Methods, apparatus and equipment for high-speed maneuvering target detection with rapid adaptive iteration

Abstract

This invention provides a fast adaptive iterative method, apparatus, and device for high-speed maneuvering target detection. The method includes: constructing a Doppler compensation function based on the linear frequency modulation characteristics of the deambiguous signal time-domain matrix; randomly selecting a subset of points in the two-dimensional signal as initial points; performing Doppler compensation and one-dimensional constant false alarm rate (CFAR) detection on the deambiguous signal time-domain matrix based on the initial points, the Doppler compensation function, and a coordinate ascending search algorithm to obtain a final reference point; calculating the target signal velocity and target signal acceleration based on the final reference point, and using the target signal velocity and target signal acceleration as the target detection result. In this invention, the above steps concentrate the calculation process around data with higher confidence, enabling rapid acquisition of a set of final reference points that have passed the threshold, resulting in better robustness, reduced data computation load on the radar signal processor, and lower hardware requirements for the radar signal processor.

Description

Technical Field

[0001] This invention relates to the field of radar target detection technology, and specifically to a method, apparatus, and equipment for detecting high-speed maneuvering targets with rapid adaptive iteration. Background Technology

[0002] When high-speed maneuvering targets accumulate, the targets have strong speed and acceleration, which will affect the radar echo phase, producing primary and secondary phase modulation, respectively. This causes the radar accumulated spectral peak to spread in the speed and acceleration domains, resulting in a decrease in the radar spectral peak signal-to-noise ratio, which reduces the signal-to-noise ratio before radar detection and the probability of detection afterward.

[0003] In the existing technology, the method to solve this problem is to uniformly construct multiple velocity and acceleration sequences, construct a phase compensation factor based on the sequence, and then search for the velocity and acceleration parameters of the high-speed maneuvering target based on the phase compensation factor, and compensate for the target echo in the parameter domain. However, in order to obtain high-precision motion parameters and low signal-to-noise ratio accumulation gain loss, this method requires a large number of search points, which brings great computational pressure to the radar signal processor. Summary of the Invention

[0004] To address the aforementioned problems in the prior art, this invention provides a method, apparatus, and device for rapid adaptive iteration of high-speed maneuvering target detection.

[0005] The technical problem to be solved by this invention is achieved through the following technical solution:

[0006] In a first aspect, the present invention provides a fast adaptive iterative method for detecting high-speed maneuvering targets, comprising:

[0007] Acquire radar echo data; radar echo data is two-dimensional data.

[0008] The radar echo data is sequentially down-converted and pulse compressed to obtain the frequency domain echo signal.

[0009] Two-dimensional signals are obtained by using preset virtual slow time axis equations, preset velocity ambiguity number equations, and frequency domain echo signals.

[0010] Distance compensation is performed on the two-dimensional signal based on the fuzzy compensation function to obtain the time-domain matrix of the defuzzified signal;

[0011] A Doppler compensation function is constructed by utilizing the linear frequency modulation characteristics of the time-domain matrix of the deambiguous signal; and some points are randomly selected in the two-dimensional signal as initial points.

[0012] Based on the initial point, Doppler compensation function, and coordinate ascending search algorithm, Doppler compensation and one-dimensional constant false alarm rate (CFAR) detection are performed on the time-domain matrix of the unambiguous signal to obtain the final reference point.

[0013] The target signal velocity and target signal acceleration are calculated based on the final reference point, and the target signal velocity and target signal acceleration are used as the target detection results.

[0014] Optionally, the radar echo data is represented as:

[0015]

[0016] Wherein s(t,t m ) represents radar echo data, rect(·) represents the window function, t represents fast time, τ represents radar echo delay, Tp represents pulse duration, exp represents the exponential function, j represents the imaginary number, μ represents the frequency modulation, and f c This indicates the carrier frequency of the radar transmission signal.

[0017] Optionally, the radar echo data is sequentially down-converted and pulse compressed to obtain a frequency domain echo signal, including:

[0018] The radar echo data is down-converted to obtain the baseband echo signal;

[0019] The baseband echo signal is pulse-compressed to obtain the frequency domain echo signal.

[0020] The frequency domain echo signal is represented as:

[0021]

[0022] Wherein S(f,t) m ) represents the frequency domain echo signal, t m Let |P(f)| represent the slow time interval, and f represent the distance frequency variable relative to the fast time interval t. 2 This represents the signal energy in the fast time-frequency domain versus the slow time-time domain, where j represents the imaginary number, c represents the speed of light, and f represents the frequency domain. c R0 represents the carrier frequency of the radar transmitted signal, and R0 represents the moving target at time t. m =0, v represents the initial radial distance of the target, a represents the radial acceleration of the target, and exp represents the exponential function.

[0023] Optionally, the preset virtual slow time axis equation is expressed as:

[0024] τ m f c =(f c +f)t m ;

[0025] Where, τ m Represents a virtual slow-time variable;

[0026] The preset speed fuzzy number equation is expressed as:

[0027] v = M amb v a +v0;

[0028] Among them, M amb V represents the folding factor. a Indicates blind speed, v a =λPRF / 2, where λ represents the radar transmission wavelength, PRF represents the pulse repetition frequency, and v0 represents the unambiguous velocity.

[0029] Two-dimensional signals are represented as:

[0030]

[0031] Wherein, S1(f,τ) m ) represents a two-dimensional signal.

[0032] Optionally, distance compensation is performed on the two-dimensional signal based on the fuzzy compensation function to obtain the time-domain matrix of the defuzzified signal, including:

[0033] Multiplying the two-dimensional signal by the ambiguity compensation function yields the time-domain matrix of the deambigued signal;

[0034] The fuzzy compensation function is expressed as:

[0035]

[0036] Among them, H m (f,τ m ) represents the fuzzy compensation function, and M represents the fuzzy number variable.

[0037] Alternatively, the Doppler compensation function can be expressed as:

[0038]

[0039] H dechirp (t m ) represents the Doppler compensation function, rect(·) represents the window function, and T p f represents the pulse duration. dc f represents the centroid of the Doppler frequency. dr This indicates the Doppler modulation frequency.

[0040] Optionally, based on the initial point, Doppler compensation function, and coordinate ascent search algorithm, Doppler compensation and one-dimensional constant false alarm rate (CFAR) detection are performed on the time-domain matrix of the deblurred signal to obtain the final reference point, including:

[0041] S201. Extract the azimuth signal of the same distance cell from the time domain matrix of the defuzzified signal according to the preset rules to obtain multiple sub-azimuth matrices;

[0042] S202. Compensate multiple sub-azimuth matrices according to the Doppler compensation function to obtain the first sub-azimuth matrix;

[0043] S203. Perform one-dimensional CA-CFAR detection on the first sub-direction matrix based on the initial point, and take the target point that is greater than the first preset threshold value as the first current reference point.

[0044] S204. Perform one-dimensional CA-CFAR detection on the first sub-azimuth matrix based on the first current reference point, and take the target point that is greater than the second preset threshold value as the second current reference point.

[0045] S205. The first current reference point and the second current reference point are combined to form an intermediate reference point;

[0046] S206. Using the intermediate reference point as the initial point, repeat S203-S205 until the coordinate difference between the intermediate reference point of the current iteration and the intermediate reference point of the previous iteration is less than the preset coordinate change threshold.

[0047] S207. Use the intermediate reference point corresponding to the most recent execution of S205 as the final reference point.

[0048] Optionally, the first sub-orientation matrix is ​​represented as:

[0049]

[0050] Among them, s n (t m ) 1 Let A represent the first sub-azimuth matrix, A represent the signal amplitude after pulse compression, rect(·) represent the window function, and t represent the first sub-azimuth matrix. m T represents slow time. p f represents the pulse duration. dc f represents the centroid of the Doppler frequency. dr Indicates Doppler frequency modulation. This represents the centroid of the Doppler frequency compensated by the Doppler compensation function. This represents the Doppler modulation frequency compensated by the Doppler compensation function.

[0051] In a second aspect, the present invention provides a high-speed maneuvering target detection device with rapid adaptive iteration, which includes: an acquisition unit, a pulse compression processing unit, a range compensation unit, a construction unit, a Doppler compensation unit, and a calculation unit.

[0052] The acquisition unit is used to: acquire radar echo data; the radar echo data is two-dimensional data;

[0053] The pulse compression processing unit is used to: sequentially perform down-conversion and pulse compression processing on the radar echo data to obtain the frequency domain echo signal;

[0054] The acquisition unit is also used to: acquire two-dimensional signals by using a preset virtual slow time axis equation, a preset velocity ambiguity number equation, and a frequency domain echo signal;

[0055] The distance compensation unit is used to: perform distance compensation on a two-dimensional signal based on a fuzzy compensation function to obtain the time-domain matrix of the defuzzified signal;

[0056] The construction unit is used to: construct a Doppler compensation function by utilizing the linear frequency modulation characteristics of the time-domain matrix of the deambiguous signal; and randomly select some points in the two-dimensional signal as initial points;

[0057] The Doppler compensation unit is used to: perform Doppler compensation and one-dimensional constant false alarm rate (CFAR) detection on the time-domain matrix of the unambiguous signal based on the initial point, Doppler compensation function, and coordinate ascending search algorithm, and obtain the final reference point;

[0058] The calculation unit is used to: calculate the target signal velocity and target signal acceleration based on the final reference point, and use the target signal velocity and target signal acceleration as the target detection result.

[0059] Thirdly, the present invention provides a high-speed maneuvering target detection device with rapid adaptive iteration, comprising: a processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, and when the high-speed maneuvering target detection device with rapid adaptive iteration is running, the processor communicates with the storage medium through the bus, and the processor executes the machine-readable instructions to perform the steps of the high-speed maneuvering target detection method with rapid adaptive iteration described in the first aspect above.

[0060] This invention provides a fast adaptive iterative method, apparatus, and device for high-speed maneuvering target detection. The method includes: acquiring radar echo data; the radar echo data is two-dimensional data; performing down-conversion and pulse compression processing on the radar echo data sequentially to obtain a frequency domain echo signal; acquiring a two-dimensional signal using a preset virtual slow time axis equation, a preset velocity ambiguity number equation, and the frequency domain echo signal; performing range compensation on the two-dimensional signal based on a ambiguity compensation function to obtain a deambiguous signal time domain matrix; constructing a Doppler compensation function using the linear frequency modulation characteristics of the deambiguous signal time domain matrix; randomly selecting some points in the two-dimensional signal as initial points; performing Doppler compensation and one-dimensional constant false alarm rate (CFAR) detection on the deambiguous signal time domain matrix based on the initial points, the Doppler compensation function, and a coordinate ascending search algorithm to obtain a final reference point; calculating the target signal velocity and target signal acceleration based on the final reference point, and using the target signal velocity and target signal acceleration as the target detection result. In this invention, a fuzzy compensation function is used to perform range compensation on the two-dimensional signal, which improves the accuracy of the constructed Doppler compensation function. Secondly, based on the Doppler compensation function, the coordinate ascending search algorithm, and the one-dimensional CA-CFAR detection algorithm, the final reference point is obtained through rapid adaptive iteration. Finally, the target signal velocity and target signal acceleration are calculated based on the final reference point. Based on the above steps, the calculation process is more focused on the data with high confidence, and the set of final reference points that have passed the threshold can be quickly obtained, which has better robustness, reduces the data calculation load of the radar signal processor, and lowers the hardware requirements of the radar signal processor.

[0061] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0062] Figure 1 A flowchart illustrating a fast adaptive iterative high-speed maneuvering target detection method provided in an embodiment of the present invention;

[0063] Figure 2 This is a schematic diagram illustrating the adaptive iterative search process in an embodiment of the present invention;

[0064] Figure 3 The result diagram is shown for the high-speed maneuvering target detection method based on fast adaptive iteration provided in the embodiments of the present invention;

[0065] Figure 4 This is a comparison chart of the detection performance of the method of the present invention and the traditional uniform search method;

[0066] Figure 5 A schematic diagram of the structure of a high-speed maneuvering target detection device with rapid adaptive iteration provided in an embodiment of the present invention;

[0067] Figure 6This is a schematic diagram of the structure of a high-speed maneuvering target detection device with rapid adaptive iteration provided in an embodiment of the present invention. Detailed Implementation

[0068] The present invention will be further described in detail below with reference to specific embodiments, but the implementation of the present invention is not limited thereto.

[0069] To reduce the amount of data computation by the radar signal processor and lower its hardware requirements, this invention provides a fast adaptive iterative method for detecting high-speed moving targets. Figure 1 This is a flowchart illustrating a fast adaptive iterative high-speed maneuvering target detection method provided in an embodiment of the present invention. Figure 1 As shown, it includes:

[0070] S101, Acquire radar echo data.

[0071] Radar echo data is two-dimensional data.

[0072] In this embodiment of the invention, the radar transmission signal can be represented as:

[0073]

[0074] Where s(t) represents the radar transmitted signal, rect(·) represents the window function, μ represents the frequency modulation, and T p f represents the pulse duration. c Let represent the radar transmit signal carrier frequency, t represent fast time, exp represent the exponential function, and j represent the imaginary number.

[0075] Suppose a moving target in slow time t m When the initial radial distance is R0 (=0), the initial radial velocity of the target is ν, and the radial acceleration is a (very small). Higher-order motion parameters are not considered here. In this case, the radial distance between the maneuvering target and the radar can be expressed as follows:

[0076]

[0077] Where R0 represents the initial radial distance, v represents the initial radial velocity of the target, a represents the radial acceleration of the target, and R s (t m () indicates the radial distance between the moving target and the radar.

[0078] The radar echo data can then be represented as:

[0079]

[0080] Wherein s(t,t m) represents radar echo data, τ represents radar echo delay, τ = 2R s (t m ) / c, where c is the speed of light, R s (t m R represents the radial distance between the radar and the target during the observation period. s (t m ) is slow time t m The function rect(·) represents the window function.

[0081] S102. The radar echo data is sequentially down-converted and pulse compressed to obtain the frequency domain echo signal.

[0082] Optionally, the radar echo data is represented as:

[0083]

[0084] Wherein s(t,t m ) represents radar echo data, rect(·) represents the window function, t represents fast time, τ represents radar echo delay, and T p The pulse duration is represented by exp, the exponential function is represented by j, the imaginary number is represented by μ, and the frequency modulation is represented by f. c This indicates the carrier frequency of the radar transmission signal.

[0085] Optionally, S102 may specifically include:

[0086] The radar echo data is down-converted to obtain the baseband echo signal.

[0087] In this embodiment of the invention, the baseband echo signal can be represented as:

[0088]

[0089] Wherein, s'(t,t m ) represents the baseband echo signal, λ = c / f c This indicates the wavelength of the radar signal.

[0090] The baseband echo signal is pulse-compressed, and the radial distance between the target and the radar is substituted to obtain the frequency domain echo signal, which is represented as:

[0091]

[0092] Wherein S(f,t) m ) represents the frequency domain echo signal, t m Let |P(f)| represent the slow time interval, and f represent the distance frequency variable relative to the fast time interval t. 2This represents the signal energy in the fast time-frequency domain versus the slow time-time domain, where j represents the imaginary number, c represents the speed of light, and f represents the frequency domain. c R0 represents the carrier frequency of the radar transmitted signal, and R0 represents the moving target at time t. m =0, v represents the initial radial distance of the target, a represents the radial acceleration of the target, and exp represents the exponential function.

[0093] S103. Obtain a two-dimensional signal by using a preset virtual slow time axis equation, a preset velocity ambiguity number equation, and a frequency domain echo signal.

[0094] Optionally, the preset virtual slow time axis equation is expressed as:

[0095] τ m f c =(f c +f)t m ;

[0096] Where τm represents the virtual slow time variable;

[0097] The preset speed fuzzy number equation is expressed as:

[0098] v = Mambv a + v0;

[0099] Among them, Mam b V represents the folding factor. a Indicates blind speed, v a =λPRF / 2, where λ represents the radar transmission wavelength, PRF represents the pulse repetition frequency, and v0 represents the unambiguous velocity.

[0100] Two-dimensional signals are represented as:

[0101]

[0102] Where S1(f,τm) represents a two-dimensional signal.

[0103] S104. Based on the fuzzy compensation function, distance compensation is performed on the two-dimensional signal to obtain the time-domain matrix of the defuzzified signal.

[0104] Optionally, S104 may specifically include:

[0105] Multiplying the two-dimensional signal by the ambiguity compensation function yields the time-domain matrix of the deambigued signal;

[0106] The fuzzy compensation function is expressed as:

[0107]

[0108] Among them, H m (f,τm ) represents the fuzzy compensation function, and M represents the fuzzy number variable.

[0109] The time-domain matrix representation of the defuzzified signal is as follows:

[0110]

[0111] Wherein, S1'(f,τ) m ) represents the time-domain matrix of the defuzzified signal, and M represents the fuzzy number variable.

[0112] As can be seen, when M = M amb When the third exponent term in the above formula is 1, the coupling between the distance frequency domain and the velocity is eliminated, and compensation for distance movement is achieved.

[0113] For S1'(f,τ) m Performing an inverse Fourier transform yields the time-domain expression for the time-domain matrix of the defuzzified signal:

[0114]

[0115] S105. Construct a Doppler compensation function by utilizing the linear frequency modulation characteristics of the time-domain matrix of the deambiguous signal; and randomly select some points in the two-dimensional signal as initial points.

[0116] Alternatively, the Doppler compensation function can be expressed as:

[0117]

[0118] H dechirp (t m ) represents the Doppler compensation function, rect(·) represents the window function, and T p f represents the pulse duration. dc f represents the centroid of the Doppler frequency. dr This indicates the Doppler modulation frequency.

[0119] S106. Based on the initial point, Doppler compensation function, and coordinate ascending search algorithm, Doppler compensation and one-dimensional constant false alarm rate detection are performed on the time-domain matrix of the deblurred signal to obtain the final reference point.

[0120] Optionally, S106 may specifically include:

[0121] S201. Extract the azimuth signal of the same distance cell from the time domain matrix of the defuzzified signal according to the preset rules to obtain multiple sub-azimuth matrices.

[0122] S202. Based on the Doppler compensation function, multiple sub-azimuth matrices are compensated to obtain the first sub-azimuth matrix.

[0123] S203. Perform one-dimensional CA-CFAR detection on the first sub-direction matrix based on the initial point, and take the target point that is greater than the first preset threshold value as the first current reference point.

[0124] S204. Perform one-dimensional CA-CFAR detection on the first sub-azimuth matrix based on the first current reference point, and take the target point that is greater than the second preset threshold value as the second current reference point.

[0125] S205. The first current reference point and the second current reference point are combined to form an intermediate reference point.

[0126] S206. Using the intermediate reference point as the initial point, repeat S203-S205 until the coordinate difference between the intermediate reference point of the current iteration and the intermediate reference point of the previous iteration is less than the preset coordinate change threshold.

[0127] S207. Use the intermediate reference point corresponding to the most recent execution of S205 as the final reference point.

[0128] It should be noted that, in the embodiments of the present invention, each row of the deambiguous signal time domain matrix can be used as a unit to extract the same row of matrix data from the deambiguous signal time domain matrix to obtain multiple sub-azimuth matrices.

[0129] In this embodiment of the invention, the sub-orientation matrix can be represented as:

[0130]

[0131] Wherein s(t) m ) represents the sub-azimuth matrix, and A represents the signal amplitude after pulse compression processing.

[0132]

[0133] Furthermore, it should be noted that the Doppler compensation function is constructed by utilizing the linear frequency modulation (LFM) characteristics of the time-domain matrix of the deambiguous signal. Specifically, this can be achieved by utilizing the LFM characteristics of the sub-azimuth matrix. Specifically, according to the above equation, the sub-azimuth matrix exhibits obvious LFM signal characteristics, and Doppler motion correction can be achieved using the de-lFM method. Since the true Doppler frequency centroid and Doppler modulation frequency are not actually known, a Doppler compensation function is constructed, forming a two-dimensional matrix of Doppler frequency centroid and Doppler modulation frequency.

[0134] In this embodiment of the invention, the azimuth signal is extracted by searching for possible Doppler frequency centroids and Doppler modulation frequencies, thereby transforming the time-domain signal into the Doppler frequency-Doppler modulation frequency domain. To reduce computational load, this invention differs from traditional traversal methods by employing an adaptive iterative search using a coordinate ascending method. Therefore, this method requires randomly selecting a certain number of points from the constructed two-dimensional Doppler frequency centroid-Doppler modulation frequency signal as initial points for subsequent operations. It should be noted that the selection of initial points should be flexibly adjusted based on factors such as the actual number of signal points and signal strength.

[0135] Furthermore, an adaptive iterative initial point selection based on the coordinate ascent method is employed, along with Doppler compensation and one-dimensional constant false alarm rate (CFAR) detection, and the detection results are recorded. Assuming there are n reference points, the coordinates of each initial point in the two-dimensional matrix of Doppler frequency centroid-Doppler modulation frequency are (f... dr,n ,f dc,n ), f dr,n f represents the x-coordinate of the nth initial point. dc,n The y-coordinate of the nth initial point. This can be broken down into the following steps:

[0136] For the initial point (f) dr,n ,f dc,n Using its Doppler frequency centroid (the x-axis of the two-dimensional matrix of Doppler frequency centroid-Doppler modulation frequency) and all possible Doppler modulation frequencies as the traversal range for demodulation processing, Doppler compensation is performed on the distance-compensated time-domain two-dimensional signal, that is, demodulation processing is performed on all points on the Doppler frequency centroid axis where the reference point is located.

[0137] The azimuth signal s(t) m Multiply by the Doppler compensation function H dechirp (t m The first sub-orientation matrix can be obtained.

[0138] Optionally, the first sub-orientation matrix is ​​represented as:

[0139]

[0140] Among them, s n (t m ) 1 Let A represent the first sub-azimuth matrix, A represent the signal amplitude after pulse compression, rect(·) represent the window function, and t represent the first sub-azimuth matrix. m T represents slow time. p f represents the pulse duration. dc f represents the centroid of the Doppler frequency. dr Indicates Doppler frequency modulation. This represents the centroid of the Doppler frequency compensated by the Doppler compensation function. This represents the Doppler modulation frequency compensated by the Doppler compensation function.

[0141] The Doppler-compensated signal s n (t m ) 1 Perform one-dimensional CA-CFAR detection, record target points that cross the threshold, and select new reference points. This can be done in the following two steps:

[0142] 1. For the compensated signal s n (t m ) 1 One-dimensional constant false alarm rate (CFAR) detection was performed to obtain target points that crossed the threshold.

[0143] 1.1 For target points that exceed the threshold, select the point with the largest ratio of signal amplitude to the corresponding detection threshold, and record the coordinates of that point (f). d ′ r,n ,f d ′ c,n ), which serves as the first current reference point.

[0144] 1.2. For the first current reference point (f) d ′ r,n ,f d ′ c,n Using its Doppler modulation frequency (Doppler frequency centroid - Doppler modulation frequency two-dimensional matrix y-axis) and all possible Doppler frequency centroids as the traversal range, Doppler compensation is performed on the distance-compensated time-domain two-dimensional signal. That is, demodulation processing is performed on all points on the Doppler modulation frequency axis where the first current reference point is located to obtain the Doppler y-axis compensated signal s. n (t m ) 2 .

[0145] 2. Compensation signal s for the Doppler y-axis n (t m ) 2 Perform one-dimensional CA-CFAR detection, record target points that cross the threshold, and select a second current reference point. This can be done in the following two steps:

[0146] 2.1 Regarding s n ′(t m One-dimensional constant false alarm rate (CFAR) detection was performed separately.

[0147] 2.2, for each s n ′(t m After performing constant false alarm rate (CFAR) detection, select the point with the largest ratio of signal amplitude to the corresponding threshold, and record the coordinates of that point (f). d ′ r,n ,f d ′ c,n), as the second current reference point.

[0148] S107. Calculate the target signal velocity and target signal acceleration based on the final reference point, and use the target signal velocity and target signal acceleration as the target detection result.

[0149] This invention provides a fast adaptive iterative method for high-speed maneuvering target detection, comprising: performing range compensation on a two-dimensional signal using a fuzzy compensation function to improve the accuracy of the constructed Doppler compensation function; secondly, based on the Doppler compensation function, a coordinate ascending search algorithm, and a one-dimensional CA-CFAR detection algorithm, rapidly adaptively iterating to accumulate and obtain a final reference point; and finally, calculating the target signal velocity and target signal acceleration based on the final reference point. Based on the above steps, the calculation process is more focused on data with high confidence, enabling rapid acquisition of a set of final reference points that have passed the threshold, exhibiting better robustness, reducing the data computation load of the radar signal processor, and lowering the hardware requirements of the radar signal processor.

[0150] It should be noted that, in this embodiment of the invention, when the final reference point coordinates are... hour, Indicates the x-coordinate of the final reference point. This represents the ordinate of the final reference point. The target signal velocity and target signal acceleration can then be calculated using the following formulas:

[0151]

[0152] Where v' represents the velocity of the target signal and a' represents the acceleration of the target signal.

[0153] The fast adaptive iterative high-speed maneuvering target detection method provided in this invention combines a demodulation accumulation algorithm (Doppler compensation for two-dimensional time-domain signals), a coordinate ascending search algorithm, and a one-dimensional CA-CFAR (constant false alarm rate) detection algorithm. This method enables rapid adaptive iterative accumulation, search, and detection, acquiring a target range-velocity point cloud set that exceeds a threshold. This achieves fast and robust detection of high-speed maneuvering targets, significantly reducing the computational load of such radar signal processing algorithms and accelerating the hardware development process of radar signal processing modules. It is of great significance for the detection of high-speed maneuvering targets.

[0154] To verify the effectiveness of the method provided in this embodiment of the invention, simulation experiments were conducted. This embodiment constructs echo data based on a signal model, and then performs target enhancement and detection on it. Please refer to... Figure 2 , Figure 3 and Figure 4 .

[0155] Figure 2This diagram illustrates the adaptive iterative search process according to an embodiment of the present invention. The initial points are selected by sampling along the diagonal, with ten initial points chosen at a uniform step size, as shown by the red hollow circle. Next, these initial points are used as reference points. Based on the coordinates of the initial points, a Doppler compensation function is constructed using the coordinates of points on the x-axis or y-axis where they lie. Doppler compensation and one-dimensional CFAR detection are then performed to obtain new reference points. This iterative search is repeated continuously. Points undergoing Doppler compensation during the entire iteration process are shown by the black dotted line, and the updated reference points are shown by the blue triangles. Finally, the coordinates of the reference points no longer change, the search ends, and the spectral peak of the target is obtained, as shown by the red solid circle. It can be seen that through adaptive iterative search, continuously seeking the maximum value, the position and amplitude of the target spectral peak are finally obtained, completing target detection.

[0156] Figure 3 The figure shows the results of the high-speed maneuvering target detection method based on fast adaptive iteration provided in the embodiments of the present invention. Figure 3 Figure (a) shows the accumulated sampling diagram of the high-speed maneuvering target detection method. It reflects the search results of the high-speed maneuvering target detection method in the Doppler frequency modulation parameter domain and the Doppler centroid parameter domain using the coordinate ascending method. The x-axis represents the Doppler frequency modulation parameter domain, the y-axis represents the Doppler centroid domain, and the z-axis represents the amplitude of the target spectrum peak in the parameter domain. Figure 3 Figure (b) is a schematic diagram of the results accumulated by the high-speed maneuvering target detection method after threshold detection. The red surface is the CFAR detection threshold surface, and the exposed part of the surface reflects the target spectral peak crossing the threshold. The x-axis represents the Doppler frequency modulation parameter domain, the y-axis represents the Doppler centroid domain, and the z-axis represents the target spectral peak amplitude and the detection and false alarm threshold amplitude in the parameter domain. Figure 3 As shown in Figure (a), the method of this invention can filter out most regions with low confidence in the parameter domain space and quickly find the target spectral peak. Furthermore, from... Figure 3 Figure (b) shows that by combining the CFAR detector, the target can be effectively detected and the target range-velocity point cloud set can be obtained.

[0157] Figure 4 This chart compares the detection performance of the method of this invention with that of the traditional uniform search method. It mainly statistically analyzes the relationship between the accumulated gain loss and the number of search points used in the two algorithms. The x-axis represents the number of search points in the two-dimensional parameter domain of Doppler frequency centroid-Doppler modulation frequency, and the y-axis represents the accumulated gain loss in the two-dimensional parameter domain of Doppler frequency centroid-Doppler modulation frequency. Figure 4As can be seen, under the same cumulative gain loss, the method of this invention (adaptive iterative search) requires significantly fewer search points than the traditional uniform search method, reflecting the ability of this invention to quickly detect targets. Correspondingly, under the same search point conditions, the method proposed in this invention has a significantly smaller cumulative gain loss than the traditional uniform search method, reflecting the accuracy of this invention in detecting targets.

[0158] The method provided in this embodiment of the invention can be applied to electronic devices. Specifically, the electronic device can be a desktop computer, a portable computer, a smart mobile terminal, a server, etc., and this embodiment of the invention does not limit the application to such devices.

[0159] Based on the same inventive concept, embodiments of the present invention also provide a high-speed maneuvering target detection device with rapid adaptive iteration. Figure 5 This is a schematic diagram of a high-speed maneuvering target detection device with rapid adaptive iteration provided in an embodiment of the present invention. Figure 5 As shown, it includes: an acquisition unit 501, a pulse compression processing unit 502, a distance compensation unit 503, a construction unit 504, a Doppler compensation unit 505, and a calculation unit 506;

[0160] The acquisition unit 501 is used to: acquire radar echo data; the radar echo data is two-dimensional data;

[0161] The pulse compression processing unit 502 is used to: sequentially perform down-conversion and pulse compression processing on the radar echo data to obtain a frequency domain echo signal;

[0162] The acquisition unit 501 is also used to: acquire a two-dimensional signal by means of a preset virtual slow time axis equation, a preset velocity ambiguity number equation and a frequency domain echo signal;

[0163] The distance compensation unit 503 is used to: perform distance compensation on the two-dimensional signal based on the fuzzy compensation function to obtain the time-domain matrix of the defuzzy signal;

[0164] The construction unit 504 is used to: construct a Doppler compensation function by utilizing the linear frequency modulation characteristics of the time-domain matrix of the deambiguous signal; and randomly select some points in the two-dimensional signal as initial points;

[0165] The Doppler compensation unit 505 is used to: perform Doppler compensation and one-dimensional constant false alarm detection on the time-domain matrix of the unambiguous signal based on the initial point, the Doppler compensation function and the coordinate ascending search algorithm, to obtain the final reference point;

[0166] The calculation unit 506 is used to: calculate the target signal velocity and target signal acceleration based on the final reference point, and use the target signal velocity and target signal acceleration as the target detection result.

[0167] Figure 6This is a schematic diagram of a high-speed maneuvering target detection device with rapid adaptive iteration provided in an embodiment of the present invention. It includes a processor 610, a storage medium 620, and a bus 630. The storage medium 620 stores machine-readable instructions executable by the processor 610. When the high-speed maneuvering target detection device with rapid adaptive iteration is running, the processor 610 communicates with the storage medium 620 via the bus 630. The processor 610 executes the machine-readable instructions to perform the steps of the above-described method embodiment. The specific implementation and technical effects are similar and will not be described in detail here.

[0168] The storage medium may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the storage medium may also be at least one storage device located remotely from the aforementioned processor.

[0169] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0170] It should be noted that the terms "first," "second," etc., are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of apparatuses and methods consistent with some aspects of the invention.

[0171] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Furthermore, those skilled in the art can combine and integrate the different embodiments or examples described in this specification.

[0172] Although the invention has been described herein in conjunction with various embodiments, those skilled in the art, by reviewing the accompanying drawings and the disclosure, will understand and implement other variations of the disclosed embodiments in carrying out the claimed invention. In the description of the invention, the word "comprising" does not exclude other components or steps, "a" or "an" does not exclude a plurality, and "a plurality" means two or more, unless otherwise explicitly specified. Furthermore, while different embodiments may describe certain measures, this does not mean that these measures cannot be combined to produce good results.

[0173] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the inventive concept, and all such modifications and substitutions should be considered within the scope of protection of the present invention.

Claims

1. A fast adaptive iterative method for detecting high-speed maneuvering targets, characterized in that, include: Acquire radar echo data; The radar echo data is two-dimensional data; The radar echo data is sequentially down-converted and pulse compressed to obtain a frequency domain echo signal; Two-dimensional signals are obtained by using a preset virtual slow time axis equation, a preset velocity ambiguity number equation, and the frequency domain echo signal. Distance compensation is performed on the two-dimensional signal based on the fuzzy compensation function to obtain the time-domain matrix of the defuzzified signal; A Doppler compensation function is constructed using the linear frequency modulation characteristics of the time-domain matrix of the deambiguous signal; and some points are randomly selected in the two-dimensional signal as initial points. Based on the initial point, the Doppler compensation function, and the coordinate ascending search algorithm, Doppler compensation and one-dimensional constant false alarm rate (CFAR) detection are performed on the time-domain matrix of the unambiguous signal to obtain the final reference point. The target signal velocity and target signal acceleration are calculated based on the final reference point, and the target signal velocity and target signal acceleration are used as the target detection result. The Doppler compensation function is expressed as follows: ； Represents the Doppler compensation function. Represents the window function. Indicates the pulse duration. Indicates the centroid of the Doppler frequency. Indicates the Doppler modulation frequency; The process of performing Doppler compensation and one-dimensional constant false alarm rate (CFAR) detection on the time-domain matrix of the unambiguous signal based on the initial point, the Doppler compensation function, and the coordinate ascent search algorithm to obtain the final reference point includes: S201. Extract the azimuth signal of the same distance unit from the time domain matrix of the defuzzified signal according to the preset rules to obtain multiple sub-azimuth matrices; S202. The multiple sub-azimuth matrices are compensated according to the Doppler compensation function to obtain the first sub-azimuth matrix; S203. Perform one-dimensional CA-CFAR detection on the first sub-azimuth matrix based on the initial point, and take the target point that is greater than the first preset threshold value as the first current reference point; S204. Perform one-dimensional CA-CFAR detection on the first sub-azimuth matrix based on the first current reference point, and take the target point that is greater than the second preset threshold value as the second current reference point. S205. The first current reference point and the second current reference point are combined to form an intermediate reference point; S206. Using the intermediate reference point as the initial point, repeat S203-S205 until the coordinate difference between the intermediate reference point in the current iteration and the intermediate reference point in the previous iteration is less than a preset coordinate change threshold. S207. Take the intermediate reference point corresponding to the most recent execution of S205 as the final reference point.

2. The fast adaptive iterative high-speed maneuvering target detection method according to claim 1, characterized in that, The radar echo data is represented as follows: ； in, Represents radar echo data, Represents the window function. Indicates a fast time. Indicates radar echo delay. Indicates the pulse duration. Represents an exponential function. represents an imaginary number, Indicates frequency modulation. This indicates the carrier frequency of the radar transmission signal.

3. The fast adaptive iterative high-speed maneuvering target detection method according to claim 1, characterized in that, The process of sequentially down-converting and pulse compression the radar echo data to obtain a frequency domain echo signal includes: The radar echo data is down-converted to obtain the baseband echo signal; The baseband echo signal is pulse-compressed to obtain the frequency domain echo signal; The frequency domain echo signal is represented as: ； in, Represents the frequency domain echo signal. Indicates slow time. Indicates relative to fast time Distance frequency variable, This represents the signal energy in the fast time-frequency domain and the slow time-time domain. represents an imaginary number, Represents the speed of light. Indicates the carrier frequency of the radar transmitted signal. Indicates the moving target is in The initial radial distance at that time, Indicates the target's initial radial velocity. This represents the radial acceleration of the target. This represents an exponential function.

4. The fast adaptive iterative high-speed maneuvering target detection method according to claim 3, characterized in that, The preset virtual slow time axis equation is expressed as: ； in, Represents a virtual slow-time variable; The preset speed fuzzy number equation is expressed as: ； in, Indicates the folding factor. Indicates blind speed, , Indicates the radar transmission wavelength. Indicates the pulse repetition frequency. Indicates unambiguous speed. ; The two-dimensional signal is represented as: ； in, It represents a two-dimensional signal.

5. The fast adaptive iterative high-speed maneuvering target detection method according to claim 4, characterized in that, Distance compensation is performed on the two-dimensional signal based on the fuzzy compensation function to obtain the time-domain matrix of the defuzzified signal, including: Multiplying the two-dimensional signal by the ambiguity compensation function yields the time-domain matrix of the deambiguous signal; The fuzzy compensation function is expressed as follows: ； in, Represents the fuzzy compensation function. Represents fuzzy number variables.

6. The fast adaptive iterative high-speed maneuvering target detection method according to claim 1, characterized in that, The first sub-orientation matrix is ​​represented as: ； in, This represents the first sub-orientation matrix. This indicates the signal amplitude after pulse compression processing. Represents the window function. Indicates slow time. Indicates the pulse duration. Indicates the centroid of the Doppler frequency. Indicates Doppler frequency modulation. This represents the centroid of the Doppler frequency compensated by the Doppler compensation function. This represents the Doppler modulation frequency compensated by the Doppler compensation function.

7. A high-speed maneuvering target detection device with rapid adaptive iteration, characterized in that, The fast adaptive iterative high-speed maneuvering target detection device includes: an acquisition unit, a pulse compression processing unit, a range compensation unit, a construction unit, a Doppler compensation unit, and a calculation unit; The acquisition unit is used to: acquire radar echo data; the radar echo data is two-dimensional data; The pulse compression processing unit is used to: sequentially perform down-conversion and pulse compression processing on the radar echo data to obtain a frequency domain echo signal; The acquisition unit is further configured to: acquire a two-dimensional signal by means of a preset virtual slow time axis equation, a preset velocity ambiguity number equation, and the frequency domain echo signal; The distance compensation unit is used to: perform distance compensation on the two-dimensional signal based on the fuzzy compensation function to obtain the time-domain matrix of the defuzzy signal; The construction unit is used to: construct a Doppler compensation function based on the linear frequency modulation characteristics of the time-domain matrix of the deambiguous signal; and randomly select some points in the two-dimensional signal as initial points; The Doppler compensation unit is used to: perform Doppler compensation and one-dimensional constant false alarm detection on the time-domain matrix of the unambiguous signal based on the initial point, the Doppler compensation function and the coordinate ascending search algorithm, to obtain the final reference point; The calculation unit is used to: calculate the target signal velocity and the target signal acceleration based on the final reference point, and use the target signal velocity and the target signal acceleration as the target detection result; The Doppler compensation function is expressed as follows: ； Represents the Doppler compensation function. Represents the window function. Indicates the pulse duration. Indicates the centroid of the Doppler frequency. Indicates the Doppler modulation frequency; The process of performing Doppler compensation and one-dimensional constant false alarm rate (CFAR) detection on the time-domain matrix of the unambiguous signal based on the initial point, the Doppler compensation function, and the coordinate ascent search algorithm to obtain the final reference point includes: S201. Extract the azimuth signal of the same distance unit from the time domain matrix of the defuzzified signal according to the preset rules to obtain multiple sub-azimuth matrices; S202. The multiple sub-azimuth matrices are compensated according to the Doppler compensation function to obtain the first sub-azimuth matrix; S203. Perform one-dimensional CA-CFAR detection on the first sub-azimuth matrix based on the initial point, and take the target point that is greater than the first preset threshold value as the first current reference point; S204. Perform one-dimensional CA-CFAR detection on the first sub-azimuth matrix based on the first current reference point, and take the target point that is greater than the second preset threshold value as the second current reference point. S205. The first current reference point and the second current reference point are combined to form an intermediate reference point; S206. Using the intermediate reference point as the initial point, repeat S203-S205 until the coordinate difference between the intermediate reference point in the current iteration and the intermediate reference point in the previous iteration is less than a preset coordinate change threshold. S207. Take the intermediate reference point corresponding to the most recent execution of S205 as the final reference point.

8. A high-speed maneuvering target detection device with rapid adaptive iteration, characterized in that, include: The device includes a processor, a storage medium, and a bus. The storage medium stores machine-readable instructions executable by the processor. When the fast adaptive iterative high-speed maneuvering target detection device is running, the processor communicates with the storage medium via the bus, and the processor executes the machine-readable instructions to perform the steps of the fast adaptive iterative high-speed maneuvering target detection method as described in any one of claims 1-6.

Images