A CIR moving target detection and tracking method for ultra-wideband radar

Abstract

The present invention provides an ultra-wideband radar CIR moving target detection and tracking method, comprising: performing wavelet decomposition and reconstruction on CIR data to remove direct current to obtain a first signal; shifting the first signal to obtain a second signal, and calculating the complex correlation coefficient between the second signal and the first signal; when the complex correlation coefficient exceeds a preset multiple of the first background noise, it is judged that a moving target appears, otherwise it is considered that no moving target appears, and a new complex correlation coefficient is obtained again, and the new complex correlation coefficient is weighted with the first background noise to obtain a second background noise; recording the range gate result where the moving target is located; performing fast Fourier transform on the range gate signal where the moving target is located to obtain the speed of the moving target; performing median filtering on the range gate result to obtain a new range gate result; fitting based on the new range gate result and the speed to achieve target tracking. The ultra-wideband radar CIR moving target detection and tracking method provided by the present invention realizes more accurate moving target detection and tracking by wavelet decomposition and reconstruction to remove direct current.

Description

Technical Field

[0001] The invention relates to the field of ultra-wideband technology, and in particular to an ultra-wideband radar CIR moving target detection and tracking method. Background Art

[0002] The CIR (Channel Impulse Response) based on UWB (Ultra Wideband) radar can detect moving targets. Generally speaking, the static environment is detected first to obtain environmental noise, and then the detection signal and the environmental signal are differentiated to detect the environmental energy change, thereby realizing the detection of moving objects. However, this method is usually only applicable to the case where the static environment does not change, and the differential method will double the Gaussian white noise energy in the signal, resulting in inaccurate detection results. In addition, there is a method for detecting objects based on energy peak detection. When the signal-to-noise ratio is high, the distance between the object and the radar can be directly determined based on the energy change of different distance gates, but when the signal-to-noise ratio is low, it is difficult to achieve accurate detection. The detection method that compares different frequency components by analyzing the CIR spectrum will be affected by DC and its nearby frequency components, and will be limited by the effect of DC removal.

[0003] Therefore, it is necessary to provide an ultra-wideband radar CIR moving target detection and tracking method to solve the above problems. Summary of the invention

[0004] The present invention provides an ultra-wideband radar CIR moving target detection and tracking method, which realizes more accurate moving target detection and tracking through wavelet decomposition and reconstruction to remove direct current.

[0005] The embodiment of the present invention provides an ultra-wideband radar CIR moving target detection and tracking method, comprising the following steps:

[0006] Perform wavelet decomposition and reconstruction on the real part and the imaginary part of each range gate of the CIR data to remove the DC and obtain the first signal;

[0007] Shifting the first signal to obtain a second signal, and calculating a complex correlation coefficient between the second signal and the first signal;

[0008] When the complex correlation coefficient exceeds the preset multiple of the first background noise, it is determined that a moving target appears, otherwise it is considered that no moving target appears, and a new complex correlation coefficient is obtained again, and the new complex correlation coefficient is weighted with the first background noise to obtain a second background noise;

[0009] Record the range gate result of the moving target;

[0010] Performing a fast Fourier transform on the range gate signal where the moving target is located to obtain the speed of the moving target;

[0011] Performing median filtering on the range gate result to obtain a new range gate result;

[0012] Target tracking is achieved by fitting based on the new range gate result and the speed.

[0013] Preferably, performing wavelet decomposition and reconstruction on the real part and the imaginary part of each range gate of the CIR data to remove direct current to obtain the first signal comprises:

[0014] Based on the Daubechies fourth-order wavelet, the real and imaginary parts of each range gate of the CIR data are respectively subjected to four-level wavelet decomposition to obtain four detail components and one approximate component. The waveform of the real and imaginary signals of each range gate of the CIR data is reconstructed using the first, second, third, and fourth level detail components obtained by the decomposition to remove the approximate components.

[0015] Preferably, shifting the first signal to obtain a second signal, and calculating a complex correlation coefficient between the second signal and the first signal comprises:

[0016] Take n+1 points in time t to sample the complex signal a i +jb i (i=1,2,...,n+1), where j is the imaginary unit, j 2 = -1, record the first n sampling points as Y1, record the last n sampling points as Y2. Calculate the mean of the two signals respectively:

[0017]

[0018] Wherein, EY1 and EY2 represent the mean values ​​of the first n sampling points and the last n sampling points, respectively.

[0019] Preferably, the variance and covariance of the first n-point sampling signal and the last n-point sampling complex signal are calculated by the following formula:

[0020]

[0021] Among them, EY1 and EY2 represent the mean of the first n sampling signal and the last n sampling complex signal, respectively. It means taking conjugate operation on Y, DY1 and DY2 are the variance of the complex signal of the first n sampling points and the last n sampling points respectively, and Cov(Y1,Y2) is the covariance of the complex signal of the first n sampling points and the last n sampling points.

[0022] Preferably, the multiple correlation coefficient is calculated by the following formula:

[0023]

[0024] in, represents the multiple correlation coefficient.

[0025] Preferably, the mean values ​​EY1 and EY2 of the first n sampling signal and the last n sampling complex signal are equivalent to 0, and the complex correlation coefficient is expressed as follows:

[0026]

[0027] Preferably, the first background noise is obtained in the following manner:

[0028] Complex correlation coefficients are collected for a static environment within a preset time, and weighted average is performed on the complex correlation coefficients collected for the static environment within the preset time to obtain the first background noise.

[0029] Preferably, performing median filtering on the range gate result to obtain a new range gate result includes:

[0030] Determine a time frame with a time length of t, select the range gate result in the time frame from the initial time, replace the middle range gate result with the median of all range gate results in this period of time, shift the time frame backward in sequence and repeat the above operation to remove the high-frequency jump part of the range gate result.

[0031] Preferably, performing a fast Fourier transform on the range gate signal where the moving target is located to obtain the speed of the moving target comprises calculating by the following formula:

[0032]

[0033] Where, v represents the radial velocity of the moving target relative to the radar, and f c represents the center frequency of the ultra-wideband pulse, c represents the speed of light, and f d represents the Doppler frequency deviation caused by the motion of the moving target, f d When it is a positive number, it means that the moving target is moving towards the direction close to the radar. d When it is a negative number, it means that the moving target is moving in a direction away from the radar.

[0034] Preferably, the distance result is obtained by fitting a first distance and a second distance, the first distance is determined based on a range gate at the current moment, and the second distance is calculated by the following formula:

[0035]

[0036] Wherein, D represents the second distance, D0 represents the initial distance of the moving target relative to the radar, N represents the total number of time periods, and v n It represents the moving speed of the moving target in the nth period of time, and t1 represents the length of each period of time.

[0037] Compared with the prior art, the technical solution of the embodiment of the present invention has the following beneficial effects:

[0038] The ultra-wideband radar CIR moving target detection and tracking method provided by an embodiment of the present invention includes: performing wavelet decomposition and reconstruction on the real part and the imaginary part of each range gate of CIR data to remove direct current to obtain a first signal; shifting the first signal to obtain a second signal, and calculating the complex correlation coefficient between the second signal and the first signal; when the complex correlation coefficient exceeds a preset multiple of the first background noise, it is judged that a moving target appears, otherwise it is considered that no moving target appears, and a new complex correlation coefficient is obtained again, and the new complex correlation coefficient is weighted with the first background noise to obtain a second background noise; recording the range gate result where the moving target is located; performing fast Fourier transform on the range gate signal where the moving target is located to obtain the speed of the moving target; performing median filtering on the range gate result to obtain a new range gate result; fitting based on the new range gate result and the speed to achieve target tracking, and removing direct current through wavelet decomposition and reconstruction to achieve more accurate moving target detection and tracking;

[0039] Furthermore, the shifted complex correlation coefficient is calculated for the signal after the DC is removed to reduce the interference of noise on the signal;

[0040] Furthermore, median filtering of the range gate changes can remove repeated jumps in the range gate results caused by signal instability;

[0041] Furthermore, based on the range gate results of the object, the waveforms of different time periods are subjected to fast Fourier transform to obtain the speed information of the moving target;

[0042] Furthermore, the weighting based on the range gate result and the Doppler frequency shift result can improve the detection accuracy as much as possible without changing the radar sampling rate, thus achieving more accurate target detection and tracking. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are some embodiments of the present invention, not all embodiments. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor.

[0044] Figure 1 A schematic flow chart of an ultra-wideband radar CIR moving target detection and tracking method provided by an embodiment of the present invention;

[0045] Figure 2A schematic diagram of a real signal on the 20th range gate in an ultra-wideband radar CIR moving target detection and tracking method provided by an embodiment of the present invention;

[0046] Figure 3 A schematic diagram of an imaginary signal on the 20th range gate in an ultra-wideband radar CIR moving target detection and tracking method provided by an embodiment of the present invention;

[0047] Figure 4 A schematic diagram of the modulus value of a signal after DC removal in an ultra-wideband radar CIR moving target detection and tracking method provided by an embodiment of the present invention;

[0048] Figure 5 A schematic diagram of the modulus value of the shifted complex correlation coefficient of a signal in an ultra-wideband radar CIR moving target detection and tracking method provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] The technical solution of the present invention is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0051] Based on the problems existing in the prior art, an embodiment of the present invention provides an ultra-wideband radar CIR moving target detection and tracking method, which achieves more accurate moving target detection and tracking through wavelet decomposition and reconstruction to remove DC.

[0052] Figure 1 A schematic flow chart of an ultra-wideband radar CIR moving target detection and tracking method provided by an embodiment of the present invention; Figure 2 A schematic diagram of a real signal on the 20th range gate in an ultra-wideband radar CIR moving target detection and tracking method provided by an embodiment of the present invention; Figure 3 A schematic diagram of an imaginary signal on the 20th range gate in an ultra-wideband radar CIR moving target detection and tracking method provided by an embodiment of the present invention; Figure 4 A schematic diagram of the modulus value of a signal after DC removal in an ultra-wideband radar CIR moving target detection and tracking method provided by an embodiment of the present invention; Figure 5A schematic diagram of the modulus value of the shifted complex correlation coefficient of a signal in an ultra-wideband radar CIR moving target detection and tracking method provided by an embodiment of the present invention.

[0053] View now Figure 1 The embodiment of the present invention provides an ultra-wideband radar CIR moving target detection and tracking method, comprising the following steps:

[0054] Step S101: performing wavelet decomposition and reconstruction on the real part and the imaginary part of each range gate of the CIR data to remove the DC and obtain a first signal;

[0055] Step S102: shifting the first signal to obtain a second signal, and calculating a complex correlation coefficient between the second signal and the first signal;

[0056] Step S103: when the complex correlation coefficient exceeds the preset multiple of the first background noise, it is determined that a moving target appears, otherwise it is determined that no moving target appears, and a new complex correlation coefficient is obtained, and the new complex correlation coefficient is weighted with the first background noise to obtain a second background noise;

[0057] Step S104: Record the range gate result of the moving target;

[0058] Step S105: performing a fast Fourier transform on the range gate signal where the moving target is located to obtain the speed of the moving target;

[0059] Step S106: performing median filtering on the range gate result to obtain a new range gate result;

[0060] Step S107: performing fitting based on the new range gate result and the speed to achieve target tracking.

[0061] Specifically, in step S101, it is assumed that the sampling rate of the ultra-wideband radar is F s , set the data processing cycle to That is, δ groups of data are processed each time.

[0062] In a specific implementation, performing wavelet decomposition and reconstruction on the real part and the imaginary part of each range gate of the CIR data to remove the DC to obtain the first signal includes:

[0063] Based on the Daubechies fourth-order wavelet, the real and imaginary parts of each range gate of the CIR data are respectively decomposed into four levels of wavelets to obtain four detail components and one approximate component. The real and imaginary signals of each range gate of the CIR data are reconstructed using the first, second, third, and fourth level detail components obtained by the decomposition to remove the approximate components and retain the motion information of the moving target. Compared with the high-pass filter, this can achieve real-time DC removal operation without any delay.

[0064] When a motion signal appears, the phase will change continuously, so the CIR will show a sine wave-like fluctuation. At this time, the waveform is correlated in time, and the noise is mainly Gaussian white noise, which is not correlated in time. Therefore, the signal can be shifted and the correlation coefficient can be calculated with the original signal. In places where there is no motion signal, when the shifted noise is calculated with itself, the correlation coefficient obtained will be very small; in places where motion signals appear, when the shifted signal is calculated with itself, the correlation coefficient obtained is larger. And the correlation coefficient is already a normalized data type, so in practical applications, the selection of thresholds is also relatively convenient.

[0065] In order to use both real and imaginary information to make the decision more accurate, IQ (In-phase Quadrature) signals are used for joint judgment. Therefore, the complex domain correlation coefficient is derived based on the existing real domain correlation coefficient expression for use.

[0066] In a specific implementation, shifting the first signal to obtain a second signal, and calculating a complex correlation coefficient between the second signal and the first signal includes:

[0067] Take n+1 points in time t to sample the complex signal a i +jb i (i=1,2,...,n+1), where j is the imaginary unit, j 2 = -1, record the first n sampling points as Y1, record the last n sampling points as Y2. Calculate the mean of the two signals respectively:

[0068]

[0069] Wherein, EY1 and EY2 represent the mean values ​​of the first n sampling points and the last n sampling points, respectively.

[0070] In a specific implementation, the variance and covariance of the first n-point sampling signal and the last n-point sampling complex signal are calculated by the following formula:

[0071]

[0072] Among them, EY1 and EY2 represent the mean of the first n sampling signal and the last n sampling complex signal, respectively. It means taking conjugate operation on Y, DY1 and DY2 are the variance of the complex signal of the first n sampling points and the last n sampling points respectively, and Cov(Y1,Y2) is the covariance of the complex signal of the first n sampling points and the last n sampling points.

[0073] In a specific implementation, the complex correlation coefficient is calculated by the following formula:

[0074]

[0075] in, represents the multiple correlation coefficient.

[0076] Specifically, the correlation coefficient obtained is is a complex value, The modulus value is obtained To represent the correlation of the signal.

[0077] The complex correlation coefficient is approximately calculated based on the signal after the DC is removed based on the wavelet transform. Since the DC has been removed from the signal based on the wavelet decomposition and reconstruction, in the process of calculating the complex correlation coefficient, the mean value of the signal is very small compared to the signal itself. Therefore, it can be calculated equivalently to 0.

[0078] In a specific implementation, the mean values ​​EY1 and EY2 of the first n-point sampling signal and the last n-point sampling complex signal are equivalent to 0, and the complex correlation coefficient is expressed as follows:

[0079]

[0080] Since the obtained complex correlation coefficients are all values ​​between 0 and 1, in the process of constant false alarm detection, a method of setting a fixed threshold for target judgment can be used; in addition, the constant false alarm detection method can also be used to judge the moving target, that is, first collect the complex correlation coefficients for a static environment within a period of time, and perform weighted averaging on these data as the first background noise. In the subsequent period of time, if the complex correlation coefficient exceeds the value of A1 times the first background noise, A1 can be a threshold set manually, then it is considered that a moving target has appeared. If the value does not exceed the threshold, it is considered that no moving target has appeared, and the new complex correlation coefficient is weighted with the first background noise to obtain the second background noise. Since the ultra-wideband pulse has a certain width, in the judgment process, when two or three consecutive range gates detect the target, it is considered that the moving target is really detected, reducing the false alarm probability of the radar. When a moving target is detected at multiple distances at the same time, the nearest distance is used for judgment first, because the motion signal detected by other range gates is likely to be a multipath signal of the moving target. Based on this, the change curve of the range gate where the moving target is located obtained based on the complex correlation coefficient can be portrayed.

[0081] In a specific implementation, the first background noise is obtained by:

[0082] Complex correlation coefficients are collected for a static environment within a preset time, and weighted average is performed on the complex correlation coefficients collected for the static environment within the preset time to obtain the first background noise.

[0083] Because there may be many reflection points on a person's body when he moves, and due to factors such as noise and complex correlation coefficient threshold setting, the obtained range gate change is usually not completely accurate and has a certain degree of repeated jump. Therefore, the change curve of the range gate where the moving target is located can be median filtered.

[0084] In a specific implementation, performing median filtering on the range gate result to obtain a new range gate result includes:

[0085] Determine a time frame with a time length of t, select the range gate result in the time frame from the initial time, replace the middle range gate result with the median of all range gate results in this period of time, shift the time frame backward in sequence and repeat the above operation to remove the high-frequency jump part of the range gate result.

[0086] Based on the range gate result of the object obtained by median filtering, perform fast Fourier transform on the complex signal on the range gate and calculate the peak-to-average ratio of the fast Fourier transform result. If the peak-to-average ratio of the fast Fourier transform is greater than the set threshold A2, the Doppler frequency deviation f caused by the object movement is obtained through this frequency point. d .

[0087] In a specific implementation, performing a fast Fourier transform on the range gate signal where the moving target is located to obtain the speed of the moving target includes calculating by the following formula:

[0088]

[0089] Where, v represents the radial velocity of the moving target relative to the radar, and f c represents the center frequency of the ultra-wideband pulse, c represents the speed of light, and f d represents the Doppler frequency deviation caused by the motion of the moving target, f d When it is a positive number, it means that the moving target is moving towards the direction close to the radar. d When it is a negative number, it means that the moving target is moving in a direction away from the radar.

[0090] Since the distance gate result of the object based on the complex correlation coefficient is only accurate within a macroscopic range, the distance accuracy of the ultra-wideband radar is related to the sampling rate of the signal. The distance accuracy is expressed by the following formula:

[0091]

[0092] Where d is the distance accuracy, c is the speed of light, and t represents the sampling interval. Assuming the sampling interval is 1ns, the distance accuracy is 15cm, that is, the detection distances indicated by adjacent range gates differ by 15cm. Therefore, judging the distance of an object based on the range gate is accurate in the macroscopic range, but the accuracy is not high enough. In addition, the distance between the object and the radar can be calculated based on the radial velocity of the object obtained by fast Fourier transform, and the movement of the object in each short period of time is approximately regarded as uniform motion, and the velocity in different time periods is multiplied by time to obtain the approximate displacement of the object in each time period.

[0093] In a specific implementation, the distance result is obtained by fitting a first distance and a second distance, the first distance is determined based on a range gate at the current moment, and the second distance is calculated by the following formula:

[0094]

[0095] Wherein, D represents the second distance, D0 represents the initial distance of the moving target relative to the radar, N represents the total number of time periods, and v n It represents the moving speed of the moving target in the nth period of time, and t1 represents the length of each period of time.

[0096] Although the accuracy of the target distance is high enough based on the object speed obtained by fast Fourier transform, the speed obtained based on the frequency deviation may have certain errors due to the large reflection area of ​​the human body. Therefore, the target distance curve drawn by the result may not be accurate. The target and radar distance curve drawn based on the fast Fourier transform result may differ greatly from the actual distance result in the long-term accumulation process, and the distance curve drawn based on the range gate result is not fine enough due to the resolution problem. Therefore, the detailed description of the target motion trajectory can be achieved by fitting the two, and the detection and tracking of moving targets can be completed.

[0097] Specifically, the distance of the object is characterized by combining the range gate result and the fast Fourier transform result.

[0098] First, the initial position is determined based on the range gate. For any subsequent moment, there are two ways to determine the position. One is to determine the position D1 directly based on the range gate at this moment, and the other is to obtain D2 based on the position at the previous moment and the product of the speed at the previous moment and this time. Each time D1 is compared with D2, when |D1-D2|>B, it means that the difference between the two methods exceeds the tolerance range, and D2 needs to be corrected based on D1 when calculating the distance at the next moment, where B is a manually set threshold.

[0099] For example, when D1<D2, if the speed direction is positive, that is, close to the radar direction, the speed at the next moment needs to be multiplied by a coefficient B1, where B1>1. If the speed direction is negative, that is, away from the radar direction, the speed at the next moment needs to be multiplied by a coefficient B2, where 0<B2<1, to narrow the gap between D1 and D2. When D1>D2, the weighting method is opposite to the above.

[0100] Through the above method, the high accuracy and low precision of the distance information obtained by the range gate is utilized, and the high accuracy and low precision of the distance information obtained by the speed are utilized to combine the two, so as to obtain a distance fitting result that can improve both accuracy and precision.

[0101] View now Figures 2 to 5 , Figure 2 and Figure 3 They are the real and imaginary signals at the 20th range gate respectively. The red box is the position of the motion signal. It can be seen that the signal-to-noise ratio is very poor at this time, and the signal-to-noise amplitude ratio is about 2. Figure 4 is the modulus of the signal after removing DC, that is, the energy image of the motion signal. The red box corresponds to Figure 2 and Figure 3 From the motion signal, we can see that the signal energy is very weak at this time. If we judge whether the target appears or not based only on the signal energy, the result will be very unreliable. Figure 5 is the modulus of the shifted complex correlation coefficient of the signal. It can be seen that after the shifted complex correlation coefficient processing, the signal characteristics of the corresponding position become very obvious again, which greatly improves the signal-to-noise ratio of the signal.

[0102] In summary, the ultra-wideband radar CIR moving target detection and tracking method provided by the embodiment of the present invention includes: performing wavelet decomposition and reconstruction on the real part and the imaginary part of each range gate of the CIR data to remove direct current to obtain a first signal; shifting the first signal to obtain a second signal, and calculating the complex correlation coefficient between the second signal and the first signal; when the complex correlation coefficient exceeds a preset multiple of the first background noise, it is judged that a moving target appears, otherwise it is considered that no moving target appears, and a new complex correlation coefficient is obtained again, and the new complex correlation coefficient is weighted with the first background noise to obtain a second background noise; recording the range gate result where the moving target is located; performing fast Fourier transform on the range gate signal where the moving target is located to obtain the speed of the moving target; performing median filtering on the range gate result to obtain a new range gate result; fitting based on the new range gate result and the speed to achieve target tracking, and removing direct current through wavelet decomposition and reconstruction to achieve more accurate moving target detection and tracking;

[0103] Furthermore, the shifted complex correlation coefficient is calculated for the signal after the DC is removed to reduce the interference of noise on the signal;

[0104] Furthermore, median filtering of the range gate changes can remove repeated jumps in the range gate results caused by signal instability;

[0105] Furthermore, based on the range gate results of the object, the waveforms of different time periods are subjected to fast Fourier transform to obtain the speed information of the moving target;

[0106] Furthermore, the weighting based on the range gate result and the Doppler frequency shift result can improve the detection accuracy as much as possible without changing the radar sampling rate, thus achieving more accurate target detection and tracking.

[0107] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An ultra-wideband radar CIR moving target detection and tracking method, characterized in that: The steps include: The real part and the imaginary part of each range gate of the CIR data are decomposed and reconstructed by wavelet to remove the direct current and obtain the first signal; based on the fourth-order Daubechies wavelet, the real part and the imaginary part of each range gate of the CIR data are respectively decomposed by 4 levels of wavelet to obtain 4 detail components and 1 approximate component, and the real part and the imaginary part of each range gate of the CIR data are reconstructed by waveform using the first, second, third, and fourth level detail components obtained by the decomposition to remove the approximate component; Shifting the first signal to obtain a second signal, and calculating a complex correlation coefficient between the second signal and the first signal; Take n+1 points in time t to sample the complex signal a i +jb i (i=1,2,...,n+1), where j is the imaginary unit, j 2 = -1, the first n sampling points are recorded as Y1, and the last n sampling points are recorded as Y 2， Calculate the mean of the two signals separately: Wherein, EY1 and EY2 represent the mean of the first n sampling signal and the last n sampling complex signal respectively; When the complex correlation coefficient exceeds the preset multiple of the first background noise, it is determined that a moving target appears, otherwise it is considered that no moving target appears, and a new complex correlation coefficient is obtained again, and the new complex correlation coefficient is weighted with the first background noise to obtain a second background noise; Record the range gate result of the moving target; Performing a fast Fourier transform on the range gate signal where the moving target is located to obtain the speed of the moving target; Perform median filtering on the range gate result to obtain a new range gate result; determine a time frame with a time length of t, select the range gate result in the time frame from the initial time, replace the middle range gate result with the median of all range gate results in this period of time, shift the time frame backward in sequence and repeat the above operation to remove the high-frequency jump part of the range gate result; Performing a fitting based on the new range gate result and the speed to achieve target tracking; Performing a fast Fourier transform on the range gate signal where the moving target is located to obtain the speed of the moving target includes calculating by the following formula: Where, v represents the radial velocity of the moving target relative to the radar, and f c represents the center frequency of the ultra-wideband pulse, c represents the speed of light, and f d represents the Doppler frequency deviation caused by the motion of the moving target, f d When it is a positive number, it means that the moving target is moving towards the direction close to the radar. d When it is a negative number, it means that the moving target is moving in a direction away from the radar; The distance result is obtained by fitting a first distance and a second distance, wherein the first distance is determined based on the range gate at the current moment, and the second distance is calculated by the following formula: Wherein, D represents the second distance, D0 represents the initial distance of the moving target relative to the radar, N represents the total number of time periods, and v n It represents the moving speed of the moving target in the nth period of time, and t1 represents the length of each period of time.

2. The ultra-wideband radar CIR moving target detection and tracking method according to claim 1 is characterized in that: The variance and covariance of the first n sampling signals and the last n sampling complex signals are calculated by the following formula: Among them, EY1 and EY2 represent the mean of the first n sampling signal and the last n sampling complex signal, respectively. It means taking conjugate operation on Y, DY1 and DY2 are the variance of the complex signal of the first n sampling points and the last n sampling points respectively, and Cov(Y1,Y2) is the covariance of the complex signal of the first n sampling points and the last n sampling points.

3. The ultra-wideband radar CIR moving target detection and tracking method according to claim 2 is characterized in that: The multiple correlation coefficient is calculated by the following formula: in, represents the multiple correlation coefficient.

4. The ultra-wideband radar CIR moving target detection and tracking method according to claim 3 is characterized in that: The mean values ​​EY1 and EY2 of the first n-point sampling signal and the last n-point sampling complex signal are equivalent to 0, and the complex correlation coefficient is expressed as follows: .

5. The ultra-wideband radar CIR moving target detection and tracking method according to claim 1, characterized in that: The first background noise is obtained by: Complex correlation coefficients are collected for a static environment within a preset time, and weighted average is performed on the complex correlation coefficients collected for the static environment within the preset time to obtain the first background noise.

Images