A proximity fuze ranging method based on fast warping fourier transform

Abstract

A near-fuse ranging method based on fast warping Fourier transform, comprising the following steps: a single-chip radar in a near-fuse emits a triangular wave detection signal to a detection area, the signal is received by a receiving antenna of the single-chip radar after being reflected by the detection area, a radar echo digital signal is obtained by a pre-processing of a receiver front end of the single-chip radar, and the echo digital signal is sent to a signal processor; the echo digital signal is fitted by a WFFT fitting non-uniform filter group of the signal processor, and a modulus value of a processing result is taken to obtain a spectrum diagram of target energy accumulation with high resolution in a low frequency area and diffusion in a high frequency area; a detection threshold is set, a constant false alarm rate detection is performed on the spectrum diagram of target energy accumulation with high resolution in the low frequency area and diffusion in the high frequency area, and target frequency information is extracted; and distance information is calculated and transmitted to a fuse control system through a missile-borne computer. The present application improves the signal-to-noise ratio in the high frequency area and more accurately captures high frequency information.

Description

Technical Field

[0001] This invention belongs to the field of proximity fuse ranging technology, specifically relating to a proximity fuse ranging method based on fast warped Fourier transform. Background Technology

[0002] In proximity fuze ranging, it is necessary to obtain the missile's distance to the ground in real time to carry out pre-set combat missions. However, due to limitations in warhead size, low power consumption, and missile platform response time, the fuze must be miniaturized, low-power, and have high real-time performance. Therefore, it is necessary to study high-precision and fast ranging calculation methods for proximity fuzes.

[0003] Proximity-based ranging utilizes a linear frequency modulated continuous wave (LFM) system for real-time, precise ranging. Its working principle involves performing beat frequency calculations on the echo signal reflected from the target and the reference signal transmitted by the antenna. The target's distance information is then extracted through signal processing. In signal processing, the distance is directly proportional to the frequency of the beat signal. The closer the distance, the lower the beat signal frequency; conversely, the farther the distance, the higher the beat signal frequency. Since the beat signal contains target information, accurately measuring its frequency is crucial for its accurate extraction.

[0004] In signal processing, the frequency point where the spectral peak is located in the frequency domain is found through the difference-frequency discrete Fourier transform to calculate the target's distance information. Therefore, spectral accuracy is particularly important. The higher the spectral accuracy, the more accurate the frequency point where the spectral peak is located, and the more accurate the calculated target information.

[0005] During missile flight, since the impact altitude threshold is generally on the order of meters, the spectrum near the impact altitude, i.e., the low-frequency spectrum, should be the primary focus to obtain better ranging accuracy. Furthermore, at long distances during ranging, the low echo signal-to-noise ratio makes distance information difficult to observe, thus limiting the performance of the radar ranging system.

[0006] Ideal difference-frequency signals are generally sinusoidal signals. The frequency of a single-frequency signal can be accurately estimated using spectral analysis methods such as maximum likelihood estimation, wavelet analysis, or neural networks, but these methods are difficult to apply due to their massive computational demands. The Fast Fourier Transform (FFT) is the most commonly used method for spectral analysis. However, FFT suffers from spectral leakage and picket-fence effects, resulting in equally spaced distance spectra with distance resolution as the interval, thus introducing ranging errors. To improve spectral accuracy, FFT requires increasing the number of signal processing points, which inevitably leads to increased computation. Current solutions to this problem mainly include Zoom-FFT (ZFFT) and Chirp-Z Transform (CZT).

[0007] Traditional ZFFT uses a unit complex exponential signal to perform complex modulation on the initial signal, and then shifts the center frequency of the selected frequency band to zero frequency based on the frequency-shifting property of Fourier transform. To obtain a refined spectrum near zero frequency after the frequency shift, the sampling rate needs to be reduced. Simultaneously, to prevent aliasing in the sampled spectrum, low-pass filtering is required. After low-pass filtering, the sampling rate is reduced to a fraction of the original sampling rate. Sampling, specifically filtering and resampling to reduce the sampling frequency, increases spectral resolution and thus frequency resolution. However, in actual calculations, to achieve the original target number of sampling points N, the length of the sampled data needs to be increased to D times the original number of sampling points, and then the increased data is sampled. Therefore, the actual physical resolution of the signal is not improved. Another improved ZFFT based on fitting first fits the distance, then shifts the frequency based on the distance information, and repeatedly low-pass filters and resampling to reduce the sampling frequency. Although this reduces some computation while improving ranging accuracy, the filter order is very high, making it difficult to implement. Furthermore, the information obtained is consistent with the traditional FFT, and it does not obtain more accurate distance information. Therefore, ZFFT is essentially just a computationally efficient FFT and does not achieve the goal of increasing spectral resolution to obtain more accurate distance information in practical applications. Moreover, ZFFT only focuses on a specific frequency band, and at low cutoff frequencies, it requires a high-order low-pass filter, resulting in high algorithm complexity and increasing the difficulty of real-time signal processing.

[0008] CZT improves spectral accuracy and refines local frequency bands by increasing the number of FFT operation points within the selected frequency band through interpolation within the selected frequency band without increasing data length. An improved method is to perform FFT processing first, select key observation frequency bands, and then perform CZT processing. However, CZT requires continuous sampling at equal angles along the spiral line, which can block other areas from being monitored when focusing on a specific region. Furthermore, in practical applications, echo information from distant (high-frequency) targets cannot be lost. CZT requires a high sampling rate when observing high-frequency regions, which increases the data volume and computational load, making real-time processing impossible.

[0009] Ideally, difference-frequency signals are typically sinusoidal. While methods like maximum likelihood estimation, wavelet analysis, or neural networks can accurately estimate the frequency of a single-frequency signal, their computational complexity limits their application. The Fast Fourier Transform (FFT) is the most commonly used method for spectral analysis. However, FFT suffers from spectral leakage and picket-fence effects, resulting in equally spaced distance spectra with distance resolution as the interval, leading to ranging errors. Improving spectral accuracy with FFT requires increasing the number of signal processing points, inevitably increasing computational complexity. Therefore, current solutions to these problems primarily include Zoom-FFT (ZFFT) and Chirp-Z Transform (CZT).

[0010] Traditional ZFFT uses a unit complex exponential signal to perform complex modulation on the initial signal, and then shifts the center frequency of the selected frequency band to zero frequency based on the frequency-shifting property of Fourier transform. To obtain a refined spectrum near zero frequency after the frequency shift, the sampling rate needs to be reduced. Simultaneously, to prevent aliasing in the sampled spectrum, low-pass filtering is required. After low-pass filtering, the sampling rate is reduced to a fraction of the original sampling rate. Sampling, specifically filtering and resampling to reduce the sampling frequency, increases spectral resolution and thus frequency resolution. However, in actual calculations, to achieve the original target number of sampling points N, the length of the sampled data needs to be increased to D times the original number of sampling points, and then sampled again. Therefore, the actual physical resolution of the signal is not improved. Another improved ZFFT based on fitting first fits the distance, then shifts the frequency based on the distance information, and repeatedly performs low-pass filtering and resampling to reduce the sampling frequency. Although this reduces some computation while improving ranging accuracy, the algorithm has a high filter order, making it difficult to implement. Furthermore, the information obtained is consistent with the traditional FFT, and it does not obtain more accurate distance information. Therefore, ZFFT is essentially just an FFT algorithm that saves some computation; it does not achieve the goal of increasing spectral resolution to obtain more accurate distance information in practical applications. Moreover, ZFFT only focuses on a specific frequency band, and at low cutoff frequencies, it requires a high-order low-pass filter, increasing algorithm complexity and the difficulty of real-time signal processing.

[0011] CZT improves spectral accuracy and refines local frequency bands by increasing the number of FFT operation points within the selected frequency band through interpolation within the selected frequency band without increasing data length. An improved algorithm performs FFT processing first, selects key observation frequency bands, and then performs CZT processing. However, CZT requires continuous sampling at equal angles along the spiral line, which can block other areas from being monitored when focusing on a specific region. Furthermore, in practical applications, echo information from distant (high-frequency) targets cannot be lost. CZT requires a high sampling rate when observing high-frequency regions, which increases the data volume and computational load, making real-time processing impossible. Summary of the Invention

[0012] To overcome the problems existing in the prior art, the present invention aims to provide a proximity fuse ranging method based on Fast Warped Fourier Transform (WFT), which can improve the spectral accuracy of the low-frequency region of interest while not losing information in the high-frequency region. It meets the real-time requirements of proximity fuse ranging, and the low-frequency region (near-range region) of interest has a dense spectrum, high resolution, high accuracy, and small error. The spectral resolution of the high-frequency region (far-range region) is lower. By fitting a non-uniform filter bank using WFFT, the target energy diffused in the high-frequency region is accumulated, improving the signal-to-noise ratio in the high-frequency region and more accurately capturing high-frequency information.

[0013] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0014] A proximity fuse ranging method based on fast warped Fourier transform includes the following steps;

[0015] Step 1. The single-chip radar in the proximity fuse transmits a triangular wave detection signal to the detection area. After the signal is reflected by the detection area, it is received by the single-chip radar receiving antenna. The front end of the single-chip radar receiver preprocesses the echo signal to obtain the radar echo digital signal, and sends the echo digital signal to the signal processor.

[0016] Step 2. The echo digital signal sent by the single-chip radar is fitted with a non-uniform filter bank by the signal processor using WFFT, and the modulus value of the processing result is taken to obtain a spectrum diagram of target energy accumulation with high resolution in the low frequency region and diffusion in the high frequency region.

[0017] Step 3. Set a detection threshold and perform constant false alarm rate (CFAR) detection on the spectrum of target energy accumulation with high resolution in the low-frequency region and diffusion in the high-frequency region to extract target frequency information;

[0018] Step 4. Calculate the extracted target frequency information to obtain the range information, and transmit the range information to the fuze control system through the missile-borne computer.

[0019] In step 1, the single-chip radar in the proximity fuse consists of a transmitting antenna, a receiving antenna, an on-chip radio frequency channel (low-noise amplifier, mixer, filter), and an on-chip digital signal processing unit, which constitute the proximity fuse ranging device. The proximity fuse ranging device illuminates the ground at an incident angle of α to transmit detection signals, receive echo signals reflected from the target, and output distance information.

[0020] The single-chip radar is used to generate radar signals, receive mixing and amplification filtering; the digital signal processing unit realizes difference frequency intermediate frequency signal processing, high-level binding, generation of modulation triangular waves and output of detonation pulse signals.

[0021] The digital signal processing unit communicates with the onboard computer via a data bus, receives control commands from the onboard computer, and executes corresponding operations. The single-chip radar generates a modulated triangular wave, which is radiated outward through the transmitting antenna via the on-chip radio frequency channel. The receiving channel of the radio frequency channel adopts the De-chirp receiving method, and mixes the signal received by the receiving antenna with the local oscillator signal of the transmitting channel to obtain the difference frequency echo signal, which is output to the digital signal processing unit for ADC sampling.

[0022] The signal processor in step 2 includes WFFT operation and modulo operation;

[0023] The WFFT operation is used to map non-uniformly distributed points on a unit circle in a plane to a warped shape. Points uniformly distributed on a unit circle in a plane;

[0024] Modulo operation is used to calculate the modulus of the result after WFFT operation.

[0025] Step 2 specifically involves:

[0026] WFFT is a fast algorithm of WDFT, so it's used in practical applications as WFFT. However, the principle section describes WDFT. WDFT improves the spectral accuracy within a specific spectral region without increasing the number of signal processing points by selecting appropriate all-pass warping functions and parameters. This improves the frequency resolution of key sampling areas without increasing the original data length or performing interpolation fitting, thus accelerating signal processing and meeting the accuracy and real-time requirements of proximity fuse ranging. WDFT uses the all-pass warping function (AWF) to map the non-uniformly distributed points on the original z-plane unit circle to the warped area. Points uniformly distributed on a unit circle in a plane;

[0027] N-point WDFT of a sequence x(n) of length N This is equivalent to correcting X(z) using the following transformation. After that, Spectral sample values ​​from N uniformly distributed points:

[0028]

[0029] in It is an M-order real-coefficient all-pass function (because the transformation function is used to map the unit circle on the z-plane to...). If the unit circle is on the plane, then the mapping must be an all-pass function;

[0030] Applying the above equation to the z-transform, we get:

[0031]

[0032] yes Frequency sampling values ​​at N uniformly distributed points, i.e. exist The value at the point, so

[0033]

[0034] If we use a first-order all-pass function with complex parameters, let

[0035]

[0036] Where 'a' is called the warping control factor, the N-point WDFT transform is expressed as:

[0037]

[0038] Where a = re jθ (0≤r<1, 0, θ<2π), where r is the magnitude of a and θ is its phase; z=e jω This indicates that z rotates in the complex plane at an angular frequency ω; express In the complex plane at angular frequency Rotate; make a = re jθ z = e jω , Substituting into equation (7), we get:

[0039]

[0040] Taking the root mean square of both sides of equation (9) and converting it to trigonometric functions according to Euler's formula, we have:

[0041]

[0042] Taking the ratio of the imaginary part to the real part of both sides of equation (10), we have:

[0043]

[0044] Simplifying equation (11) yields the frequency mapping expression:

[0045]

[0046] And according to a = re jθ Simplifying equation (12) yields:

[0047]

[0048] After warping Angular frequency of a plane The mapping relationship, or conversion relationship, between the beam center and the angular frequency ω of the original z-plane. In step 1, the angle α is the angle between the beam center and the ground (ground grazing angle), and the specific formula is as follows: The missile altitude is h, the slant distance from the scene center is L, and the beam center is M.

[0049] Step 3 includes constant false alarm rate (CFAR) detection, which is used to set a detection threshold, determine whether it is a target signal, and perform corresponding processing accordingly.

[0050] Step 4 calculates the distance based on the target frequency information processed in step 3, and outputs the distance information to transmit the detonation command to the fuse control system.

[0051] Step 4 specifically involves: the fuze ranging utilizes a linear frequency modulated triangular wave system for real-time and accurate ranging.

[0052] The fuze ranging method calculates the beat frequency by performing a beat operation on the echo signal reflected from the target and the reference signal transmitted by the antenna, and then extracts the target's distance information through signal processing; the specific process is as follows:

[0053] The relationship between the time delay Δt of the transmitted signal and the echo signal and the target distance L is as follows:

[0054] Δt=2R / c (1)

[0055] Where c is the speed of light; the difference frequency signal output by the mixer is f. b ,have:

[0056]

[0057] Where T is the period of the linear frequency modulated triangular wave, and B is the frequency modulation bandwidth; the target distance L can be calculated from (1) and (2) as follows:

[0058]

[0059] The target distance L and the frequency difference f b Proportional.

[0060] The beneficial effects of this invention are:

[0061] In the practical application scenarios of this invention, compared with other methods to improve spectral accuracy, WFFT is easy to calculate, efficient, and accurate. Moreover, it has high resolution, high accuracy, and small error for the low-frequency region (near-distance region) that is of particular interest. While ensuring that information in the high-frequency region (far-distance region) is not lost, the WFFT fits the non-uniform filter bank to accumulate the target energy diffused in the high-frequency region, which is equivalent to improving the signal-to-noise ratio in the high-frequency region. This allows for more accurate capture of high-frequency information and improves the ranging accuracy in the far-distance region.

[0062] This is something other methods cannot achieve. For example, ZFFT is merely a computationally efficient FFT that doesn't fulfill the practical application goal of increasing spectral resolution to obtain more accurate distance information. Furthermore, ZFFT only focuses on a specific frequency band and has high requirements for low-pass filters, resulting in high algorithm complexity and increased difficulty in real-time processing. CZT, on the other hand, requires continuous sampling at equal angles along the spiral line. Therefore, when focusing on a specific area, other areas will be blocked and unable to be monitored. Simultaneously, in practical applications, echo information from distant (high-frequency) targets cannot be lost. However, CZT requires a high sampling rate when observing high-frequency regions, which increases the data volume and computational load, making real-time processing impossible. Attached Figure Description

[0063] Figure 1 is a schematic diagram of the ranging model of the fuze of the present invention; wherein Figure 1-1 This is the initial terminal guidance phase fuze ranging model; Figure 1-2 This is a dynamic terminal guidance phase fuze ranging model.

[0064] Figure 2 This is a schematic diagram of the components of a fuze ranging system.

[0065] Figure 3 This is a flowchart of digital signal processing.

[0066] Figure 4 Flowchart for calculating the WFFT spectrum.

[0067] Figure 5 A schematic diagram showing the distribution of frequency points on the unit circle in the z-plane in the cases of FFT(a) and WFFT(b).

[0068] Figure 6 This diagram illustrates the distribution of frequency points on the spectrum in the two cases: FFT(a) and WFFT(b). Detailed Implementation

[0069] The present invention will now be described in further detail with reference to the accompanying drawings.

[0070] As shown in Figure 1: A proximity fuse ranging method based on warped Fourier transform includes the following steps;

[0071] Step 1. The single-chip radar in the proximity fuse transmits a triangular wave detection signal to the detection area. After the signal is reflected by the detection area, it is received by the single-chip radar receiving antenna. The front end of the single-chip radar receiver preprocesses the echo signal to obtain the radar echo digital signal, and sends the echo digital signal to the signal processor.

[0072] Step 2. The echo digital signal sent by the single-chip radar is fitted with a non-uniform filter bank by the signal processor using WFFT, and the modulus value of the processing result is taken to obtain a spectrum diagram of target energy accumulation with high resolution in the low frequency region and diffusion in the high frequency region.

[0073] Step 3. Set a detection threshold and perform constant false alarm rate (CFAR) detection on the spectrum of target energy accumulation with high resolution in the low-frequency region and diffusion in the high-frequency region to extract target frequency information;

[0074] Step 4. Calculate the extracted target frequency information to obtain the range information, and transmit the range information to the fuze control system through the missile-borne computer.

[0075] In step 1, the single-chip radar in the proximity fuse consists of a transmitting antenna, a receiving antenna, an on-chip radio frequency channel (low-noise amplifier, mixer, filter), and an on-chip digital signal processing unit, which constitute the proximity fuse ranging device. The proximity fuse ranging device illuminates the ground at an incident angle of α to transmit detection signals, receive echo signals reflected from the target, and output distance information.

[0076] The single-chip radar is used to generate radar signals, receive mixing and amplification filtering; the digital signal processing unit realizes difference frequency intermediate frequency signal processing, high-level binding, generation of modulation triangular waves and output of detonation pulse signals.

[0077] The digital signal processing unit communicates with the onboard computer via a data bus, receives control commands from the onboard computer, and executes corresponding operations. The single-chip radar generates a modulated triangular wave, which is radiated outward through the transmitting antenna via the on-chip radio frequency channel. The receiving channel of the radio frequency channel adopts the De-chirp receiving method, and mixes the signal received by the receiving antenna with the local oscillator signal of the transmitting channel to obtain the difference frequency echo signal, which is output to the digital signal processing unit for ADC sampling.

[0078] The signal processor in step 2 includes WFFT operation, modulus operation, constant false alarm rate detection, and distance output;

[0079] The WFFT operation is used to map non-uniformly distributed points on a unit circle in a plane to a warped shape. Points uniformly distributed on a unit circle in a plane;

[0080] Modulo operation is used to calculate the modulus of the result after WFFT operation;

[0081] Constant false alarm rate (CFAR) detection is used to set a detection threshold, determine whether it is a target signal, and perform corresponding processing accordingly.

[0082] Distance output is used to calculate and output distance information, which is then transmitted to the detonation command to the fuse control system.

[0083] In step 2, the WFFT fitting of the non-uniform filter bank specifically involves:

[0084] WFFT is a fast algorithm of WDFT, so it is used in practical applications. However, the principle section describes WDFT. WDFT improves the spectral accuracy in a specific spectral region without increasing the number of signal processing points by selecting appropriate all-pass warp functions and parameters. This improves the frequency resolution of key sampling areas, and does not increase the length of the original data or perform interpolation fitting, thus speeding up signal processing and meeting the accuracy and real-time requirements of proximity fuse ranging.

[0085] The fuze ranging utilizes a linear frequency modulated triangular wave system for real-time and accurate ranging.

[0086] The fuze ranging method calculates the beat frequency by performing a beat operation on the echo signal reflected from the target and the reference signal transmitted by the antenna, and then extracts the target's distance information through signal processing.

[0087] The angle α is the angle between the beam center and the ground (ground grazing angle), and the specific formula is as follows: The missile altitude is h, the slant distance from the scene center is L, and the beam center is M.

[0088] This method is constructed through the following system:

[0089] 1. Missile-borne fuze ranging model:

[0090] The proximity fuse rangefinder mounted on the warhead illuminates the ground at an incident angle of α to emit detection signals, receive echo signals reflected from the target, and output distance information.

[0091] The fuze ranging system utilizes a linear frequency modulated triangular wave system for real-time and accurate ranging.

[0092] Its working principle is to perform beat frequency calculation on the echo signal reflected from the target and the reference signal transmitted by the antenna, and then extract the target's range information through signal processing; the specific process is as follows:

[0093] The relationship between the time delay Δt of the transmitted signal and the echo signal and the target distance L is as follows:

[0094] Δt=2R / c (1)

[0095] Where c is the speed of light. The difference frequency signal output by the mixer is f. b ,have:

[0096]

[0097] Where T is the period of the linear frequency modulated triangular wave, and B is the frequency modulation bandwidth. The target distance L can be calculated from (1) and (2) as follows:

[0098]

[0099] It can be seen that the target distance L and the difference frequency f b The frequency of the beat signal is directly proportional to the target information, so the measurement of the beat signal frequency is particularly important for accurately extracting target information.

[0100] Figure 1 shows the geometric relationship of missile-borne radar ranging, where the missile altitude is h, the slant range of the scene center is L, the beam center is M, and the angle between the beam center and the ground (ground grazing angle) is M.

[0101] Figure 1-1 and 1-2 It is a range-finding model of the fuze. The proximity fuze range-finding device on the warhead illuminates the ground at an incident angle of α to emit detection signals, receive echo signals reflected from the target, and output distance information.

[0102] 2. Proximity fuse ranging system:

[0103] The proposed fuse structure diagram is as follows: Figure 2 As shown, it consists of a transmitting antenna, a receiving antenna, an on-chip radio frequency channel (low-noise amplifier, mixer, filter, etc.), and an on-chip digital signal processing unit. The fuze adopts the form of linear frequency modulated continuous wave transmission signal.

[0104] The single-chip radar has functions such as generating radar signals, receiving and mixing, and amplifying and filtering. The digital signal processing unit realizes the processing of difference frequency intermediate frequency signals, detonation height setting, generation of modulated triangular waves, and output of detonation pulse signals. The digital signal processing unit communicates with the onboard computer through a data bus, receives control commands from the onboard computer, and executes corresponding operations. The single-chip radar generates modulated triangular waves, which are radiated outward through the transmitting antenna via the on-chip radio frequency channel. The receiving channel of the radio frequency channel adopts the De-chirp receiving method, mixing the signal received by the receiving antenna with the local oscillator signal of the transmitting channel to obtain the difference frequency echo signal, which is output to the digital signal processing unit for ADC sampling.

[0105] 3. Digital Signal Processor:

[0106] Digital signal processing flowchart, in which:

[0107] Preprocessing is used to convert the echo signal received by the radar receiver front end into a digital echo signal;

[0108] The WFFT operation is used to map non-uniformly distributed points on a unit circle in a plane to a warped shape. Points uniformly distributed on a unit circle in a plane;

[0109] Modulo operation is used to calculate the modulus of the result after WFFT operation;

[0110] Constant false alarm rate (CFAR) detection is used to set a detection threshold, determine whether it is a target signal, and perform corresponding processing accordingly.

[0111] Distance output is used to calculate and output distance information, transmitting the proximity detonation command to the fuse control system. 4. WFFT Algorithm Principle:

[0112] WFFT is a fast algorithm for implementing Warped Discrete Fourier Transform (WDFT), a non-uniform Discrete Fourier Transform (NDFT). Compared with the traditional DFT, WDFT can improve the spectral accuracy in a specific spectral region without increasing the number of signal processing points by selecting appropriate all-pass warping functions and parameters. This improves the frequency resolution of key sampling areas, and it does not increase the length of the original data or perform interpolation fitting, thus speeding up signal processing and meeting the accuracy and real-time requirements of proximity fuse ranging.

[0113] The advantage of WDFT lies in its ability to uniformly distribute frequency points across non-uniform frequency samples by altering their location, thereby achieving higher accuracy. First-order all-pass warp functions are typically used to fit non-uniformly distributed filter banks, resulting in denser spectra in the near-range region (low-frequency region), leading to high resolution, high accuracy, and small errors; and target energy accumulation in the far-range region (high-frequency region), thus more accurately capturing information about distant targets.

[0114] WDFT uses the All-pass Warping Function (AWF) to map non-uniformly distributed points on the original z-plane unit circle to the warped surface. Points uniformly distributed on a unit circle in a plane.

[0115] like Figure 4As shown: N-point WDFT of a sequence x(n) of length N. This is equivalent to correcting X(z) using the following transformation. After that, Spectral sample values ​​from N uniformly distributed points:

[0116]

[0117] in It is an M-order real-coefficient all-pass function (because the transformation function is used to map the unit circle on the z-plane to...). If the unit circle is on the plane, then the mapping must be an all-pass function. Its flowchart is as follows: Figure 4 As shown:

[0118] Applying the above equation to the z-transform, we get:

[0119]

[0120] yes Frequency sampling values ​​at N uniformly distributed points, i.e. exist The value at the point, so

[0121]

[0122] If we use a first-order all-pass function with complex parameters, let

[0123]

[0124] Where 'a' is called the warping control factor, the N-point WDFT transform can be expressed as:

[0125]

[0126] Where a = re jθ (0≤r<1, 0, θ<2π), where r is the magnitude of a and θ is its phase. z=e jω This indicates that z rotates in the complex plane at an angular frequency ω; z = e jω , Substituting into equation (7), we get:

[0127]

[0128] Taking the root mean square of both sides of equation (9) and converting them to trigonometric functions, we have:

[0129]

[0130] Taking the ratio of the imaginary part to the real part of both sides of equation (10), we have:

[0131]

[0132] Simplifying equation (11) yields the frequency mapping expression:

[0133]

[0134] And according to a = re jθ Simplifying equation (12) yields:

[0135] In this invention, θ is set to 0, meaning the warpage control factor a is a real number. When a is positive, the spectral resolution in the low-frequency region is higher, at the cost of reduced spectral resolution in the high-frequency region; the opposite is true when a is negative. When a is zero, the WDFT transform becomes the traditional DFT transform. Moreover, to ensure that A(z) is a stable system, the condition |a| < 1 must be satisfied.

[0136] The characteristics of the range information output after signal processing by this method are: the operation speed of WFFT based on FFT is fast, which meets the real-time requirements of proximity fuse ranging; the low-frequency region (near-range region) of focus has dense spectrum, high resolution, high accuracy and small error; the spectrum resolution of the high-frequency region (far-range region) becomes lower, and the target energy diffused in the high-frequency region can be accumulated by fitting a non-uniform filter bank with WFFT, which is equivalent to improving the signal-to-noise ratio in the high-frequency region and can capture high-frequency information more accurately.

[0137] Simulation explanation of the technical solution of this invention:

[0138] 1. Distribution of sampling points on the unit circle in the z-plane:

[0139] By simulating the distribution of sampling points on the unit circle, the distribution of sampling points can be more intuitively and clearly reflected. In the simulation, the number of sampling points N is taken as 16, with the horizontal axis representing the real part and the vertical axis representing the imaginary part. Figure 5 Compare the frequency distribution under the two cases: FFT (Figure a) and WFFT (Figure b).

[0140] Figure 5 The distribution of frequency points on the unit circle in the z-plane is compared between FFT(a) and WFFT(b). Simulation results show that when the warpage control factor is 0, the sampling points are uniformly distributed on the unit circle, that is, when a is 0, the WFFT transform becomes the traditional FFT transform.

[0141] 2. Spectrum simulation of FFT and WFFT algorithms:

[0142] By simulating the spectral distribution of the sampling points, the application in real-world scenarios can be more intuitively and clearly reflected. In the simulation, the number of sampling points N is set to 64, with the horizontal axis representing frequency and the vertical axis representing amplitude. For example... Figure 6 Compare the frequency distribution in the spectrum under the two cases of FFT(a) and WFFT(b).

[0143] In the simulation of this method, it can be clearly seen that after applying WFFT, the low-frequency region (near distance region) has a dense spectrum, high resolution, high accuracy and small error. The spectral resolution of the high-frequency region (far distance region) becomes lower. However, by fitting the non-uniform filter bank with WFFT, the target energy diffused in the high-frequency region is accumulated, which is equivalent to improving the signal-to-noise ratio of the high-frequency region and can capture high-frequency information more accurately.

[0144] This invention combines WFFT fitting with a non-uniformly distributed filter bank for proximity fuse ranging, which aligns with practical application scenarios. The range information output after signal processing using this method has the following characteristics:

[0145] 1. Real-time performance: This method meets the real-time requirements of fuze ranging, that is, it can quickly and accurately process signals and output distance information in real-time scenarios, ensuring that the fuze can respond and trigger in a timely manner in practical applications.

[0146] 2. Improving the spectral resolution of key observation bands without affecting processing speed: This method can improve the frequency resolution of key observation areas, enabling more accurate analysis and capture of the spectral characteristics of the low-frequency region (near-range area) of focus during distance information processing. This allows for the acquisition of low-frequency spectrum with dense spectrum and high resolution, resulting in high-precision distance information with small errors, thereby improving the accuracy of fuse triggering.

[0147] 3. No loss of spectral information in non-key observation areas: This method improves the frequency resolution of key observation areas while ensuring that spectral information in non-key sampling areas is not lost. This allows for full utilization of the entire spectral information in the signal, ensuring comprehensive and accurate processing of distance information across the entire ranging area.

[0148] 4. Accurate capture of high-frequency information: While ensuring that high-frequency (long-distance) information is not lost, this method accumulates the target energy diffused in the high-frequency region by fitting a non-uniform filter bank with WFFT, which is equivalent to improving the signal-to-noise ratio in the high-frequency region. This allows for more accurate capture of high-frequency information and improves the ranging accuracy in long-distance regions.

[0149] In summary, the distance information output after signal processing using this method has the characteristics of real-time performance, improved frequency resolution, no loss of spectral information in non-key observation areas (high-frequency areas), and accurate capture of high-frequency information. Thus, it can achieve high-precision and high-accuracy distance measurement in proximity fuse ranging.

[0150] The above content is only for illustrating the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. Any modifications made to the technical solution based on the technical concept proposed in this invention shall fall within the scope of protection of the claims of this invention.

Claims

1. A proximity fuse ranging method based on fast warped Fourier transform, characterized in that, Includes the following steps; Step 1. The single-chip radar in the proximity fuse transmits a triangular wave detection signal to the detection area. After the signal is reflected by the detection area, it is received by the single-chip radar receiving antenna. The front end of the single-chip radar receiver preprocesses the echo signal to obtain the radar echo digital signal, and sends the echo digital signal to the signal processor. Step 2. The echo digital signal sent by the single-chip radar is fitted with a non-uniform filter bank by the signal processor using WFFT, and the modulus value of the processing result is taken to obtain a spectrum diagram of target energy accumulation with high resolution in the low frequency region and diffusion in the high frequency region. Step 3. Set a detection threshold and perform constant false alarm rate (CFAR) detection on the spectrum of target energy accumulation with high resolution in the low-frequency region and diffusion in the high-frequency region to extract target frequency information; Step 4. Calculate the extracted target frequency information to obtain the range information, and transmit the range information to the fuze control system through the missile-borne computer.

2. The proximity fuse ranging method based on fast warped Fourier transform according to claim 1, characterized in that, In step 1, the single-chip radar in the proximity fuse consists of a transmitting antenna, a receiving antenna, an on-chip radio frequency channel, and an on-chip digital signal processing unit, which constitute a proximity fuse ranging device. The proximity fuse ranging device illuminates the ground at an incident angle of α to transmit a detection signal, receive the echo signal reflected from the target, and output the distance information.

3. The proximity fuse ranging method based on fast warped Fourier transform according to claim 2, characterized in that, The single-chip radar is used to generate radar signals, receive mixing and amplification filtering; the digital signal processing unit realizes difference frequency intermediate frequency signal processing, high-level binding, generation of modulation triangular waves and output of detonation pulse signals.

4. The proximity fuse ranging method based on fast warped Fourier transform according to claim 2, characterized in that, The digital signal processing unit communicates with the onboard computer via a data bus, receives control commands from the onboard computer, and executes corresponding operations. The single-chip radar generates a modulated triangular wave, which is radiated outward through the transmitting antenna via the on-chip radio frequency channel. The receiving channel of the radio frequency channel adopts the De-chirp receiving method, and mixes the signal received by the receiving antenna with the local oscillator signal of the transmitting channel to obtain the difference frequency echo signal, which is output to the digital signal processing unit for ADC sampling.

5. The proximity fuse ranging method based on fast warped Fourier transform according to claim 2, characterized in that, In step 1, the angle α is the angle between the beam center and the ground, and the specific formula is as follows: The missile altitude is h, the slant distance from the scene center is L, and the beam center is M.

6. The proximity fuse ranging method based on fast warped Fourier transform according to claim 1, characterized in that, The signal processor in step 2 includes WFFT operation and modulo operation; The WFFT operation is used to map non-uniformly distributed points on a unit circle in a plane to a warped shape. Points uniformly distributed on a unit circle in a plane; Modulo operation is used to calculate the modulus of the result after WFFT operation.

7. The proximity fuse ranging method based on fast warped Fourier transform according to claim 6, characterized in that, Step 2 specifically involves: WFFT is a fast algorithm of WDFT. In practical applications, it is called WFFT, but the principle section describes WDFT. WDFT uses an all-pass warp function to map non-uniformly distributed points on the original z-plane unit circle to the warped version. Points uniformly distributed on a unit circle in a plane; N-point WDFT of a sequence x(n) of length N This is equivalent to correcting X(z) using the following transformation. After that, Spectral sample values ​​from N uniformly distributed points: in It is an M-order all-pass function with real coefficients; Applying the above equation to the z-transform, we get: WDFT yes Frequency sampling values ​​at N uniformly distributed points, i.e. exist The value at the point, so If we use a first-order all-pass function with complex parameters, let Where 'a' is called the warping control factor, the N-point WDFT transform is expressed as: Where a = re jθ (0≤r<1,0≤θ<2π), where r is the magnitude of a and θ is its phase; z=e jω This indicates that z rotates in the complex plane with an angular frequency ω; a = re jθ z = e jω , Substituting into equation (7), we get: Taking the root mean square of both sides of equation (9) and converting them to trigonometric functions, we have: Taking the ratio of the imaginary part to the real part of both sides of equation (10), we have: Simplifying equation (11) yields the frequency mapping expression: And according to a = re jθ Simplifying equation (12) yields: After warping Angular frequency of a plane The mapping relationship, or conversion relationship, between the angular frequency ω of the original z-plane and the z-plane.

8. The proximity fuse ranging method based on fast warped Fourier transform according to claim 1, characterized in that, Step 3 includes constant false alarm rate (CFAR) detection, which is used to set a detection threshold, determine whether it is a target signal, and perform corresponding processing accordingly.

9. The proximity fuse ranging method based on fast warped Fourier transform according to claim 1, characterized in that, Step 4 calculates the distance based on the target frequency information processed in step 3, and outputs the distance information to transmit the detonation command to the fuse control system.

10. A proximity fuse ranging method based on fast warped Fourier transform according to claim 9, characterized in that, Step 4 specifically involves: the fuze ranging utilizes a linear frequency modulated triangular wave system for real-time and accurate ranging. The fuze ranging method calculates the beat frequency by performing a beat operation on the echo signal reflected from the target and the reference signal transmitted by the antenna, and then extracts the target's distance information through signal processing; the specific process is as follows: The relationship between the time delay Δt of the transmitted signal and the echo signal and the target distance L is as follows: Δt=2R / c (1) Where c is the speed of light; the difference frequency signal output by the mixer is f. b ,have: Where T is the period of the linear frequency modulated triangular wave, and B is the frequency modulation bandwidth; the target distance L can be calculated from (1) and (2) as follows: The target distance L and the frequency difference f b Proportional.

Images