A Terahertz Ultra-Low Range Sidelobe Pulse Compression Method Based on Phase Correction

Abstract

The present invention discloses a terahertz ultra-low range sidelobe pulse compression method based on phase correction, comprising the following steps: S1, collecting a reference signal of an internal calibration branch of a system and a target echo signal of a receiver, and extracting the residual phases of the two signals; S2, according to the residual phases of the two signals, convolving the reference signal of the internal calibration branch with the target echo signal of the receiver for matched filtering processing to obtain a matched filtering result; S3, according to the matched filtering result, optimizing and designing an ultra-low range sidelobe filter to obtain an optimal ultra-low range sidelobe filter coefficient; S4, using the optimal ultra-low range sidelobe filter coefficient to process the two signals collected by the system, and further obtaining a processing result of the ultra-low range sidelobe suppression ratio. The present invention proposes a method combining system nonlinear phase suppression and optimal filter design, which can not only adapt to a terahertz radar system but also meet the requirements of weak target detection.

Description

Technical Field

[0001] The present invention relates to the field of radar detection, and particularly to the field of terahertz active detection and imaging technology applicable to weak target signals. Background Art

[0002] With the development of terahertz technology, the application of terahertz radar systems has gradually evolved from laboratory functional verification to target detection in real scenarios. However, due to the relatively high terahertz frequency band, the nonlinear characteristics of the system are usually more obvious, and the phase noise and spurious performance are poor, resulting in the transmitted waveform being amplitude and phase modulated by non-ideal factors of the system. In addition, with the development needs of numerical weather prediction and the application of stealth technology, the requirements for the weak target detection ability of radar systems are getting higher and higher, and the requirements for the weak target detection ability under the conditions of strong clutter interference and strong target interference are also getting higher and higher. Usually, the matched filtering of the radar system suppresses the range sidelobes through a window function. However, since it is difficult for the system to transmit an ideal waveform, the window function often has a much worse suppression effect on the range sidelobes than expected. Therefore, to meet the requirements of improving the weak target detection ability of the radar system, requirements are also put forward for the range sidelobe suppression technology after matched filtering.

[0003] In the current methods, range sidelobe suppression is performed based on the target echo, but it is difficult to play an effective role in the case of strong interference and weak targets; the method of adding window functions at both the radar reflection end and the receiving end can obtain ultra-low sidelobes, but this method causes a large energy loss of the system transmitter and is rarely applied to actual radar systems; the method of using a high-order polynomial curve fitting for the phase to achieve the purpose of non-linear waveform design has a good suppression effect on the range sidelobes using non-linear frequency modulation signals, but the application scenario of this form of waveform is small, mainly because this form of waveform is too sensitive to system errors and noise, and to a certain extent wastes system bandwidth resources, resulting in a higher system complexity.

[0004] Therefore, in view of the requirements such as strong clutter interference and weak target detection, the present invention proposes a method combining system non-linear phase suppression and optimal filtering design, which can not only adapt to terahertz radar systems but also meet the requirements of weak target detection. In addition, the method of the present invention is also suitable for the application scenarios of general radar systems. Summary of the Invention

[0005] When facing requirements such as strong clutter interference and weak target detection, due to the amplitude and phase modulation of the terahertz radar system on low-frequency signals, the performance indicators such as phase noise and spurs of the radar system are much lower than those of low-frequency microwave radar systems, resulting in the performance of conventional range sidelobe compression being difficult to achieve the expected effect.

[0006] To overcome the above problems, the present invention proposes a terahertz ultra-low range sidelobe pulse compression method based on phase correction, including the following steps:

[0007] S1. Collect the reference signal of the internal calibration branch of the radar system and the target echo signal of the receiver, and extract the residual phases of the two signals;

[0008] S2. According to the residual phases of the two signals, convolve the reference signal of the internal calibration branch with the target echo signal of the receiver for calculation, perform matched filtering processing, and obtain a matched filtering result;

[0009] S3. According to the matched filtering result, optimize and design an ultra-low range side lobe filter to obtain the optimal ultra-low range side lobe filter coefficients;

[0010] S4. Use the optimal ultra-low range side lobe filter coefficients to process the two signals collected by the system, and then obtain the processing result of the ultra-low range side lobe suppression ratio.

[0011] Among them, the step S1 further includes the following steps:

[0012] S11. Collect the reference signal of the internal calibration branch and the target echo signal simultaneously, and perform preprocessing such as filtering on the two signals;

[0013] S12. Obtain the phase information of the reference signal and the target echo signal, and calculate the residual phases of the two signals.

[0014] Among them, the step S3 further includes the following steps:

[0015] S31. Convolve the matched filtering result with the ultra-low range side lobe filter coefficients to obtain the data value filtered by this filter;

[0016] S32. Set the main lobe width of the data value filtered in S31, and set the main lobe region of this data value to zero;

[0017] S33. Select the objective function of the optimized filter function, optimize the ultra-low range side lobe filter coefficients, repeat steps S31 to S32, and after multiple iterative processes, obtain the optimal ultra-low range side lobe filter coefficients.

[0018] Among them, the step S31 further includes the following steps:

[0019] S311. Discretize the matched filtering result and the ultra-low range side lobe filter coefficients;

[0020] S312. Convolve the discretized matched filtering result with the ultra-low range side lobe filter coefficients to obtain the data value filtered by the filter.

[0021] Among them, the reference signal of the internal calibration branch is s ref (T), and its expression is where s ref (t) is a reference signal, t is a time variable, γ represents a frequency modulation slope, and f IF is the center frequency after down-conversion of the echo signal, and h(t) represents the coefficient of the ultra-low range sidelobe filter. represents the phase modulation caused by both the internal calibration branch and the transmitter; the target echo signal is s r (t), and its expression is where s r (t) is the echo signal of the radar system. is the phase modulation caused by both the receiver and the transmitter.

[0022] where the residual phase is can be expressed as where imag represents taking the imaginary part of a complex number, and rea] represents taking the real part of a complex number.

[0023] where the matched filtering result is x(t), and its expression is where "*" is the convolution operation.

[0024] where the data value after filtering by the filter is y[n], and its calculation formula is y[n] = x[n] * h[n], where x[n] is the data after discretizing the matched filtering result x(t), h[n] is the data after discretizing the coefficient of the ultra-low range sidelobe filter in continuous time, n is the sequence value, and its maximum value does not exceed the maximum sampling value N, and N is determined according to the actual situation.

[0025] where the optimization objective function is J, and its expression is J = |y′[1]| 2 + |y′[2]| 2 + … + |y′[N]| 2 , where y′[n] is a time series, and y′[N / 2 + 1 - L ml , y′[N / 2 + 1 - L ml + 1], ……, y′[N / 2 + 1 + L ml is 0, where L ml is half of the main lobe width.

[0026] where the specific step S4 is: using the optimal ultra-low range sidelobe filter coefficient in step S3 to suppress the sidelobe of the matched filtering result, and finally obtaining the processing result of the ultra-low range sidelobe suppression ratio.

[0027] In summary, the present invention proposes a method combining system non-linear phase suppression and optimal filter design, which can not only adapt to the terahertz radar system but also meet the requirements of weak target detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a principle block diagram of matched filtering processing based on the internal standard system architecture;

[0029] Figure 2 It is the result diagram of matched filtering of the ideal linear frequency modulation signal with double windows added in the transmitting and receiving time domains;

[0030] Figure 3 It is the result diagram of matched filtering of the system test data by an ultra-low sidelobe filter. Specific implementation manners

[0031] The following will combine with the Figures 1 to 3 in the embodiments of the present invention to elaborate in detail on the technical solutions, structural features, achieved objectives and effects in the embodiments of the present invention.

[0032] It should be noted that the attached drawings adopt a very simplified form and all use non-precise scales, only for conveniently and clearly assisting in explaining the purpose of the embodiments of the present invention, and are not used to limit the limiting conditions for implementing the present invention. Therefore, they do not have technical substance significance. Any modification of the structure, change of the proportional relationship or adjustment of the size, without affecting the effects that the present invention can produce and the objectives that can be achieved, should still fall within the scope covered by the technical content disclosed by the present invention.

[0033] It should be noted that in the present invention, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements clearly listed, but also includes other elements not clearly listed, or further includes elements inherent to such process, method, article or device.

[0034] A terahertz ultra-low range sidelobe pulse compression method based on phase correction includes the following steps: S1. Collect signals of a terahertz radar system and extract the residual phases of the two signals of the receiver and the internal standard branch;

[0035] In a similar terahertz radar multiple frequency multiplication system architecture, it often occurs that the waveform of the transmitted signal is affected by high-frequency devices and it is difficult to transmit a relatively ideal signal waveform. The waveform amplitude and phase of the transmitted signal will be modulated, and the phase modulation effect is more obvious. Taking the linear frequency modulation waveform as an example, in the case of ignoring amplitude modulation, the waveform of the transmitted signal is:

[0036]

[0037] Among them, s t (t) represents the transmitted signal of the system, where t is the time variable, and f c represents the center frequency of the system transmitter, γ represents the frequency modulation slope, represents the phase modulation term caused by the non-ideal factors of the system transmitter. In this formula, it is assumed that the amplitude of the transmitted signal is in an ideal state with an amplitude of 1.

[0038] In a terahertz radar system, the ideal modulation of the transmitted signal is usually selected as the ideal reference signal for matched filtering, and it is assumed that the center frequency of the down-converted echo signal of this system is f IF , then the ideal reference signal is expressed as:

[0039]

[0040] Among them, s refideal (t) represents the ideal reference signal for matched filtering, where t is the time variable, and γ represents the frequency modulation slope. However, based on this s refideal (t) for performing matched filtering on the above-mentioned transmitted signal s t (t), the system phase factor cannot be eliminated, thus affecting the effect of matched filtering.

[0041] Therefore, the present invention uses the signal obtained by the internal calibration system architecture as the target echo reference signal, so as to be able to largely suppress the influence brought by the system phase modulation. The principle of this internal calibration system architecture is as Figure 1 shown. The internal calibration branch obtains the transmitted signal of the transmitter by coupling output from the transmitter, and the receiver obtains the echo signal from the target irradiated by the radar; the transmitted signal obtained by the internal calibration branch is weighted in the time domain or frequency domain by an ultra-low range sidelobe filter and used as a reference signal to perform matched filtering on the echo signal of the target. Step S1 further includes the following steps:

[0042] S11. Simultaneously collect the reference signal and the target echo using a signal collector, and perform preprocessing such as filtering on the two signals;

[0043] In the terahertz radar system of this embodiment, a linear frequency modulation signal is used. The intermediate frequency of the target echo of the receiver and the intermediate frequency of the internal calibration branch are 3.6 MHz, and the frequency modulation bandwidth is 4.8 MHz. In a linear frequency modulation form, after collecting and preprocessing the two signals, the reference signal of the internal calibration branch obtained is

[0044]

[0045] Among them, s ref (t) is the reference signal, h(t) represents the coefficient of the ultra-low range sidelobe filter, represents the phase modulation jointly caused by the internal calibration branch and the transmitter.

[0046] Figure 1 The target echo signal obtained from

[0047]

[0048] where s r (t) is the echo signal of the radar system, is the phase modulation jointly caused by the receiver and the transmitter.

[0049] S12. Obtain the phase information of the two signals preprocessed by the receiver and the internal standard branch respectively, and calculate the residual phase of the two signals

[0050] The matched filtering of the internal standard system architecture has eliminated the phase linear modulation caused by system transmission during the preprocessing. Therefore, only the phase residual after the non-linear matching of the internal standard branch and the receiver needs to be processed. Since the general receiver and the internal standard branch work relatively stably, the waveform behavior difference can be obtained by measurement, and the residual phase can be expressed as

[0051]

[0052] where is the residual phase of the receiver and the internal standard branch, imag represents taking the imaginary part of a complex number, and real represents taking the real part of a complex number.

[0053] The above method can also be obtained through software simulation: Use Matlab software to generate a chirp signal with an intermediate frequency of 3.6 MHz and a frequency modulation bandwidth of 4.8 MHz. The characteristics of the ideal state of the filter can be analyzed. Using the method of adding double Hamming windows in the transmit and receive time domains has the best effect on the ideal chirp signal, and the maximum sidelobe level is 62.26 dB.

[0054] S13. Perform complex conjugate processing on the residual phase obtained in step S12 and the target echo signal to eliminate the influence of the residual phase on the range resolution broadening.

[0055] S2. Perform matched filtering processing on the target echo signal of the receiver with the reference signal of the internal standard branch, and obtain the matched filtering result x(t);

[0056] By convolving s r (t) and s ref (t), a matched filtering result with lower range sidelobes can be obtained. x(t) can be expressed as:

[0057]

[0058] Among them, x(t) represents the result of matched filtering, and "*" is the convolution symbol. is the residual phase between the receiver and the internal standard branch.

[0059] S3. According to the above-mentioned matched filtering result, optimize and design an ultra-low range sidelobe filter, that is, a filter with a range sidelobe suppression ratio better than 60 dB, to obtain the optimal ultra-low range sidelobe filter coefficients. The matched filtering result is as Figure 2 shown;

[0060] The ultra-low range sidelobe filter is an off-line design, and the designed filter can be used online in the terahertz radar system. The design process of this filter is as follows:

[0061] S31. Convolve the above-mentioned matched filtering result with the ultra-low range sidelobe filter to obtain the data value after being filtered by this filter;

[0062] S311. Discretize the matched filtering result and the ultra-low range sidelobe filter respectively;

[0063] Take x(t) as the input function for optimizing and designing the ultra-low range sidelobe filter, and discretize x(t) into x[n]. The x[n] is the input data of the ultra-low range sidelobe filter. n represents the sequence value, and the maximum value of n does not exceed the maximum sampling value N, and N is determined by the actual sampling.

[0064] Similarly, discretize the coefficients of the continuous-time ultra-low range sidelobe filter to obtain h[n], and initialize the ultra-low range sidelobe filter h[n].

[0065] S312. Input the data x[n], and convolve it with the initialized ultra-low range sidelobe filter coefficients h[n] to obtain the data value y[n] after being filtered by the filter. Specifically, y[n] = x[n] * h[n], where "*" is the convolution calculation symbol.

[0066] S32. Set the main lobe width of the y[n], and set the main lobe region of y[n] to zero;

[0067] Define the main lobe width of the y[n] as 2L ml , the center position of the main lobe is N / 2 + 1. Then, in the time series y'[n], y'[N / 2 + 1 - L ml , y'[N / 2 + 1 - L ml + 1],..., y'[N / 2 + 1 + L ml are 0.

[0068] S33. Select the objective function J of the optimized filter function, and obtain the optimal ultra-low range sidelobe filter coefficients h opt [n] through multiple iterations;

[0069] Since the position of the maximum peak value of the sidelobe of y[n] is uncertain, the integrated sidelobe ISL is selected as the optimization objective function J. Among them, the expression of ISL is:

[0070] J = |y′[1]| 2 +|y′[2]| 2 +…+|y′[N]| 2

[0071] Taking ISL as the objective function J for optimization and taking the filter energy h[1] 2 …+h[n] 2 = 1 as the constraint condition. Since the constrained objective function belongs to a quadratic hyperplane function, its derivative with respect to the filter coefficients is calculated and the derivative result is set to zero to determine the calculation equation. It can be expressed as:

[0072] Ah = λh

[0073] In the formula, λ is the eigenvector value of matrix A, where matrix A is the result of the product of the collected data and the main lobe set to zero. After initializing the filter coefficients, the equation is obtained by continuously updating the derivative of the objective function, and the filter coefficients are updated. For example, after initializing the coefficients and solving Ah = λh, the first ultra-low distance sidelobe filter coefficients h1[n] can be obtained. Then, replacing h[n] with h1[n] and repeating the operation of step S3, in this way, each time with a better h x [n] replaces h x-1 [n]. After multiple iterative processes, the optimal ultra-low distance sidelobe filter coefficients h opt [n] can be obtained. These coefficients can be used for sidelobe suppression in signal processing. The number of iterations is usually a preset value. In this embodiment, the number of iterations is set to 10.

[0074] S4. Process the signals collected by the system using the obtained optimal ultra-low distance sidelobe filter coefficients h opt [n];

[0075] Using the optimal ultra-low distance sidelobe filter coefficients h Opt [n] obtained in S3, perform sidelobe suppression on the result x[n] obtained after processing and matched filtering of the two signals collected in real time, and finally obtain the processing result of the ultra-low distance sidelobe suppression ratio, as Figure 3 shown.

[0076] Although the content of the present invention has been described in detail through the above preferred embodiments, it should be recognized that the above description should not be considered as a limitation of the present invention. After those skilled in the art have read the above content, various modifications and substitutions to the present invention will be obvious. Therefore, the protection scope of the present invention should be defined by the appended claims.

Claims

1. A terahertz ultra-low range sidelobe pulse compression method based on phase correction, characterized in that It includes the following steps: S1. Collect the reference signal of the internal calibration branch of the radar system and the target echo signal of the receiver, and extract the residual phases of the two signals, where the two signals are the reference signal of the internal calibration branch and the target echo signal of the receiver; S2. According to the residual phases of the two signals, convolve the reference signal of the internal calibration branch with the target echo signal of the receiver for matched filtering processing to obtain a matched filtering result; S3. According to the matched filtering result, perform multiple iterations to optimize and design an ultra-low range sidelobe filter to obtain the optimal ultra-low range sidelobe filter coefficients; S4. Use the optimal ultra-low range sidelobe filter coefficients to process the two signals collected by the system, and then obtain the processing result of the ultra-low range sidelobe suppression ratio.

2. The terahertz ultra-low range sidelobe pulse compression method based on phase correction according to claim 1, wherein, The step S1 further includes the following steps: S11. Collect the reference signal of the internal calibration branch and the target echo signal of the receiver simultaneously, and perform pre-filtering processing on the two signals; S12. Obtain the phase information of the pre-processed reference signal and the target echo signal, and calculate the residual phases of the two signals.

3. A terahertz ultra-low range sidelobe pulse compression method based on phase correction according to claim 2, characterized in that The step S3 further includes the following steps: S31. Convolve the matched filtering result with the ultra-low range sidelobe filter coefficients to obtain the data value after being filtered by the filter; S32. Set the main lobe width of the data value filtered in S31, and set the main lobe region of this data value to zero; S33. Select the objective function for optimizing the filter function, optimize the ultra-low range sidelobe filter coefficients, repeat steps S31 - S32, and after multiple iterative processes, obtain the optimal ultra-low range sidelobe filter coefficients.

4. The terahertz ultra-low range sidelobe pulse compression method based on phase correction according to claim 3, wherein The step S31 further includes the following steps: S311. Discretize the matched filtering result and the ultra-low range sidelobe filter coefficients; S312. Convolve the discretized matched filtering result with the ultra-low range sidelobe filter coefficients to obtain the data value after being filtered by the filter.

5. A terahertz ultra-low range sidelobe pulse compression method based on phase correction according to claim 3, characterized in that The reference signal of the internal standard branch is s ref (t), and its expression is where s ref (t) is a reference signal, t is a time variable, γ represents a frequency modulation slope, f IF is the center frequency after down-conversion of the echo signal, h(t) represents the coefficients of an ultra-low range sidelobe filter, represents the phase modulation jointly caused by the internal calibration branch and the transmitter; The target echo signal is s r (t), and its expression is where s r (t) is the echo signal of the radar system, is the phase modulation jointly caused by the receiver and the transmitter.

6. The terahertz ultra-low range sidelobe pulse compression method based on phase correction according to claim 5, characterized in that, The residual phase is which can be expressed as Where imag represents obtaining the imaginary part of a complex number, and real represents obtaining the real part of a complex number.

7. A terahertz ultra-low range sidelobe pulse compression method based on phase correction according to claim 6, characterized in that The matched filtering result is x(t), and its expression is where "*" is the convolution operation.

8. A terahertz ultra-low range sidelobe pulse compression method based on phase correction according to claim 7, characterized in that The data value after being filtered by the filter is y[n], and its calculation formula is y[n] = x[n] * h[n], where x[n] is the data after discretizing the matched filtering result x(t), h[n] is the data after discretizing the continuous-time ultra-low range sidelobe filter coefficients, n is the sequence value, and its maximum value does not exceed the maximum sampling value N.

9. The method for terahertz ultra-low range sidelobe pulse compression based on phase correction according to claim 8, wherein The objective function of the optimized filter function is J, and its expression is J = |y′[1]| 2 +|y′[2]| 2 +…+|y′[N]| 2 , where y′[n] is a time series, and y′[N / 2+1-L ml , y′[N / 2+1-L ml +1], ……, y′[N / 2+1+L ml is 0, where L ml is half of the main lobe width.

10. A terahertz ultra-low range sidelobe pulse compression method based on phase correction according to claim 8, characterized in that, The step S4 is specifically: use the optimal ultra-low range sidelobe filter coefficients in step S3 to suppress the sidelobes of the matched filtering result, and finally obtain the processing result of the ultra-low range sidelobe suppression ratio.

Images