to-noise ratio. And then the one-dimensional distance image is subjected to SO-CFAR with better edge detection performance. For each point to be detected in the one-dimensional distance image, P points on two sides of the point are taken as protection units, and Q points are taken as reference units. And respectively estimating a noise value by Q/2 reference units on the left side and Q/2 reference units on the right side of the point to be detected, and then selecting the smaller value of the two as a noise estimation value. For the unmanned aerial vehicle target, the SO-CFAR can detect a plurality of peak points, and the maximum distance and the minimum distance corresponding to the peak points are selected to define the integral distance distribution range [ r ] of the unmanned aerial vehiclemin,rmax]Preparing for later detection of the rotor position of the unmanned aerial vehicle;
step five: calculating a distance-entropy curve
In the radar RD diagram, a uniform target appears as a strong scattering point with a very concentrated energy distribution. The target with the micro-motion phenomenon has the micro-Doppler effect, a Doppler sideband with the bandwidth positively correlated with the speed can be generated in the RD diagram, and the unmanned aerial vehicle main body and the rotor wing can be distinguished by detecting the phenomenon. When using the radar that range resolution is high enough to test unmanned aerial vehicle, unmanned aerial vehicle chance distributes in a plurality of range unit, if only the unmanned aerial vehicle main part then can not produce the Doppler sideband in certain range unit, if contain the rotor in certain range unit, then can produce the Doppler sideband. The entropy is a statistic which can describe the dispersion degree of energy distribution, and for a distance unit with a rotor wing, a Doppler sideband exists, so that the energy distribution is the widest, the dispersion degree is the greatest, and the entropy value is also larger; for the distance unit that only has the unmanned aerial vehicle main part, do not have the Doppler sideband, therefore energy distribution range is less than the former, and the dispersion degree is weak, and entropy value is also less. The distance units with rotors and those without rotors can be effectively distinguished by this statistic of entropy.
The method for calculating the target distance-entropy curve comprises the following steps:
1) for a frame of RD image, first record the maximum value A of the frame amplitudemaxAnd Amin。
2) In [ A ]min,Amax]8-bit uniform quantization is carried out on the whole RD diagram in the range, and the mathematical expression is as follows:
where a is the amplitude of a point and i represents the quantization result.
3) Sequentially calculating the entropy of the quantized RD image according to the distance units, wherein the calculation formula is as follows:
wherein r iskDenotes the kth distance unit, njRepresenting the number of points falling within the jth quantization interval, NrangeRepresenting the total number of points for that range bin.
The advantage of applying the distance-entropy curve is that the entropy can describe the distribution characteristics of energy in each distance unit, the energy distribution range in the distance unit with only noise is small, and the entropy value is minimum; only the unmanned aerial vehicle main body has a large energy distribution range of the distance unit, and the entropy value is larger than that of the noise unit; the distance unit energy distribution range of the rotor wing is the largest, and the entropy value is the largest. Therefore, the difference between the main body and the rotor of the unmanned aerial vehicle can be effectively described by applying the distance-entropy curve, and preparation is made for subsequent rotor area detection.
Step six: detecting rotor area
And step five, estimating the energy distribution dispersion degree in a certain distance unit by adopting entropy, calculating a distance-entropy curve corresponding to one frame of RD graph, and detecting a plurality of peak values by applying OS-CFAR on the distance-entropy curve, wherein the corresponding distance unit is the distance unit with the inching target. OS-CFAR is a peak detection method with high calculation speed and strong robustness, which selects a plurality of units on two sides of a unit to be detected as reference units, sorts the amplitudes of all the reference units and selects the nth unitkAnd the amplitude of each reference unit is used as a noise estimation value, then the amplitude of the unit to be detected is compared with the noise value, and if the amplitude is larger than a set threshold value, the unit to be detected is regarded as a peak value. Wherein n iskAnd the threshold needs to be set according to the false alarm probability required by the scene. Considering that other targets with micromotion, such as birds, may exist in the scene, the unmanned aerial vehicle distance distribution range obtained in step four needs to be further used for eliminating false targets. And regarding the peak points detected by the OS-CFAR, if the peak points are distributed in the distance range of the unmanned aerial vehicle, the peak points are considered to be the area where the rotor wing is located.
The method for detecting the rotor wing area comprises the following steps: and performing OS-CFAR detection on the distance-entropy curve obtained by calculation in the fifth step to obtain a plurality of peak points. And D, combining the whole distance distribution range of the unmanned aerial vehicle obtained in the step four, and if the detected peak point is in the distance distribution range of the unmanned aerial vehicle, judging that the detected peak point is the region where the rotor wing is located.
Claims (8)
1. An unmanned aerial vehicle rotor wing detection method based on a millimeter wave radar is characterized by comprising the following steps:
the method comprises the following steps: transmitting millimeter wave electromagnetic signals
Installing millimeter wave radar equipment outdoors, directing the wave beam of the radar to the airspace where the unmanned aerial vehicle to be monitored is located, and then driving the millimeter wave radar to transmit millimeter wave electromagnetic signals; the millimeter wave radar adopts a linear frequency modulation continuous wave LFMCW signal system, baseband signals are converted into radio frequency analog signals after DA conversion, frequency mixing and power amplification, and the radio frequency analog signals are transmitted to an area to be observed by a transmitting antenna;
step two: receiving scattered echo signals of unmanned aerial vehicle
The unmanned aerial vehicle scattering echo signal is a backscattering echo received after a transmitting signal is reflected by the unmanned aerial vehicle, and the echo comprises an unmanned aerial vehicle main body scattering signal and a rotor wing scattering signal; in the first step, electromagnetic waves transmitted by a radar are reflected back to radar equipment by an unmanned aerial vehicle, out-of-band noise is filtered through a filter, then intermediate-frequency signals and baseband signals are sequentially obtained through two stages of mixers, and finally the intermediate-frequency signals and the baseband signals are converted into digital signals through AD sampling, namely echo signals of the unmanned aerial vehicle;
step three: target range-doppler analysis of each frame of data
The echo of the unmanned aerial vehicle in the step two is packaged according to the format of a frame, and the data dimensionality is fast time, slow time and a frame sequence number; the echo comprises radar radial distance and radial velocity information of each target in an observation space, and the distance and velocity information of each target can be extracted through target distance-Doppler analysis; fourier transformation is carried out on each frame of signal along the fast time dimension and the slow time dimension respectively to obtain a radar RD image, and preparation is made for subsequent unmanned aerial vehicle distance range detection and rotor wing area detection;
step four: determining a range of a distance distribution of a drone
Superposing the radar RD images obtained in the third step along the Doppler dimension to obtain a one-dimensional range profile with a larger signal-to-noise ratio, and then performing one-dimensional SO-CFAR detection, wherein the geometric dimension of the unmanned aerial vehicle is larger than the radar range resolution, SO that a plurality of peak points can be detected in the area where the unmanned aerial vehicle is located; finding a point farthest from the radar and a point closest to the radar in the peak points to define two end points of unmanned aerial vehicle distance distribution, namely the distance distribution range of the unmanned aerial vehicle, so as to prepare for subsequent rotor wing area detection;
step five: calculating a distance-entropy curve
8-bit uniform quantization is carried out on the radar RD image obtained in the third step, then the amplitude distribution entropy of each distance unit in the quantized RD image is calculated in sequence, and a distance-entropy curve is obtained to prepare for the subsequent detection of the rotor wing area of the unmanned aerial vehicle;
step six: detecting rotor area
Comprehensively considering the distance-entropy curve calculated in the fifth step and the unmanned aerial vehicle distance distribution range detected in the fourth step; and detecting a plurality of peak values by applying the OS-CFAR on the distance-entropy curve, and if the peak value points appear in the unmanned aerial vehicle distance distribution range defined in the fourth step, determining that the peak value points are in the region where the rotor wing is located, otherwise, determining that the peak value points are in the region where only the unmanned aerial vehicle main body is located.
2. The unmanned aerial vehicle rotor detection method based on the millimeter wave radar as claimed in claim 1, wherein: the first step is specifically as follows: let the slope of the radar sweep frequency be k, the period be T, and the initial phase be
For any time t, the LFMCW radar transmits a signal sT(t) is expressed as:
where A is the signal amplitude, f0For the signal initial frequency, M is 1, 2.., M denotes the mth pulse currently transmitted by the radar; j is an imaginary unit.
3. The unmanned aerial vehicle rotor detection method based on the millimeter wave radar as claimed in claim 1, wherein: the second step is specifically as follows: the rotor unmanned aerial vehicle targets more than one scattering point, and the actual echo is superposition of echo signals of the scattering points of the main body of the unmanned aerial vehicle and the rotor blades; provided with an unmanned aerial vehicle with N rotors, each rotorEach rotor has K blades, so that the echo signal S of the target is transmitted to a single receiving channelR(t) is expressed as:
wherein sigmanIs the electromagnetic scattering intensity, t, of the n-th scattering centern0And delaying the transmission of the bi-directional signal from the nth scattering center to the radar.
4. The unmanned aerial vehicle rotor detection method based on the millimeter wave radar as claimed in claim 1, wherein: the third step is specifically as follows: LFMCW radar echoes are packaged according to a frame format, and data dimensions are fast time, slow time and a frame number; the fast time represents a plurality of sampling points arranged according to sampling intervals in a single pulse, the slow time represents a plurality of pulses arranged according to a pulse repetition period in a frame, and the frame represents how many groups of such pulse sequences are transmitted together; the radar echo comprises radar radial distance and radial velocity information of each target in an observation space, and the distance and velocity information of each target can be extracted through target distance-Doppler analysis;
in the LFMCW signal system radar, the fast time frequency and the slow time frequency of an echo baseband signal are respectively in direct proportion to the radial distance and the speed of a target; therefore, Fourier transformation of two dimensions is carried out on the echo signals to obtain the frequencies of the two dimensions of the echo signals, and then frequency information is converted into distance and speed information according to radar parameters.
5. The unmanned aerial vehicle rotor detection method based on the millimeter wave radar as claimed in claim 1, wherein: the fourth step is specifically as follows: in order to determine the distance area where the rotor wing of the unmanned aerial vehicle is located, the overall distance distribution range of the unmanned aerial vehicle must be judged firstly; when target detection is carried out on the basis of the radar RD image obtained in the step three, firstly, a one-dimensional range profile with a larger signal-to-noise ratio is obtained through superposition along the Doppler dimension, and then one-dimensional CFAR detection is carried out; for a certain distance unit to be detected, firstly, taking a plurality of units around the unit as protection units, and then selecting a plurality of units outside the protection units as reference units for estimating the noise intensity; comparing the value of the unit to be detected with the noise intensity, calculating a threshold value according to the false alarm rate required by the system, and if the result is greater than the threshold value, determining that the unit is an effective scattering point;
after obtaining the RD image of the unmanned aerial vehicle, because the step cares about the distance distribution range of the unmanned aerial vehicle and does not care about the speed distribution range of the unmanned aerial vehicle, the RD image is superposed along the Doppler dimension to obtain a one-dimensional range image with higher signal-to-noise ratio; then, SO-CFAR with good edge detection performance is carried out on the one-dimensional distance image; for each point to be detected in the one-dimensional distance image, taking P points on two sides of the point as a protection unit, and taking Q points as a reference unit; respectively estimating a noise value by Q/2 reference units on the left side and Q/2 reference units on the right side of the point to be detected, and then selecting the smaller value of the two as a noise estimation value; for the unmanned aerial vehicle target, the SO-CFAR can detect a plurality of peak points, and the maximum distance and the minimum distance corresponding to the peak points are selected to define the integral distance distribution range [ r ] of the unmanned aerial vehiclemin,rmax]And preparing for later detection of the rotor position of the unmanned aerial vehicle.
6. The unmanned aerial vehicle rotor detection method based on the millimeter wave radar as claimed in claim 1, wherein: the step five specifically comprises the following steps: in a radar RD image, a uniform-speed target is represented as a strong scattering point with very concentrated energy distribution; the target with the micro-motion phenomenon has the micro-Doppler effect, a Doppler sideband with the bandwidth positively correlated with the speed is generated in the RD image, and the main body and the rotor of the unmanned aerial vehicle can be distinguished by detecting the phenomenon; when the radar with high enough distance resolution is used for testing the unmanned aerial vehicle, the unmanned aerial vehicle can be distributed in a plurality of distance units, if only the unmanned aerial vehicle body is arranged in a certain distance unit, a Doppler sideband can not be generated, and if a rotor wing is arranged in a certain distance unit, a Doppler sideband can be generated; the distance unit with the rotor wing and the distance unit without the rotor wing can be effectively distinguished through the statistic of entropy;
the method for calculating the target distance-entropy curve comprises the following steps:
1) for a frame of RD image, first record the maximum value A of the frame amplitudemaxAnd Amin；
2) In [ A ]min,Amax]8-bit uniform quantization is carried out on the whole RD diagram in the range, and the mathematical expression is as follows:
where a is the amplitude of a point and i represents the quantization result;
3) sequentially calculating the entropy of the quantized RD image according to the distance units, wherein the calculation formula is as follows:
wherein r iskDenotes the kth distance unit, njRepresenting the number of points falling within the jth quantization interval, NrangeRepresenting the total number of points for that range bin.
7. The unmanned aerial vehicle rotor detection method based on the millimeter wave radar as claimed in claim 1, wherein: the sixth step is specifically as follows: estimating the energy distribution dispersion degree in a certain distance unit by adopting entropy, calculating a distance-entropy curve corresponding to a frame of RD graph, and detecting a plurality of peak values by applying OS-CFAR on the distance-entropy curve, wherein the corresponding distance unit is the distance unit with the micro-motion target; the OS-CFAR selects a plurality of units on two sides of the unit to be detected as reference units, sorts the amplitudes of all the reference units and selects the nth unitkThe amplitude of each reference unit is used as a noise estimation value, then the amplitude of the unit to be detected is compared with the noise value, and if the amplitude is larger than a set threshold value, the unit to be detected is regarded as a peak value; wherein n iskAnd the threshold value needs to be set according to the false alarm probability required by the scene; considering the target with possible micro motion in the scene, the range of the unmanned plane distance distribution obtained in step four needs to be further used for eliminating the virtual targetA false target; and regarding the peak points detected by the OS-CFAR, if the peak points are distributed in the distance range of the unmanned aerial vehicle, the peak points are considered to be the area where the rotor wing is located.
8. An unmanned aerial vehicle rotor detection system based on a millimeter wave radar is characterized by comprising a millimeter wave signal receiving and transmitting unit, an RD processing unit, an unmanned aerial vehicle distance range detection unit and a rotor area detection unit;
the millimeter wave signal receiving and transmitting unit is connected with the RD processing unit, is a front-end unit for RD processing and is used for completing the radio frequency processing function of the front end; the unit controls the relevant parameters of LFMCW wave beams of linear frequency modulation continuous waves through a PC upper computer, radar baseband signals to be transmitted are generated by a high-speed DA, high-frequency millimeter wave signals capable of being transmitted are generated by a mixer, then the signals are transmitted to an antenna through a power amplifier, and electromagnetic waves are radiated to a target space by the antenna; the transmitted millimeter wave signal is reflected by a target, received by an antenna, converted into a baseband signal through a filter and a mixer, and transmitted to a PC (personal computer) end for storage after AD (analog-to-digital) sampling, so as to prepare for subsequent distance-Doppler processing;
the distance-Doppler RD processing unit is mutually connected with the millimeter wave receiving and transmitting unit, the unmanned aerial vehicle distance range detection unit and the rotor wing area detection unit, inputs radar echo signals received by the millimeter wave receiving unit and is AD sampling data arranged according to a time sequence; the unit rearranges data sequence according to fast time sampling point number, Chirp number and frame number defined by a radar system, wherein each dimension sequence of the data is fast time-slow time-frame number; then, Fourier transformation is carried out on the data respectively along the fast time and the slow time, extraction of RD information in radar echo signals is completed, and an RD image of each frame of data is generated; the generated RD image is used as the input of a subsequent unmanned plane distance range detection unit and a rotor wing area detection unit;
the unmanned aerial vehicle distance range detection unit is connected with the RD processing unit and the rotor wing area detection unit and is used for detecting the whole distance distribution range of the unmanned aerial vehicle; the unit superposes the radar RD images along Doppler dimensions to obtain a one-dimensional range profile with a higher signal-to-noise ratio, and two end points of a range distribution range of the unmanned aerial vehicle are obtained by adopting minimum selection constant false alarm detection SO-CFAR detection to prepare for subsequent rotor wing area detection;
the rotor wing area detection unit is connected with the RD processing unit and the unmanned aerial vehicle distance range detection unit and is used for finishing accurate extraction of the area where the rotor wing of the unmanned aerial vehicle is located; the unit calculates a distance-entropy curve based on a radar RD diagram, performs order statistics on the distance-entropy curve, detects the distance unit with the micro-motion target through a constant false alarm rate (OS) -CFAR (computational fluid dynamics) detection, and finally determines the distance unit with the rotor wing by combining the distance range of the unmanned aerial vehicle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110778809.7A CN113567976B (en) | 2021-07-09 | 2021-07-09 | Unmanned aerial vehicle rotor wing detection system based on millimeter wave radar and detection method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110778809.7A CN113567976B (en) | 2021-07-09 | 2021-07-09 | Unmanned aerial vehicle rotor wing detection system based on millimeter wave radar and detection method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113567976A true CN113567976A (en) | 2021-10-29 |
| CN113567976B CN113567976B (en) | 2023-08-29 |
Family
ID=78164294
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110778809.7A Active CN113567976B (en) | 2021-07-09 | 2021-07-09 | Unmanned aerial vehicle rotor wing detection system based on millimeter wave radar and detection method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113567976B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4275396A (en) * | 1979-10-12 | 1981-06-23 | Jacomini Omar J | Helicopter rotating blade detection system |
| CN108957443A (en) * | 2018-07-16 | 2018-12-07 | 北京航空航天大学 | A kind of estimation method based on double the unmanned plane rotor length for sending out double receipts coherent radars and revolving speed |
| CN110146853A (en) * | 2019-06-03 | 2019-08-20 | 浙江大学 | A kind of aircraft rotor fine motion feature extracting method |
| KR20190122288A (en) * | 2018-04-20 | 2019-10-30 | 한국항공대학교산학협력단 | Method of detecting rotorcraft and rotorcraft detection system |
| CN111198359A (en) * | 2020-01-11 | 2020-05-26 | 桂林电子科技大学 | Rotor unmanned aerial vehicle detection method based on target Doppler width |
-
2021
- 2021-07-09 CN CN202110778809.7A patent/CN113567976B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4275396A (en) * | 1979-10-12 | 1981-06-23 | Jacomini Omar J | Helicopter rotating blade detection system |
| KR20190122288A (en) * | 2018-04-20 | 2019-10-30 | 한국항공대학교산학협력단 | Method of detecting rotorcraft and rotorcraft detection system |
| CN108957443A (en) * | 2018-07-16 | 2018-12-07 | 北京航空航天大学 | A kind of estimation method based on double the unmanned plane rotor length for sending out double receipts coherent radars and revolving speed |
| CN110146853A (en) * | 2019-06-03 | 2019-08-20 | 浙江大学 | A kind of aircraft rotor fine motion feature extracting method |
| CN111198359A (en) * | 2020-01-11 | 2020-05-26 | 桂林电子科技大学 | Rotor unmanned aerial vehicle detection method based on target Doppler width |
Non-Patent Citations (3)
| Title |
|---|
| 余启;饶彬;罗鹏飞;: "杂波背景下基于LFMCW雷达的无人机检测性能分析", 无线电工程, no. 12 * |
| 宋晨;周良将;吴一戎;丁赤飚;: "基于时频集中度指标的多旋翼无人机微动特征参数估计方法", 电子与信息学报, no. 08 * |
| 贺治华;王德奇;: "毫米波雷达对直升机旋翼回波检测研究", 电子学报, no. 06 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113567976B (en) | 2023-08-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111352102B (en) | Multi-target number detection method and device based on frequency modulation continuous wave radar | |
| Drozdowicz et al. | 35 GHz FMCW drone detection system | |
| CN103176178B (en) | Radar moving target radon-fractional Fourier transform long-time phase-coherent accumulation detection method | |
| US20160069994A1 (en) | Sense-and-avoid systems and methods for unmanned aerial vehicles | |
| Aldowesh et al. | Slow-moving micro-UAV detection with a small scale digital array radar | |
| CN111896926B (en) | Low-altitude target detection method and system based on strong clutter suppression | |
| Kwag et al. | Multi-mode SDR radar platform for small air-vehicle Drone detection | |
| Sun et al. | A fast approach for detection and parameter estimation of maneuvering target with complex motions in coherent radar system | |
| Capria et al. | Ship detection with DVB-T software defined passive radar | |
| CN109085568B (en) | Frequency modulation continuous wave multi-target detection method based on secondary frequency mixing | |
| CN108776332B (en) | Method for detecting marine maneuvering target by using high-frequency ground wave radar | |
| KR102053881B1 (en) | Ground-based Array Antenna System and Method for Obtaining an Image of Detection Region in the Sky using the Ground-based Array Antenna System | |
| CN110488239B (en) | Target detection method based on frequency modulation continuous wave radar | |
| Anderson | Target classification, recognition and identification with HF radar | |
| CN113567976B (en) | Unmanned aerial vehicle rotor wing detection system based on millimeter wave radar and detection method thereof | |
| KR102103636B1 (en) | Method of detecting rotorcraft and rotorcraft detection system | |
| Alivizatos et al. | Towards a range-doppler UHF multistatic radar for the detection of non-cooperative targets with low RCS | |
| Strømøy | Hitchhiking bistatic radar | |
| EP4396602A1 (en) | Radar system and associated apparatus and methods | |
| CN114624694A (en) | Radar signal processing method based on unequal interval modulation DDMA-MIMO | |
| Latthe et al. | A review on ISAR imaging techniques for low RCS targets | |
| Li et al. | A Feature-based Filtering Algorithm with 60GHz MIMO FMCW Radar for Indoor Detection and Trajectory Tracking | |
| Bi et al. | Millimeter wave radar technology | |
| CN114895262B (en) | Clutter suppression method based on space-time spectrum characteristics and orthogonal projection of wind motor | |
| CN1603858A (en) | Optimal weighted value estimation method for optimum processing in airborne radar target detection |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |