An Insect Weak Wingbeat Frequency Radar Measurement Method Based on Micro-Doppler Parameter Space Search

Abstract

The present invention relates to the technical field of radar measurement, and particularly relates to a radar measurement method for the weak flapping frequency of insects based on micro-Doppler parameter space search. A method for measuring the weak flapping frequency of insects based on micro-Doppler parameter space search according to the present invention provides an effective measurement means for measuring the flapping frequency of insects. Compared with the existing methods for measuring the flapping frequency of insects, this method can greatly improve the measurement accuracy and measurement success rate of the flapping frequency of insect echoes under low signal-to-noise ratio conditions, and is used to solve the problem that the flapping frequency of insects cannot be accurately measured under low signal-to-noise ratio.

Description

Technical Field

[0001] The present invention relates to the technical field of radar measurement, and particularly to a method for measuring the weak wing-beating frequency of insects based on micro-Doppler parameter space search by radar. Background Art

[0002] China is the most important insect migration field in the world. Major migratory pests of crops such as rice, wheat, and corn migrate long distances in China, which is an important reason for the outbreak of insect pests in different regions and seriously threatens the national food production safety. The core of accurate early warning of migratory insect pests is to obtain accurate measurements of insect migratory behavior. Radar can detect targets by transmitting and receiving electromagnetic waves, and can achieve large-airspace and long-distance detection of migratory behavior without affecting insect migration, which is the most effective means to monitor insect migratory behavior.

[0003] The wing-beating behavior is a necessary condition for insects to achieve migratory behavior. Its wing-beating frequency reflects the speed of the wings flapping up and down during insect flight, which is one of the important ethological parameters to measure the flight ability of insects and can be used to identify different insect species. Traditional insect radar measurement of wing-beating frequency is mainly based on the amplitude or phase fluctuation of the echo caused by insect wing-beating. The success rate and accuracy of the wing-beating frequency measurement method based on echo amplitude fluctuation are easily affected by the echo signal-to-noise ratio and wing-beating amplitude, and are only applicable to the measurement of wing-beating frequency of medium and large insects; the wing-beating frequency measurement method based on phase fluctuation uses echo phase information and greatly improves the measurement accuracy of wing-beating frequency, but still faces the problems of low measurement success rate and accuracy under low signal-to-noise ratio. Summary of the Invention

[0004] In view of this, the present invention provides a method for measuring the weak wing-beating frequency of insects by radar through micro-Doppler parameter space search. This method first uses time-frequency analysis to transform the insect echo collected by the radar into the micro-Doppler parameter space to obtain a micro-Doppler spectrogram; furthermore, it uses the method of Generalized Radon Fourier Transform (GRFT) parameter space search to transform the micro-Doppler spectrogram into the wing-beating parameter (wing-beating frequency / wing-beating amplitude) plane, and realizes long-time coherent accumulation of weak insect wing-beating echoes during this process; finally, by detecting the peak value in the wing-beating parameter plane, the extraction of the insect wing-beating frequency is ultimately realized.

[0005] The technical solution of the present invention is as follows:

[0006] A method for measuring the weak wing-beating frequency of insects based on micro-Doppler parameter space search by radar, comprising the following steps:

[0007] Step 1: Perform time-frequency analysis on the insect radar echo to obtain the micro-Doppler spectrogram of the insect;

[0008] Step 2: Perform a GRFT parameter space search based on the micro-Doppler spectrogram of the insect obtained in Step 1 to obtain the wing flapping parameter plane of the insect;

[0009] Step 3: Perform peak detection on the wing flapping parameter plane obtained in Step 2 to obtain the wing flapping amplitude and wing flapping frequency of the insect, and complete the radar measurement of the weak wing flapping frequency of the insect based on the micro-Doppler parameter space search.

[0010] In the said Step 1, the insect radar echo is:

[0011]

[0012] where k is a constant related to the radar parameters, σ0 is the radar cross section (Radar Cross Section) when the insect is not flapping its wings, η is the ratio of the RCS fluctuation caused by the insect's wing flapping to the RCS when not flapping its wings, t is the echo time, f w is the insect wing flapping frequency, and A w is the insect wing flapping amplitude;

[0013] The obtained micro-Doppler spectrogram S(τ, f) of the insect is:

[0014]

[0015] where τ is the micro-Doppler spectrogram time, f is the micro-Doppler spectrogram frequency, and T w is the time-frequency analysis time window;

[0016] When performing time-frequency analysis, the short-time Fourier transform method is used to transform the insect echo collected by the radar into the time-Doppler domain;

[0017] In the said Step 2, the method for obtaining the wing flapping parameter plane of the insect is:

[0018] Let the range of the insect wing flapping amplitude be A w ∈(0, A max , and the range of the insect wing flapping frequency be f w ∈(0, f max . According to the insect radar echo, when the searched wing flapping amplitude is A' w and the wing flapping frequency is f' w , the instantaneous Doppler frequency f d of the insect is:

[0019]

[0020] In the micro-Doppler spectrogram S(τ, f), along the instantaneous Doppler frequency f corresponding to the searched parameters A' w and f' w d ​(τ), the instantaneous time-frequency value of the insect obtained by extraction is:

[0021]

[0022] Based on the Radon transform, all the searched wingbeat amplitudes A' are traversed w and the wingbeat frequencies f' w , and the micro-Doppler spectrogram S(τ, f) is transformed into the wingbeat parameter plane R Af (A' w , f' w ), and the phase of the time-frequency value is compensated during the transformation, and the result is:

[0023]

[0024] That is, the ranges of the insect wingbeat amplitude and the wingbeat frequency are set. According to the expression of the insect micro-Doppler spectrogram and the corresponding relationship between the instantaneous time-frequency value and the wingbeat amplitude and the wingbeat frequency, the GRFT transformation results at the corresponding wingbeat amplitude and frequency are calculated, and by traversing all the searched parameters, the micro-Doppler spectrogram is transformed into the wingbeat parameter plane;

[0025] In the third step described above, peak detection on the wingbeat parameter plane means extracting the peak coordinates (A w , f w ) of the wingbeat parameter plane.

[0026] Beneficial effects

[0027] (1) The method for measuring the weak wingbeat frequency of insects based on micro-Doppler parameter space search of the present invention uses the method of time-frequency analysis, comprehensively utilizes the amplitude and phase information in the insect radar echo, obtains the micro-Doppler spectrogram of the insects, and realizes the conversion of insect behavior observation from a single dimension such as the time domain or the frequency domain to a multi-dimensional time-frequency domain;

[0028] (2) The method for measuring the weak wingbeat frequency of insects based on micro-Doppler parameter space search of the present invention uses the method of GRFT parameter space search for the micro-Doppler spectrogram, realizes the long-time accumulation of the instantaneous time-frequency value of the insects, greatly improves the detection signal-to-noise ratio of the wingbeat frequency, and thus improves the measurement accuracy and success rate of the wingbeat frequency;

[0029] (3) The present invention discloses a method for measuring the weak wingbeat frequency of insects based on micro-Doppler parameter space search; the invention gives the steps for realizing the accurate measurement of the insect wingbeat frequency, obtains the micro-Doppler spectrogram by performing time-frequency analysis on the insect radar echo; then performs GRFT parameter space search on the micro-Doppler spectrogram and transforms it into the wingbeat parameter space; finally, extracts the wingbeat frequency by detecting the peak of the wingbeat parameter space, thereby realizing the accurate measurement of the insect wingbeat frequency;

[0030] (4) Combining with the measured data, the present invention gives the actual effect of measuring the insect's weak wingbeat frequency by the radar measurement method based on the micro-Doppler parameter space search and the comparison with the traditional method, which verifies that using this method can greatly improve the success rate and accuracy of measuring the insect's wingbeat frequency;

[0031] (5) A method for measuring the weak wingbeat frequency of insects based on the micro-Doppler parameter space search of the present invention provides an effective measurement means for measuring the insect's wingbeat frequency. Compared with the existing methods for measuring the insect's wingbeat frequency, this method can greatly improve the measurement accuracy and success rate of the wingbeat frequency of the insect echo under the condition of low signal-to-noise ratio, and is used to solve the problem that the insect's wingbeat frequency cannot be accurately measured under low signal-to-noise ratio. Brief Description of the Drawings

[0032] Figure 1 is the flow chart for implementing the radar measurement method for the weak wingbeat frequency of insects based on the micro-Doppler parameter space search of the present invention;

[0033] Figure 2 is the micro-Doppler spectrogram of the insect echo with a simulated wingbeat frequency of 25 Hz and a wingbeat amplitude of 5 mm;

[0034] Figure 3 is Figure 2 the parameter space search result and the peak position of the micro-Doppler spectrogram shown;

[0035] Figure 4 is the graph of the variation of the success rate of wingbeat frequency measurement with the echo signal-to-noise ratio;

[0036] Figure 5 is the graph of the variation of the measurement accuracy of wingbeat frequency with the echo signal-to-noise ratio. Detailed Embodiment

[0037] The present invention will be described in detail below with reference to the drawings and by way of examples.

[0038] Example

[0039] The present invention provides a method for measuring the weak wingbeat frequency of insects based on the micro-Doppler parameter space search, as Figure 1 shown, including: time-frequency analysis of the insect radar echo, GRFT parameter space search of the micro-Doppler spectrogram, and peak detection in the wingbeat parameter plane, which are three steps.

[0040] First, perform time-frequency analysis on the migratory insect radar echo collected by the radar to obtain the instantaneous Doppler spectrogram of the target, as Figure 2 shown, where the dotted line represents the instantaneous Doppler frequency of the insect. When performing time-frequency analysis, the radar measures the wingbeat frequency as f w and the wingbeat amplitude as Aw The insect radar echo s(t) is

[0041]

[0042] where k is a constant related to radar parameters, σ0 is the radar cross section of the insect when its wings are not flapping, η is the ratio of the RCS fluctuation caused by the insect's wing flapping to the RCS when the wings are not flapping, t is time, and f w is the insect wing flapping frequency, and A w is the insect wing flapping amplitude. Through time-frequency analysis, the micro-Doppler spectrogram of the insect radar echo s(t) is obtained as:

[0043]

[0044] where τ and f are time and Doppler frequency respectively.

[0045] Secondly, using the method of GRFT parameter space search, the micro-Doppler spectrogram shown in formula (1.2) is transformed to the wing flapping parameter plane, as Figure 3 shown. In this process, the ranges of the insect wing flapping amplitude and wing flapping frequency are set as A w ∈(0, A max and f w ∈(0, f max . According to the insect radar echo model shown in formula (1.1), when the searched wing flapping amplitude is A' w and the wing flapping frequency is f' w , the instantaneous Doppler frequency f d of the insect is:

[0046]

[0047] In the micro-Doppler spectrogram S(τ, f), along the instantaneous Doppler frequency f w corresponding to the searched parameters A' w and f' d (τ), the instantaneous time-frequency value of the insect target can be extracted as:

[0048]

[0049] Based on the idea of Radon transform, all the searched wing flapping amplitudes A' w and wing flapping frequencies f' w are traversed, and the micro-Doppler spectrogram S(τ, f) is transformed to the wing flapping parameter plane R Af (A' w , f' w ), and the phase of the time-frequency value shown in formula (1.4) is compensated during the transformation process, and the result is:

[0050]

[0051] Finally, according to the measured results of the target flapping parameters, the position of the maximum amplitude in the flapping parameter plane is obtained. Therefore, the flapping parameter plane R is extracted. Af (A' w , f' w ) at the point of maximum amplitude is obtained, so as to measure the flapping frequency information contained in the insect radar echo.

[0052] From the above implementation steps, it can be seen that in the process of searching the micro-Doppler parameter space, this method compensates the phase term in the instantaneous time-frequency value and accumulates it, realizing the long-time accumulation of the insect radar echo, effectively improving the signal-to-noise ratio of the insect echo, and thus improving the measurement accuracy of the flapping frequency.

[0053] Therefore, the present invention provides a method for measuring the weak flapping frequency of insects based on the search of the micro-Doppler parameter space. The implementation steps will be described below with specific embodiments:

[0054] To verify the measurement accuracy and correctness of the above-mentioned flapping frequency extraction method, a darkroom observation experiment of the radar on Agrotis ypsilon is carried out, and a method for measuring the weak flapping frequency of insects based on the search of the micro-Doppler parameter space of the present invention is used for the insect radar echo to complete the measurement of the flapping frequency of the insects. The radar parameters used in the darkroom observation are shown in Table 1;

[0055] Table 1 Radar parameters

[0056]

[0057]

[0058] Since the experiment was carried out in a dark room and the echo signal-to-noise ratio of the insects was relatively high, when analyzing the measurement accuracy of the four wingbeat frequency extraction methods for low signal-to-noise ratio targets, a semi-empirical and semi-simulative method was adopted, that is: noise was superimposed on the echoes collected in the dark room to simulate echo data under different signal-to-noise ratios. Based on the wingbeat frequency extraction method proposed in the present invention, Monte Carlo simulation was performed on the noisy echoes. The signal-to-noise ratio was set to increase from -9 dB to 15 dB at an interval of 0.02 dB, and each signal-to-noise ratio was simulated 3000 times. At the same time, the actual wingbeat frequency of the black cutworm measured by the optical observation equipment was 32 Hz. During the analysis process, the measurement success rate and measurement error of the wingbeat frequency were used as performance evaluation indicators. It was set that when the absolute difference between the measured value and the true value was less than 1 Hz, the measurement was considered successful. The measurement success rate was the ratio of the number of times the absolute difference between the measured value and the true value was less than 1 Hz to the number of simulation times, and the measurement error was the root mean square error between the measured value and the true value when the measurement was successful. Finally, the variation relationship of the wingbeat frequency measurement success rate and measurement error of the method proposed in the present invention with the signal-to-noise ratio and the comparison results with the existing methods based on amplitude fluctuation or phase fluctuation of insect radar are as Figure 4 and Figure 5 shown.

[0059] Based on the wingbeat frequency measurement results of the above dark room experimental data, the following conclusions can be obtained:

[0060] A method for measuring the weak wingbeat frequency of insects by radar based on micro-Doppler parameter space search proposed in the present invention can effectively achieve the accuracy and success rate of measuring the wingbeat frequency of insects.

[0061] In summary, the above are only the preferred embodiments of the present invention and are not used to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An insect weak wingbeat frequency radar measurement method based on micro-Doppler parameter space search, characterized in that Including the following steps: Three steps: time-frequency analysis of insect radar echoes, GRFT parameter space search of micro-Doppler spectrograms, and peak detection in the flapping parameter plane First, perform time-frequency analysis on the radar echoes of migrating insects collected by the radar to obtain the instantaneous Doppler spectrogram of the target. When performing time-frequency analysis, the radar measures the wingbeat frequency as , and the wingbeat amplitude as of the insect radar echo is: Among them, is a constant related to radar parameters, is the radar cross section of the insect when its wings are not flapping, is the ratio of the RCS fluctuation caused by the insect's wing flapping to the RCS when the wings are not flapping, t is time, is the insect wing flapping frequency, is the insect wing flapping amplitude. Through time-frequency analysis, the micro-Doppler spectrogram of the insect radar echo is as follows: wherein, is the time delay, is the Doppler frequency; Secondly, using the GRFT parameter space search method, the micro-Doppler spectrogram is transformed into the wingbeat parameter plane. The ranges of the insect wingbeat amplitude and wingbeat frequency are set to be and respectively. When the searched wingbeat amplitude is and the wingbeat frequency is , the instantaneous Doppler frequency of the insect is: In the micro-Doppler spectrogram along the searched parameters and the corresponding instantaneous Doppler frequency the instantaneous time-frequency value of the insect target extracted is: Traverse all the flapping amplitudes to be searched and the flapping frequencies , transform the micro-Doppler spectrogram to the flapping parameter plane , and compensate the phase of the time-frequency values during the transformation, and the result is: Finally, according to the measurement result of the target flapping parameters, the position of the maximum amplitude in the flapping parameter plane is determined. Therefore, the maximum amplitude point in the flapping parameter plane is extracted, so as to measure the flapping frequency information contained in the insect radar echo.

Images