River echo center frequency estimation method based on spectral peak search

Abstract

The embodiment of the present application discloses a river echo center frequency estimation method based on spectral peak search, comprising: smoothing Doppler spectrum; searching spectral peak, estimating bandwidth and centroid frequency of the first search spectral peak; judging the bandwidth of the first search spectral peak, if the spectral peak bandwidth is greater than twice the Bragg inherent frequency, it is determined that the energy of two Bragg peaks is close, the centroid frequency of the first search spectral peak is taken as the center frequency and the subsequent steps are omitted, otherwise it is determined that there is a large difference in the energy of two Bragg peaks, and step 4 is turned to; performing secondary spectral peak search, estimating the centroid frequency of the secondary search spectral peak, and turning to step 5; estimating the center frequency according to the centroid frequency of the spectral peak searched twice. The present application solves the problem of low center frequency estimation accuracy under the condition that there is a large difference in the energy of two Bragg peaks in the river echo Doppler spectrum, accurately extracts two Bragg peaks in the echo Doppler spectrum by using the threshold truncation method, and then estimates the center frequency, thereby improving the estimation accuracy of the center frequency.

Description

Technical Field

[0001] This invention belongs to the technical field of microwave radar river remote sensing, specifically relating to a method for estimating the center frequency of river echoes based on spectral peak search. Background Technology

[0002] In recent years, microwave Doppler radar has been widely used in river monitoring due to its all-weather, real-time, and high-precision characteristics. Unlike contact-based flow measurement instruments such as flow meters and acoustic Doppler current profilers, microwave Doppler radar can achieve non-contact measurement of river surface velocity and is suitable for operation in medium to high water levels. The working principle of microwave Doppler radar is based on the Doppler effect. By estimating the center frequency of the river echo within the Doppler spectrum of the radar echo, the surface velocity of the river can be obtained. The most crucial aspect of this process is estimating the center frequency of the river echo, as its accuracy directly determines the accuracy of the surface velocity measurement.

[0003] The most widely used method for center frequency estimation is the spectral moment method, which has the advantages of simple algorithm and low computational cost. However, when there is a large difference in energy between the two Bragg peaks in the echo spectrum, the estimation accuracy of the spectral moment method is low. In order to improve the flow measurement accuracy of microwave Doppler radar, it is necessary to propose a method that can still accurately estimate the center frequency when there is a large difference in energy between the two Bragg peaks. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by providing a method for estimating the center frequency of river echoes based on spectral peak search. This method solves the problem of low center frequency estimation accuracy when there is a large difference in the energy of two Bragg peaks.

[0005] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0006] A method for estimating the center frequency of river echoes based on spectral peak search includes the following steps:

[0007] Step 1: Smooth the Doppler spectrum;

[0008] Step 2: Search for spectral peaks and estimate the bandwidth and centroid frequency of the first searched spectral peak;

[0009] Step 3: Determine the bandwidth of the first search peak. If the bandwidth of the peak is greater than twice the Bragg natural frequency, the two Bragg peaks are considered to have similar energies. The centroid frequency of the first search peak is taken as the center frequency and subsequent steps are omitted. Otherwise, the two Bragg peaks are considered to have significant energy differences, and the process proceeds to step 4.

[0010] Step 4: Perform a secondary peak search, estimate the centroid frequency of the secondary search peak, and proceed to step 5;

[0011] Step 5: Estimate the center frequency based on the centroid frequencies of the spectral peaks from the two searches.

[0012] Furthermore, in step 1, the Doppler spectrum is smoothed at N points.

[0013] Furthermore, in step 2, the threshold truncation method is used to search for spectral peaks.

[0014] Furthermore, the specific method for searching spectral peaks using the threshold truncation method is as follows:

[0015] The search begins from the point of maximum power within the Doppler spectrum and proceeds to the left, targeting the last power point in the search direction that is greater than or equal to a set threshold value p. threshold_p1 The frequency corresponding to the frequency point is used as the left boundary f of the initial search spectral peak. left_p1 Then, starting from the point of maximum power within the Doppler spectrum, the search proceeds to the right to obtain the right boundary f of the first search peak. right_p1 ;

[0016] Among them, the power threshold value p threshold_p1 The calculation formula is as follows:

[0017] p threshold_p1 =p max -p threhsold

[0018] In the formula, p max p is the maximum power within the echo spectrum. threhsold The set power threshold.

[0019] Furthermore, in step 2, the formula for estimating the spectral peak bandwidth is as follows:

[0020] BW p1 =f right_p1 -f left_p1 ;

[0021] In the formula, BW p1 For the bandwidth of the initial search spectral peak, f left_p1 f represents the left boundary of the spectral peak. right_p1 This represents the right boundary of the spectral peak;

[0022] Estimating the centroid frequency using the spectral moment method:

[0023]

[0024] In the formula: f c_p1 denoted as the centroid frequency of the initial search spectral peak, f is the Doppler frequency, and p(f) is the Doppler power.

[0025] Further, in step 3, the formula for calculating Bragg's natural frequency is as follows:

[0026]

[0027] In the formula, f B denoted by Bragg's natural frequency, g is the gravitational acceleration, θ is the grazing angle of the electromagnetic wave, and λ is the electromagnetic wavelength.

[0028] Furthermore, step 4 specifically includes the following steps:

[0029] The left boundary f of the left Bragg peak is obtained by using the threshold truncation method. left_lb and right boundary f right_lb and the left boundary f of the Bragg peak on the right. left_rb and right boundary f right_rb ;

[0030] Calculate the average power and centroid frequency of the left Bragg peak based on the left and right boundaries of the left Bragg peak, and calculate the average power and centroid frequency of the right Bragg peak based on the left and right boundaries of the right Bragg peak.

[0031] Compare the average power of the left and right Bragg peaks, and use the centroid frequency of the Bragg peak with the larger average power as the centroid frequency for the secondary spectral peak search.

[0032] Furthermore, the average power of the left and right Bragg peaks is calculated as follows:

[0033]

[0034]

[0035] In the formula, p lb p is the average power of the Bragg peak on the left. rb denoted as the average power of the Bragg peak on the right, f is the Doppler frequency, and p(f) is the Doppler power.

[0036] Furthermore, in step 4, the centroid frequency is estimated using the spectral moment method, the calculation method of which is as follows:

[0037]

[0038]

[0039] In the formula, f c_lb f is the centroid frequency of the left Bragg peak. c_rb The centroid frequency of the Bragg peak on the right is given.

[0040] Further, in step 5, the average centroid frequency of the first and second search spectral peaks is calculated and used as the center frequency, i.e.

[0041] f center =(f c_p1 +f c_p2 ) / 2

[0042] In the formula: f center f is the center frequency. c_p1 To find the centroid frequency of the initial spectral peak, f c_p2 The centroid frequency of the second-order search spectral peak.

[0043] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0044] 1. This invention solves the problem of low center frequency estimation accuracy when there is a large difference in energy between the two Bragg peaks in the echo Doppler spectrum of a river. It uses the threshold truncation method to accurately extract the two Bragg peaks in the echo Doppler spectrum and then estimate the center frequency, thereby improving the estimation accuracy of the center frequency. This lays a good foundation for subsequent measurement of river surface velocity and real-time monitoring of river hydrodynamic parameters.

[0045] 2. This invention takes into account the situation where the energies of the two Bragg peaks are similar, is applicable to various echo spectral types, and has strong robustness;

[0046] 3. Microwave radar can monitor river surface velocity around the clock and is suitable for operation at medium and high water levels. Compared with traditional measuring instruments, it can save a lot of manpower and resources, which is of great significance for improving the automation level of hydrology and the efficiency of flow measurement. Attached Figure Description

[0047] Figure 1 This is a flowchart of the river echo center frequency estimation method based on spectral peak search, as described in an embodiment of the present invention.

[0048] Figure 2 The simulated echo Doppler spectrum of the two Bragg peaks with significant energy differences is shown in the embodiment of the present invention.

[0049] Figure 3 A comparison diagram of the center frequency estimated by an embodiment of the present invention and the center frequency estimated using the spectral moment method. Detailed Implementation

[0050] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, 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 of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0051] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0052] The present invention will be further described below with reference to specific embodiments, but these are not intended to limit the scope of the invention.

[0053] like Figure 1 As shown in the figure, this invention discloses a method for estimating the center frequency of river echoes based on spectral peak search, including the following steps:

[0054] Step 1: Smooth the Doppler spectrum;

[0055] The Doppler spectrum is smoothed at N points to eliminate instantaneous jumps within the Doppler spectrum. In this embodiment, N is set to 3.

[0056] Step 2: Search for spectral peaks and estimate the bandwidth and centroid frequency of the first searched spectral peak;

[0057] In this embodiment, a threshold truncation method is used to search for spectral peaks, specifically:

[0058] The search begins from the point of maximum power within the Doppler spectrum and proceeds to the left, targeting the last power point in the search direction that is greater than or equal to a set threshold value p. threshold_p1 The frequency corresponding to the frequency point is used as the left boundary f of the initial search spectral peak. left_p1 Then, starting from the point of maximum power within the Doppler spectrum, the search proceeds to the right to obtain the right boundary f of the first search peak. right_p1 ;

[0059] Among them, the power threshold value p threshold_p1 The calculation formula is as follows:

[0060] p hhreshold_p1 =p max -p threhsold ;

[0061] In the formula, p max p is the maximum power within the echo spectrum. threhsold In this embodiment, p is the set power threshold. threhsold Set to 6;

[0062] Then, the bandwidth and centroid frequency of the initial search spectral peak are estimated. The formula for estimating the spectral peak bandwidth is as follows:

[0063] BW p1 =f right_p1 -f left_p1 ;

[0064] In the formula, BW p1 The bandwidth of the initial search spectral peak;

[0065] Estimating the centroid frequency using the spectral moment method:

[0066]

[0067] In the formula: f c_p1 denoted as the centroid frequency of the initial search spectral peak, f is the Doppler frequency, and p(f) is the Doppler power.

[0068] Step 3: Determine the bandwidth of the first search peak. If the bandwidth of the peak is greater than twice the Bragg natural frequency, the two Bragg peaks are considered to have similar energies. The centroid frequency of the first search peak is taken as the center frequency and subsequent steps are omitted. Otherwise, the two Bragg peaks are considered to have significant energy differences, and the process proceeds to step 4.

[0069] In this step, the formula for calculating Bragg's natural frequency is as follows:

[0070]

[0071] In the formula, f B denoted by Bragg's natural frequency, g is the gravitational acceleration, θ is the grazing angle of the electromagnetic wave, and λ is the electromagnetic wavelength.

[0072] Step 4: Perform a secondary peak search, estimate the centroid frequency of the secondary search peak, and proceed to step 5;

[0073] In this embodiment, the left boundary f of the left Bragg peak is first obtained by using the threshold truncation method. left_lb and right boundary f right_lb and the left boundary f of the Bragg peak on the right. left_rb and right boundary f right_rb , specifically:

[0074] Using the threshold truncation method combined with the power threshold P of the left Bragg peak threshold_left In the frequency range [f left_p1 -2f B ,f left_p1 -2f r Threshold truncation is performed within ], f r To achieve Doppler resolution, the left boundary f of the left Bragg peak is obtained. left_lb and right boundary f right_lb Then, the threshold truncation method is used in conjunction with the power threshold value P of the Bragg peak on the right side. threshold_right In the frequency range [f right_p1 +2f r ,f left_p1 +2f B Threshold truncation is performed within the range to obtain the left boundary f of the right-hand Bragg peak. left_rb and right boundary f right_rb ;

[0075] Among them, the power threshold P of the Bragg peak on the left threshold_left And the power threshold P of the Bragg peak on the right threshold_right The calculation formula is as follows:

[0076] P threshold_left =p left -p threhsold ,P threshold_right =p right -p threhsold

[0077] In the formula, p left The frequency is equal to f c_p1 -2f B The power at the frequency point, p right For frequency equal to f c_p1 +2f B The power at the frequency point, p threhsold In this embodiment, p is the set power threshold. threhsold Set to 6;

[0078] Then, the average power and centroid frequency of the left Bragg peak are calculated based on the left and right boundaries of the left Bragg peak, and the average power and centroid frequency of the right Bragg peak are calculated based on the left and right boundaries of the right Bragg peak. The formula for calculating the average power is as follows:

[0079]

[0080]

[0081] In the formula, p lb p is the average power of the Bragg peak on the left. rb The average power of the Bragg peak on the right;

[0082] The centroid frequency is estimated using the spectral moment method, and its calculation formula is as follows:

[0083]

[0084]

[0085] In the formula, f c_lb f is the centroid frequency of the left Bragg peak. c_rb The centroid frequency of the Bragg peak on the right;

[0086] Finally, the average power of the left and right Bragg peaks are compared, and the centroid frequency of the Bragg peak with the larger average power is used as the centroid frequency for the secondary spectral peak search.

[0087] If prb >p lb Then f c_p2 = c_rb Otherwise f c_p2 =f c_lb

[0088] Among them, f c_p2 The centroid frequency of the second-order search spectral peak.

[0089] Step 5: Estimate the center frequency based on the centroid frequencies of the spectral peaks obtained from the two searches;

[0090] In this step, the center frequency is estimated based on the centroid frequencies of the spectral peaks from the two searches; the average centroid frequencies of the first and second search spectral peaks are calculated and used as the center frequency.

[0091] That is, f center =(f c_p1 +f c_p2 ) / 2

[0092] In the formula: f center The center frequency.

[0093] Figure 2 The simulated echo Doppler spectrum shows a significant difference in energy between the Bragg double peaks, with the power of the Bragg negative peak being 10 dB higher than that of the Bragg positive peak. Figure 2 It can be seen that two threshold cutoffs can accurately extract the two Bragg peaks in the echo spectrum. (Refer to...) Figure 2 One hundred echo Doppler spectra were simulated, and the center frequency was estimated using both the method of this embodiment and the spectral moment method. The results are as follows. Figure 3 As shown. Figure 3 The results show that the mean absolute error and mean square error of the center frequency estimated by this embodiment are both smaller than those of the spectral moment method. This indicates that when there is a large difference in the energies of the two Bragg peaks, the method of this embodiment can accurately estimate the center frequency, and the estimation accuracy is much higher than that of the spectral moment method.

[0094] The above are merely preferred embodiments of the present invention and are not intended to limit the implementation methods and protection scope of the present invention. Those skilled in the art should recognize that any equivalent substitutions and obvious changes made based on the content of this specification should be included within the protection scope of the present invention.

Claims

1. A river echo center frequency estimation method based on spectral peak search, characterized in that, The method comprises the following steps: Step 1: smoothing the Doppler spectrum; Step 2: searching the spectrum peak, estimating the bandwidth and the center frequency of the first searched spectrum peak; Step 3: judging the bandwidth of the first searched spectrum peak, if the spectrum peak bandwidth is greater than twice the Bragg inherent frequency, it is determined that the energy of two Bragg peaks is close, the center frequency of the first searched spectrum peak is taken as the center frequency and the subsequent steps are omitted, otherwise it is determined that there is a large difference in the energy of two Bragg peaks, and step 4 is turned to; Step 4: performing secondary spectrum peak search, estimating the center frequency of the secondary searched spectrum peak, and turning to step 5; Step 5: estimating the center frequency according to the center frequencies of the spectrum peaks searched twice.

2. The spectral peak search based river echo center frequency estimation method according to claim 1, characterized in that, In step 1, the Doppler spectrum is N-point smoothed.

3. The spectral peak search based river echo center frequency estimation method according to claim 1, characterized in that, In step 2, the threshold cutting method is used to search the spectrum peak.

4. The spectral peak search based river echo center frequency estimation method according to claim 3, characterized in that, The specific method for searching the spectrum peak by using the threshold cutting method is as follows: Search from the maximum power point in the Doppler spectrum to the left side, and take the frequency of the last frequency point with power greater than or equal to the set threshold value as the left boundary of the first search spectrum peak Search from the maximum power point in the Doppler spectrum to the right side, and take the frequency of the last frequency point with power greater than or equal to the set threshold value as the left boundary of the first search spectrum peak Search from the maximum power point in the Doppler spectrum to the right side, and take the frequency of the last frequency point with power greater than or equal to the set threshold value as the left boundary of the first search spectrum peak ; wherein the power threshold value is calculated as follows: wherein is the maximum power within the echo spectrum, is a set power threshold.

5. The spectral peak search based river echo center frequency estimation method according to claim 1, characterized in that, In step 2, the spectrum peak bandwidth estimation formula is as follows: ； wherein is the bandwidth of the first search for the spectral peak, is the left boundary of the spectral peak, is the right boundary of the spectral peak; The center frequency is estimated by using the spectral moment method: In the formula: is the center frequency of the first searched spectral peak, is the Doppler frequency, is the Doppler power.

6. The spectral peak search based river echo center frequency estimation method according to claim 1, characterized in that, In step 3, the calculation formula of the Bragg inherent frequency is as follows: wherein represents the Bragg eigenfrequency, g is the gravitational acceleration, is the electromagnetic wave grazing angle, is the electromagnetic wavelength.

7. The spectral peak search based river echo center frequency estimation method according to claim 1, characterized in that, Step 4 specifically comprises the following steps: The left boundary of the left Bragg peak and the right boundary of the right Bragg peak are searched using a threshold cut-off method and the left boundary of the right Bragg peak and the right boundary of the right Bragg peak and the left boundary of the right Bragg peak ; The average power, the center frequency of the left Bragg peak are calculated according to the left and right boundaries of the left Bragg peak, and the average power, the center frequency of the right Bragg peak are calculated according to the left and right boundaries of the right Bragg peak; The center frequency of the Bragg peak with larger average power is taken as the center frequency of the secondary spectrum peak search.

8. The spectral peak search based river echo center frequency estimation method according to claim 7, characterized in that, The calculation method of the average power of the left Bragg peak and the right Bragg peak is as follows: ； ； wherein Pleft is the average power of the left Bragg peak, Pright is the average power of the right Bragg peak, fD is the Doppler frequency, PD is the Doppler power.

9. The spectral peak search based river echo center frequency estimation method according to claim 7, characterized in that, In step 4, the center frequency is estimated by using the spectral moment method, and the calculation method is as follows: wherein is the centroid frequency of the left Bragg peak, is the centroid frequency of the right Bragg peak, is the Doppler frequency, is the Doppler power.

10. The spectral peak search based river echo center frequency estimation method according to claim 7, characterized in that, In step 5, the average value of the center frequencies of the first searched spectrum peak and the secondary searched spectrum peak is calculated and taken as the center frequency, that is, In the formula: is a center frequency, is a center frequency of a first search for a spectral peak, is a center frequency of a second search for a spectral peak.

Images