Maneuvering target detection parameter estimation method and device based on non-uniform sampling technology

Through non-uniform sampling technology and phase compensation function, the problem of low signal-to-noise ratio caused by distance movement and Doppler movement in radar detection is solved, and the rapid detection and efficient parameter estimation of maneuverable targets are achieved, which is suitable for low signal-to-noise ratio environments .

CN119044946APending Publication Date: 2024-11-29SHANGHAI SPACEFLIGHT ELECTRONICS & COMM EQUIP RES INST

Patent Information

Application Number
CN202411166350.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-23
Publication Date
2024-11-29

AI Technical Summary

Technical Problem

The existing radar technology is difficult to achieve rapid detection and efficient parameter estimation due to distance movement and Doppler frequency movement when detecting maneuvering targets.

Method used

Using a method based on non-uniform sampling technology, the third-order phase function and symmetric autocorrelation function are corrected by frequency conversion, pulse compression, Keystone transform, inverse fast Fourier transform, non-uniform sampling, and the phase of the target energy is eliminated. The target exists through constant false alarm rate detection.

Benefits of technology

It improves the signal-to-noise ratio and calculation efficiency of maneuverable target detection, can better adapt to low signal-to-noise ratio environment, realizes fast detection and efficient parameter estimation, and reduces the computational complexity.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119044946A_ABST
    Figure CN119044946A_ABST
Patent Text Reader

Abstract

The invention discloses a maneuvering target detection and parameter estimation method based on a non-uniform sampling technology, which is applied to radar target detection. The method mainly comprises the following steps of: correcting range walk of a pulse pressure echo signal by adopting Keystone transformation; extracting data of a distance unit where the target is located, and estimating a third-order acceleration and a second-order acceleration of the target by adopting a non-uniform sampling correction third-order phase function; on the basis, a quartic term and a cubic term of a slow time dimension of the echo signal by the phase compensation function are constructed; the first-order acceleration of a target is estimated through a non-uniform sampling symmetric autocorrelation function, a corresponding phase compensation function is constructed to complete Doppler walk correction on a quadratic term of a slow time dimension of an echo signal, and finally coherent accumulation of target echoes is realized through Fourier transform. According to the method, the framework of the multi-linear function is expanded, the nonlinearity of the multi-linear function is reduced, the radar echo accumulation signal-to-noise ratio is effectively improved, the multi-dimensional grid search process is avoided, and the calculation complexity is low.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of radar signal processing, and in particular relates to a method and device for maneuvering target detection and parameter estimation based on non-uniform sampling technology. Background Art

[0002] Maneuvering target detection has attracted more and more attention due to its demand in practical applications. These maneuvering targets have typical characteristics such as small radar scattering cross-section and high maneuverability. In order to achieve effective detection of maneuvering targets, it is necessary to use a long-term accumulation technology that exchanges time for energy to accumulate the energy of the echo data. Traditional radar detection methods face major challenges such as distance movement and Doppler frequency movement caused by too long observation time. How to achieve rapid detection of maneuvering targets with existing radars and improve the detection signal-to-noise ratio in response to the above problems is an urgent problem to be solved in the current radar detection field.

[0003] TongXin Dang proposed a method of non-uniform sampling third-order phase function (NUCPF) in his paper A Fast Two-Dimensional Velocity Estimation Method for Multi-Channel UWB SAR. This method can estimate parameters without a search process and has low computational cost. However, this method only selects two slow time dimension slices and determines the target parameter value by the peak position of the two time slices. Due to the lack of the energy accumulation process of the slow time dimension, the target detection performance is poor in low signal-to-noise ratio scenarios.

[0004] Patent application number 202311869685.9, a long-term coherent accumulation detection method for variable acceleration moving targets, proposes a method of de-linear tuning and integrating third-order phase functions, which can effectively suppress the influence of cross terms and quickly achieve coherent accumulation of energy of variable acceleration moving targets. However, this method cannot effectively handle Doppler movement caused by third-order acceleration. Summary of the invention

[0005] The purpose of the present invention is to provide a method for maneuvering target detection and parameter estimation based on non-uniform sampling technology, so as to solve the problem of low accumulated signal-to-noise ratio caused by range movement and Doppler movement in maneuvering target detection in the prior art, and to obtain better detection performance and higher computational efficiency.

[0006] To solve the above problems, the technical solution of the present invention is:

[0007] A method for detecting and estimating maneuvering targets based on non-uniform sampling technology, comprising:

[0008] In a coherent processing interval, the radar echo signal is down-converted and pulse compressed to obtain a radar frequency domain pulse compression signal, and the radar frequency domain pulse compression signal is subjected to Keystone transformation to correct the linear distance movement caused by speed.

[0009] Perform inverse fast Fourier transform on the corrected signal along the distance frequency, extract the phase signal of the distance unit where the target is located, and use the method of sampling the non-uniform sampling to correct the third-order phase function to extract the energy of the phase signal and focus it to form a peak; estimate the third-order acceleration value and second-order acceleration value of the target based on the location of the peak;

[0010] The phase compensation function is constructed by using the estimated values ​​of the target's third-order acceleration and second-order acceleration to compensate for the fourth-order and third-order terms of the echo slow-time dimension, eliminate the third-order and fourth-order Doppler movement, and obtain the compensated slow-time secondary signal.

[0011] The compensated slow-time secondary signal is subjected to non-uniform sampling symmetric autocorrelation function method to focus the signal energy into a peak, and the first-order acceleration value of the target is estimated according to the peak position.

[0012] The phase compensation function is constructed by using the target first-order acceleration estimation value to compensate the quadratic term of the echo slow time dimension, eliminate the second-order Doppler movement, and obtain the compensated slow time single frequency signal.

[0013] The slow-time single-frequency signal that has completed the Doppler movement correction is transformed by FFT along the slow time to complete the coherent accumulation of the target energy, and the existence of the target is confirmed through constant false alarm rate detection.

[0014] Among them, the corrected signal is subjected to inverse fast Fourier transform along the distance frequency, the phase signal of the distance unit where the target is located is extracted, and the method of sampling the non-uniform sampling correction third-order phase function is used to extract the energy of the phase signal and focus to form a peak; estimating the third-order acceleration value and the second-order acceleration value of the target according to the position of the peak value further includes:

[0015] After completing the distance correction, the signal S1(f,t n ) Perform IFFT transformation along the range frequency to obtain a two-dimensional time domain signal represented as:

[0016]

[0017] Where λ is the wavelength, t n is the slow time series after KT, A is the complex amplitude of the signal, t n =mT r ,m=1,2,...,M, is a slow time series, M is the number of accumulated pulses in a coherent processing interval, f cis the carrier frequency, B is the bandwidth, r0 is the initial radial distance, a1 is the radial velocity, a2 is the radial first-order acceleration, a3 is the radial second-order acceleration, and a4 is the radial third-order acceleration;

[0018] Extract the phase signal within the distance unit where the target is located, and get the slow time dimension phase signal expressed as:

[0019]

[0020] The non-uniform sampling corrected third-order phase function is expressed as:

[0021]

[0022] Among them, τ is the time delay sequence, Ω is the frequency sequence, w0 is the time delay constant, * is the complex conjugate operation,

[0023]

[0024] For R(t n ,Ω) modulus value, and get the third-order phase function that discards irrelevant phase terms:

[0025] R(t n ,Ω)=|R(t n ,Ω)|

[0026] =A 4 δ[Ω-(6a3w0+24a4w0t n )]

[0027] Where, δ(g) is the Dirac function;

[0028] For R(t n ,Ω) to perform IFFT and obtain a two-dimensional slow-time phase signal:

[0029] r(t n ,τ)=A 4 exp[-j2π(6a3w0+24a4w0t n )τ]

[0030] For r(t n ,τ) along the slow time series t n Perform scale Fourier transform and FFT transform along the time delay sequence τ to obtain a two-dimensional decoupled integral signal:

[0031]

[0032] in, is the frequency sequence corresponding to the slow time;

[0033] According to the peak position of the integrated signal Get an estimate of the third-order acceleration and the estimated value of the second-order acceleration

[0034]

[0035] Using the estimated values ​​of the target's third-order acceleration and second-order acceleration, a phase compensation function is constructed to compensate for the fourth-order and third-order terms of the echo slow time dimension. Eliminating the third-order and fourth-order Doppler motion further includes:

[0036] The phase compensation function is expressed as:

[0037]

[0038] According to the phase compensation function, the phase compensation function is combined with the distance-corrected time-domain pulse pressure signal s1(τ,t n ) are multiplied to obtain the time domain pulse pressure signal after the first compensation:

[0039]

[0040] The slow-time phase signal in the distance unit where the target is located after the first compensation is expressed as:

[0041]

[0042] According to the compensated phase signal, obtaining the target first-order acceleration estimation value by performing non-uniform sampling symmetric autocorrelation function on the phase signal further includes:

[0043] The non-uniform sampling symmetric autocorrelation function is expressed as:

[0044]

[0045] Q(t n ,τ) along the slow time series t n Perform SFT and FFT along the time delay sequence τ to obtain the two-dimensional linear frequency modulation integrated signal:

[0046]

[0047] According to the peak position of the integrated signal Get an estimate of the first-order acceleration

[0048] Using the estimated value of the first-order acceleration of the target, constructing a phase compensation function and multiplying it with the time-domain pulse pressure signal after the first compensation, eliminating the second-order Doppler movement further includes:

[0049] The phase compensation function is expressed as:

[0050]

[0051] According to the phase compensation function, the phase compensation function is combined with the time domain pulse pressure signal s2(τ,t n ) after multiplication, the time domain pulse pressure signal after the second compensation is:

[0052]

[0053] The slow-time single-frequency signal that has completed the Doppler movement correction is transformed by FFT along the slow time to complete the coherent accumulation of the target energy. The coherent accumulation result is expressed as:

[0054]

[0055] Perform CFAR detection on the coherent accumulation results to confirm the target.

[0056] Based on the same concept, the present invention also provides a mobile target detection and parameter estimation device based on non-uniform sampling technology, comprising:

[0057] The distance correction module is used to perform down-conversion and pulse compression on the radar echo signal within a coherent processing interval to obtain a radar frequency domain pulse compression signal, and perform Keystone transformation on the radar frequency domain pulse compression signal to correct the linear distance movement caused by speed;

[0058] The first target estimation module is used to perform inverse fast Fourier transform on the corrected signal along the distance frequency, extract the phase signal of the distance unit where the target is located, and sample the method of non-uniform sampling to correct the third-order phase function, extract the energy of the phase signal, and focus it to form a peak; estimate the third-order acceleration value and the second-order acceleration value of the target according to the location of the peak value;

[0059] The first compensation module is used to construct a phase compensation function using the estimated values ​​of the target third-order acceleration and second-order acceleration, compensate the fourth-order term and the third-order term of the echo slow time dimension, eliminate the third-order and fourth-order Doppler movement, and obtain the compensated slow-time secondary signal;

[0060] The target second estimation module is used to focus the signal energy to form a peak through the non-uniform sampling symmetric autocorrelation function method of the compensated slow-time secondary signal, and estimate the target first-order acceleration value according to the location of the peak;

[0061] The second compensation module is used to construct a phase compensation function using the target first-order acceleration estimation value, compensate the quadratic term of the echo slow time dimension, eliminate the second-order Doppler movement, and obtain a compensated slow time single frequency signal;

[0062] The coherent accumulation module is used to perform FFT transformation along the slow time on the slow-time single-frequency signal that has completed the Doppler movement correction, complete the coherent accumulation of the target energy, and confirm the existence of the target through constant false alarm rate detection.

[0063] Due to the adoption of the above technical solution, the present invention has the following advantages and positive effects compared with the prior art:

[0064] 1) The method for maneuvering target detection and parameter estimation based on non-uniform sampling technology in one embodiment of the present invention utilizes non-uniform resampling technology to cleverly design the kernel functions of two third-order phase functions and symmetric autocorrelation functions, and fully utilizes the phase information of the two dimensions of slow time and delayed time series to complete coherent accumulation, thereby overcoming the energy loss caused by selecting only a single time slice or a high order of correlation function in the prior art, and the low detection probability of maneuvering targets. The present invention can eliminate the Doppler movement caused by the high-order acceleration of the maneuvering target, complete the coherent accumulation of the maneuvering target echo energy, better adapt to the low signal-to-noise ratio environment, and have better maneuvering target detection performance.

[0065] 2) By performing FFT and IFFT on the third-order phase function and the symmetric autocorrelation function matrix data to integrate the target energy, the problem of the prior art needing to search for motion parameters, resulting in high computational complexity and slow detection speed of maneuvering targets, is overcome. The present invention avoids the search for motion parameters, reduces computational complexity, improves the radar's detection speed of maneuvering targets, and is more conducive to real-time applications in engineering. In addition, the framework of the multilinear function kernel is further expanded, so that the sampling samples of the multilinear function kernel are no longer limited to the known sample points of the original signal sequence, and can further utilize the new sampling points obtained by resampling at any position of the original signal time series. The processing method is more flexible, and the nonlinearity of the multilinear function kernel is reduced, alleviating the signal-to-noise ratio loss caused by nonlinear operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 A flow chart of a method for detecting and estimating maneuvering targets and parameters based on non-uniform sampling technology in one embodiment of the present invention;

[0067] Figure 2 Schematic diagram of experimental results of simulation experiment 1 in one embodiment of the present invention;

[0068] Figure 3 Schematic diagram of experimental results of simulation experiment 2 in one embodiment of the present invention;

[0069] Figure 4 It is a block diagram of a mobile target detection and parameter estimation device based on non-uniform sampling technology in one embodiment of the present invention. DETAILED DESCRIPTION

[0070] The following is a further detailed description of a method and device for detecting and estimating maneuvering targets based on non-uniform sampling technology and parameters proposed by the present invention in conjunction with the accompanying drawings and specific embodiments. The advantages and features of the present invention will become more apparent from the following description and claims.

[0071] This embodiment provides a target detection and parameter estimation method based on non-uniform sampling technology, which eliminates range migration and Doppler frequency spread and completes maneuvering target detection and target parameter estimation technology.

[0072] Please see Figure 1 , the maneuvering target detection and parameter estimation method based on non-uniform sampling technology includes:

[0073] S1: In a coherent processing interval, the radar echo signal is down-converted and pulse compressed to obtain a radar frequency domain pulse compression signal, and the radar frequency domain pulse compression signal is subjected to Keystone transformation to correct the linear distance movement caused by speed;

[0074] S2: Perform inverse fast Fourier transform on the corrected signal along the distance frequency, extract the phase signal of the distance unit where the target is located, and sample the method of non-uniform sampling to correct the third-order phase function, extract the energy of the phase signal, and focus it to form a peak; estimate the third-order acceleration value and second-order acceleration value of the target according to the location of the peak;

[0075] S3: Using the estimated values ​​of the target's third-order acceleration and second-order acceleration, a phase compensation function is constructed to compensate for the fourth-order and third-order terms of the echo slow-time dimension, eliminate the third-order and fourth-order Doppler movement, and obtain the compensated slow-time secondary signal;

[0076] S4: The compensated slow-time secondary signal is subjected to a non-uniform sampling symmetric autocorrelation function method to focus the signal energy to form a peak, and the first-order acceleration value of the target is estimated according to the location of the peak;

[0077] S5: Using the estimated value of the first-order acceleration of the target, a phase compensation function is constructed to compensate for the quadratic term of the slow-time dimension of the echo, eliminate the second-order Doppler movement, and obtain the compensated slow-time single-frequency signal;

[0078] S6: Perform FFT transformation along the slow time for the slow-time single-frequency signal that has completed the Doppler movement correction, complete the coherent accumulation of the target energy, and confirm the existence of the target through constant false alarm rate detection.

[0079] Specifically, in step S1, it is assumed that the broadband radar transmits a set of linear frequency modulation signals:

[0080] s t (τ,t m ) = rect(τ / T p )exp(jπγτ2 )exp[j2πf c (t m +τ)]

[0081] Where rect(g) represents the gate function, γ = B / T p represents the modulation frequency, B is the signal bandwidth, T p represents the pulse duration, f c is the carrier frequency, τ represents the fast time, that is, the distance time, t m =mT r ,m=1,2,...,M represents slow time, T r represents the pulse repetition period, and M represents the number of accumulated pulses.

[0082] Assuming that there is a target in the radar detection area, the baseband echo signal of a single scattering point received by the radar within a coherent processing interval is as follows:

[0083]

[0084] Where A is the complex amplitude, c is the speed of light, R(t m ) is the instantaneous radial distance of the target relative to the radar, and its specific form is as follows:

[0085]

[0086] Where r0 is the initial radial distance of the target, a1 is the target radial velocity, a2 is the target radial first-order acceleration, a3 is the target radial second-order acceleration, and a4 is the target radial third-order acceleration.

[0087] The baseband echo signal is down-converted and pulse compressed to obtain the frequency domain pulse compression signal as follows:

[0088]

[0089] Perform KT on the frequency domain pulse pressure signal, which is defined as follows:

[0090]

[0091] In the formula, t n It is the slow time series recalibrated after KT.

[0092] Under narrowband conditions, since f is much smaller than f c , then f / (f+f c )≈f / f c , then the frequency domain KT correction signal can be approximately expressed as:

[0093]

[0094] Step S2: Through the above method, the linear distance movement caused by the target speed has been eliminated. Perform IFFT along the fast time frequency of equation (19), and the target energy is concentrated in a distance unit:

[0095]

[0096] After eliminating the linear distance movement, the signal whose energy is concentrated in the same distance unit is extracted:

[0097]

[0098] The third-order phase function of the extracted signal can be expressed as follows through non-uniform sampling correction:

[0099]

[0100] Among them, τ is the time delay sequence, Ω is the frequency sequence, w0 is the time delay constant, * is the complex conjugate operation,

[0101]

[0102] For R(t n ,Ω) modulus value, and get the third-order phase function that discards irrelevant phase terms:

[0103] R(t n ,Ω)=|R(t n ,Ω)|

[0104] =A 4 δ[Ω-(6a3w0+24a4w0t n )]

[0105] Where δ(g) is the Dirac function.

[0106] For R(t n ,Ω) to perform IFFT and obtain a two-dimensional slow-time phase signal:

[0107] r(t n ,τ)=A 4 exp[-j2π(6a3w0+24a4w0t n )τ]

[0108] For r(t n ,τ) along the slow time series t n Perform SFT and FFT along the time delay sequence τ to obtain a two-dimensional decoupled integral signal:

[0109]

[0110] in, is the frequency sequence corresponding to slow time.

[0111] According to the peak position of the integrated signal Get an estimate of the third-order acceleration and the estimated value of the second-order acceleration

[0112]

[0113] It should be noted that in this embodiment, the kernel function in the non-uniform sampling modified third-order phase function

[0114]

[0115] The delay index values ​​in are non-integer values. The sampling of these non-integer index values ​​can be implemented as follows:

[0116] (1) Using traditional linear interpolation technology, the original phase signal s(t n ) is interpolated by 4 or 8 times to obtain the interpolated phase signal s′(t n ′), where the interpolated slow time series is t n ′;

[0117] (2) Calculate the sample value s(t0) at any time point t0:

[0118]

[0119] in, Indicates t that is greater than or equal to t0 and closest to t0 n ′, Indicates t that is less than or equal to t0 and closest to t0 n ′ value.

[0120] Specifically, in the non-uniform sampling corrected third-order phase function of this embodiment, the values ​​of t0 are respectively

[0121] Specifically, considering the parameter estimation accuracy and the cross-term suppression performance, a suitable value of w0 is w0=round(0.089·M).

[0122] Step S3: Construct the following phase compensation function according to the estimation result of step S2:

[0123]

[0124] Then, the phase compensation function is combined with the distance-corrected time-domain pulse pressure signal s1(τ,t n ) are multiplied to obtain the time domain pulse pressure signal after the first compensation:

[0125]

[0126] The slow time phase signal in the distance unit where the target is located after the first compensation can be expressed as:

[0127]

[0128] Step S4: The extracted secondary phase signal is subjected to non-uniform sampling symmetric autocorrelation function to obtain:

[0129]

[0130] Q(t n ,τ) along the slow time series t n Perform SFT and FFT along the time delay sequence τ to obtain the two-dimensional linear frequency modulation integrated signal:

[0131]

[0132] According to the peak position of the integrated signal The estimated value of the first-order acceleration can be obtained

[0133]

[0134] It should be noted that in this embodiment, the kernel function in the non-uniform sampling symmetric autocorrelation function

[0135]

[0136] The delay index values ​​in are non-integer. These non-integer sampling can be implemented as follows:

[0137] (1) Using traditional linear interpolation technology, the original phase signal s(t n ) is interpolated by 4 or 8 times to obtain the interpolated phase signal s′(t n ′), where the interpolated slow time series is t n ′;

[0138] (2) Calculate the sample value s(t0) at any time point t0:

[0139]

[0140] in, Indicates t that is greater than or equal to t0 and closest to t0 n ′, Indicates t that is less than or equal to t0 and closest to t0 n ′ value.

[0141] Specifically, in the non-uniform sampling symmetric autocorrelation function of this embodiment, the values ​​of t0 are respectively

[0142] Step S5: According to the estimated value of step S4, construct the following phase compensation function:

[0143]

[0144] The phase compensation function is combined with the time domain pulse compression signal s2(τ,t n ) is multiplied to obtain the time domain pulse pressure signal after the second compensation:

[0145]

[0146] Step S6: Perform FFT on the time domain pulse pressure signal after the second compensation along the slow time to complete the accumulation:

[0147]

[0148] Finally, perform CFAR detection on the accumulated results. Use the following threshold formula to calculate the detection threshold:

[0149]

[0150] Among them, η represents the detection threshold, ln represents the logarithm with the natural constant e as the base, P fa represents the false alarm probability given by the system, and σ represents the system noise variance.

[0151] The maximum value is taken from the absolute value of each element in the coherent accumulation signal to determine whether the maximum value is greater than or equal to the detection threshold. If so, it is determined that a maneuvering target exists; otherwise, it is determined that a maneuvering target does not exist.

[0152] Simulation experiment

[0153] The present invention is further described below in conjunction with simulation experiments.

[0154] (1) The simulation experiment conditions are as follows:

[0155] The operating system platform of the simulation experiment of the present invention is: MacOS10.15 operating system;

[0156] The simulation software for the simulation experiment of the present invention is: MATLAB 2021b.

[0157] The radar system parameters used in the simulation experiment of the present invention are shown in Table 1:

[0158] Table 1 Radar system parameters

[0159]

[0160] The target parameters used in the simulation experiment of the present invention are shown in Table 2:

[0161] Table 2 Target parameters

[0162] Initial radial distance Radial velocity Radial first-order acceleration Radial second-order acceleration Radial third-order acceleration 50km 250m / s <![CDATA[20m / s 2 ]]> <![CDATA[10m / s 3 ]]> <![CDATA[10m / s 4 ]]>

[0163] (2) Simulation content and result analysis

[0164] Simulation experiment 1:

[0165] The simulation experiment 1 of the present invention uses the method of the present invention and four existing technologies (generalized Radon Fourier transform method, non-uniform sampling variable scale processing method, non-uniform sampling third-order phase function method, non-uniform sampling reduction operation) to perform 500 Monte-Carlo simulations on each signal-to-noise ratio in the signal-to-noise ratio interval of [-50, 10] dB. Assume P fa =10 -6 The simulation results are plotted as a comparison chart of the maneuvering target detection probability versus signal-to-noise ratio. Figure 2 shown.

[0166] The four existing technologies in the simulation experiment are:

[0167] The prior art generalized Radon Fourier transform method refers to a maneuvering target detection method proposed by J.Xu et al. in the paper “Radar maneuvering target motion estimation based on generalized radon-Fourier transform”, which is referred to as the generalized Radon Fourier transform method.

[0168] The prior art non-uniform sampling variable scale processing method refers to the maneuvering target detection method proposed by XL Chen et al. in the paper “Sparse long-time coherent integration-based detection method for radar low-observable manoeuvring target”, which is referred to as the non-uniform sampling variable scale processing method.

[0169] The prior art non-uniform sampling third-order phase function method refers to the maneuvering target detection method proposed by TX Dang et al. in "A Fast Two-Dimensional Velocity Estimation Method for Multi-Channel UWB SAR", which is referred to as the non-uniform sampling third-order phase function method.

[0170] The prior art non-uniform sampling reduced-order operation method refers to the maneuvering target detection method proposed by RQCao et al. in “A new method for parameter estimation of high-order polynomial-phase signals”, referred to as the non-uniform sampling reduced-order operation method.

[0171] Figure 2 The horizontal axis is the signal-to-noise ratio, in dB, and the vertical axis is the detection probability. Figure 2 It can be seen that under the same signal-to-noise ratio value, the target detection probability of the present invention is higher than the non-uniform sampling variable scale processing method, the non-uniform sampling third-order phase function method and the non-uniform sampling reduction operation in the prior art, and is slightly lower than the generalized Radon Fourier transform method in the prior art. This is because the generalized Radon Fourier transform method adopts a multi-dimensional grid search method, has no signal-to-noise ratio loss caused by nonlinear operation, and is the optimal Doppler filter group under the background of Gaussian white noise.

[0172] Simulation experiment 2:

[0173] The simulation experiment 2 of the present invention uses the method of the present invention and four existing technologies (generalized Radon Fourier transform method, non-uniform sampling variable scale processing method, non-uniform sampling third-order phase function method, non-uniform sampling reduction operation) to calculate the computational complexity of each accumulated pulse number in a coherent processing interval with an accumulated pulse number interval of [50:50:3000], and plots the simulation results into a curve of computational complexity versus accumulated pulse number as shown in FIG. Figure 3 As shown. Figure 3 It can be seen that the computational complexity of the method of the present invention is lower than that of the generalized Radon Fourier transform method and the non-uniform sampling variable scale processing method, and is almost consistent with the non-uniform sampling order reduction operation, but slightly higher than the non-uniform sampling third-order phase function method. This is because the non-uniform sampling third-order phase function method directly selects two slow-time dimension slices for parameter estimation, has no accumulation process, and the algorithm flow is relatively simple.

[0174] In summary, the maneuvering target detection and parameter estimation method based on non-uniform sampling technology provided by the present invention can achieve a good balance between computational complexity and detection probability, and is a very practical radar maneuvering target detection and parameter estimation method.

[0175] Based on the same concept, the present invention also provides a mobile target detection and parameter estimation device based on non-uniform sampling technology, please refer to Figure 4 , the device comprises:

[0176] The distance correction module is used to perform down-conversion and pulse compression on the radar echo signal within a coherent processing interval to obtain a radar frequency domain pulse compression signal, and perform Keystone transformation on the radar frequency domain pulse compression signal to correct the linear distance movement caused by speed;

[0177] The first target estimation module is used to perform inverse fast Fourier transform on the corrected signal along the distance frequency, extract the phase signal of the distance unit where the target is located, and sample the method of non-uniform sampling to correct the third-order phase function, extract the energy of the phase signal, and focus it to form a peak; estimate the third-order acceleration value and the second-order acceleration value of the target according to the location of the peak value;

[0178] The first compensation module is used to construct a phase compensation function using the estimated values ​​of the target third-order acceleration and second-order acceleration, compensate the fourth-order term and the third-order term of the echo slow time dimension, eliminate the third-order and fourth-order Doppler movement, and obtain the compensated slow-time secondary signal;

[0179] The target second estimation module is used to focus the signal energy to form a peak through the non-uniform sampling symmetric autocorrelation function method of the compensated slow-time secondary signal, and estimate the target first-order acceleration value according to the location of the peak;

[0180] The second compensation module is used to construct a phase compensation function using the target first-order acceleration estimation value, compensate the quadratic term of the echo slow time dimension, eliminate the second-order Doppler movement, and obtain a compensated slow time single frequency signal;

[0181] The coherent accumulation module is used to perform FFT transformation along the slow time on the slow-time single-frequency signal that has completed the Doppler movement correction, complete the coherent accumulation of the target energy, and confirm the existence of the target through constant false alarm rate detection.

[0182] The above-mentioned device is used to implement the steps in the method of maneuvering target detection and parameter estimation based on non-uniform sampling technology. Its functions and implementation methods are similar and will not be repeated here.

[0183] The embodiments of the present invention are described in detail above in conjunction with the accompanying drawings, but the present invention is not limited to the above embodiments. Even if various changes are made to the present invention, if these changes fall within the scope of the claims of the present invention and their equivalents, they still fall within the protection scope of the present invention.

Claims

1. A method for detecting and estimating maneuvering targets based on non-uniform sampling technology, characterized in that: include: In a coherent processing interval, the radar echo signal is down-converted and pulse compressed to obtain a radar frequency domain pulse compression signal, and the radar frequency domain pulse compression signal is subjected to Keystone transformation to correct the linear distance movement caused by speed. Perform inverse fast Fourier transform on the corrected signal along the distance frequency, extract the phase signal of the distance unit where the target is located, and use the method of sampling the non-uniform sampling to correct the third-order phase function to extract the energy of the phase signal and focus it to form a peak; estimate the third-order acceleration value and second-order acceleration value of the target based on the location of the peak; The phase compensation function is constructed by using the estimated values ​​of the target's third-order acceleration and second-order acceleration to compensate for the fourth-order and third-order terms of the echo slow-time dimension, eliminate the third-order and fourth-order Doppler movement, and obtain the compensated slow-time secondary signal. The compensated slow-time secondary signal is subjected to non-uniform sampling symmetric autocorrelation function method to focus the signal energy into a peak, and the first-order acceleration value of the target is estimated according to the peak position. The phase compensation function is constructed by using the target first-order acceleration estimation value to compensate the quadratic term of the echo slow time dimension, eliminate the second-order Doppler movement, and obtain the compensated slow time single frequency signal. The slow-time single-frequency signal that has completed the Doppler movement correction is transformed by FFT along the slow time to complete the coherent accumulation of the target energy, and the existence of the target is confirmed through constant false alarm rate detection.

2. The method for detecting and estimating maneuvering targets based on non-uniform sampling technology according to claim 1, characterized in that: Perform inverse fast Fourier transform on the corrected signal along the distance frequency to extract the phase signal of the distance unit where the target is located, and use the method of sampling the non-uniform sampling to correct the third-order phase function to extract the energy of the phase signal and focus it to form a peak; Estimating the target third-order acceleration value and the second-order acceleration value according to the peak position further includes: After completing the distance correction, the signal S1(f,t n ) Perform IFFT transformation along the range frequency to obtain a two-dimensional time domain signal represented as: Where λ is the wavelength, A is the complex amplitude of the signal, and t n is the slow time series after KT transformation, t n =mT r ,m=1,2,...,M, M is the number of accumulated pulses in a coherent processing interval, T r represents the pulse repetition period, f c is the carrier frequency, B is the bandwidth, r0 is the initial radial distance, a1 is the radial velocity, a2 is the radial first-order acceleration, a3 is the radial second-order acceleration, and a4 is the radial third-order acceleration; Extract the phase signal within the distance unit where the target is located, and get the slow time dimension phase signal expressed as: The non-uniform sampling corrected third-order phase function is expressed as: Among them, τ is the time delay sequence, Ω is the frequency sequence, w0 is the time delay constant, * is the complex conjugate operation, For R(t n ,Ω) modulus value, and get the third-order phase function that discards irrelevant phase terms: R(t n ,Ω)=|R(t n ,Oh)| =A 4 δ[Ω-(6a3w0+24a4w0t n )] Where, δ(g) is the Dirac function; For R(t n ,Ω) to perform IFFT and obtain a two-dimensional slow-time phase signal: r(t n ,τ)=A 4 exp[-j2π(6a3w0+24a4w0t n )t] For r(t n ,τ) along the slow time series t n Perform scale Fourier transform and FFT transform along the time delay sequence τ to obtain a two-dimensional decoupled integral signal: Among them, f tn is the frequency sequence corresponding to the slow time; According to the peak position of the integrated signal Get an estimate of the third-order acceleration and the estimated value of the second-order acceleration 3. The method for detecting and estimating maneuvering targets based on non-uniform sampling technology according to claim 1, characterized in that: Using the estimated values ​​of the target's third-order acceleration and second-order acceleration, a phase compensation function is constructed to compensate for the fourth-order and third-order terms of the echo slow time dimension. Eliminating the third-order and fourth-order Doppler motion further includes: The phase compensation function is expressed as: According to the phase compensation function, the phase compensation function is combined with the distance-corrected time-domain pulse pressure signal s1(τ,t n ) are multiplied to obtain the time domain pulse pressure signal after the first compensation: The slow-time phase signal in the distance unit where the target is located after the first compensation is expressed as:

4. The method for detecting and estimating maneuvering targets based on non-uniform sampling technology according to claim 1, characterized in that: According to the compensated phase signal, obtaining the target first-order acceleration estimation value by performing non-uniform sampling symmetric autocorrelation function on the phase signal further includes: The non-uniform sampling symmetric autocorrelation function is expressed as: Q(t n ,τ) along the slow time series t n Perform SFT and FFT along the time delay sequence τ to obtain the two-dimensional linear frequency modulation integrated signal: According to the peak position of the integrated signal Get an estimate of the first-order acceleration 5. The method for maneuvering target detection and parameter estimation based on non-uniform sampling technology as claimed in claim 1, characterized in that: Using the estimated value of the first-order acceleration of the target, constructing a phase compensation function and multiplying it with the time-domain pulse pressure signal after the first compensation, eliminating the second-order Doppler movement further includes: The phase compensation function is expressed as: According to the phase compensation function, the phase compensation function is combined with the time domain pulse pressure signal s2(τ,t n ) after multiplication, the time domain pulse pressure signal after the second compensation is:

6. The method for maneuvering target detection and parameter estimation based on non-uniform sampling technology as claimed in claim 1, characterized in that: The slow-time single-frequency signal that has completed the Doppler movement correction is transformed by FFT along the slow time to complete the coherent accumulation of the target energy. The coherent accumulation result is expressed as: Perform CFAR detection on the coherent accumulation results to confirm the target.

7. A mobile target detection and parameter estimation device based on non-uniform sampling technology, characterized in that: include: The distance correction module is used to perform down-conversion and pulse compression on the radar echo signal within a coherent processing interval to obtain a radar frequency domain pulse compression signal, and perform Keystone transformation on the radar frequency domain pulse compression signal to correct the linear distance movement caused by speed; The first target estimation module is used to perform inverse fast Fourier transform on the corrected signal along the distance frequency, extract the phase signal of the distance unit where the target is located, and sample the method of non-uniform sampling to correct the third-order phase function, extract the energy of the phase signal, and focus it to form a peak; estimate the third-order acceleration value and the second-order acceleration value of the target according to the location of the peak value; The first compensation module is used to construct a phase compensation function using the estimated values ​​of the target third-order acceleration and second-order acceleration, compensate the fourth-order term and the third-order term of the echo slow time dimension, eliminate the third-order and fourth-order Doppler movement, and obtain the compensated slow-time secondary signal; The target second estimation module is used to focus the signal energy to form a peak through the non-uniform sampling symmetric autocorrelation function method of the compensated slow-time secondary signal, and estimate the target first-order acceleration value according to the location of the peak; The second compensation module is used to construct a phase compensation function using the target first-order acceleration estimation value, compensate the quadratic term of the echo slow time dimension, eliminate the second-order Doppler movement, and obtain a compensated slow time single frequency signal; The coherent accumulation module is used to perform FFT transformation along the slow time on the slow-time single-frequency signal that has completed the Doppler movement correction, complete the coherent accumulation of the target energy, and confirm the existence of the target through constant false alarm rate detection.

Citation Information

Patent Citations

  • Long-time coherent accumulation detection method for variable acceleration moving target

    CN117826105A

Cited By

  • Signal processing method and system, chip, readable storage medium and program product

    CN119291644A

  • Method and device for estimating third-order motion parameters of target based on high-order phase fusion

    CN121918096A

  • Method and device for estimating third-order motion parameters of a target based on high-order phase fusion

    CN121918096B